1\input texinfo @c -*-texinfo-*- 2 3@c %**start of header 4@setfilename wget.info 5@include version.texi 6@settitle GNU Wget @value{VERSION} Manual 7@c Disable the monstrous rectangles beside overfull hbox-es. 8@finalout 9@c Use `odd' to print double-sided. 10@setchapternewpage on 11@c %**end of header 12 13@iftex 14@c Remove this if you don't use A4 paper. 15@afourpaper 16@end iftex 17 18@c Title for man page. The weird way texi2pod.pl is written requires 19@c the preceding @set. 20@set Wget Wget 21@c man title Wget The non-interactive network downloader. 22 23@dircategory Network Applications 24@direntry 25* Wget: (wget). The non-interactive network downloader. 26@end direntry 27 28@copying 29This file documents the GNU Wget utility for downloading network 30data. 31 32@c man begin COPYRIGHT 33Copyright @copyright{} 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 342004, 2005, 2006, 2007, 2008, 2009 Free Software Foundation, Inc. 35 36@iftex 37Permission is granted to make and distribute verbatim copies of 38this manual provided the copyright notice and this permission notice 39are preserved on all copies. 40@end iftex 41 42@ignore 43Permission is granted to process this file through TeX and print the 44results, provided the printed document carries a copying permission 45notice identical to this one except for the removal of this paragraph 46(this paragraph not being relevant to the printed manual). 47@end ignore 48Permission is granted to copy, distribute and/or modify this document 49under the terms of the GNU Free Documentation License, Version 1.2 or 50any later version published by the Free Software Foundation; with no 51Invariant Sections, no Front-Cover Texts, and no Back-Cover Texts. A 52copy of the license is included in the section entitled ``GNU Free 53Documentation License''. 54@c man end 55@end copying 56 57@titlepage 58@title GNU Wget @value{VERSION} 59@subtitle The non-interactive download utility 60@subtitle Updated for Wget @value{VERSION}, @value{UPDATED} 61@author by Hrvoje Nik@v{s}i@'{c} and others 62 63@ignore 64@c man begin AUTHOR 65Originally written by Hrvoje Niksic <hniksic@xemacs.org>. 66Currently maintained by Micah Cowan <micah@cowan.name>. 67@c man end 68@c man begin SEEALSO 69This is @strong{not} the complete manual for GNU Wget. 70For more complete information, including more detailed explanations of 71some of the options, and a number of commands available 72for use with @file{.wgetrc} files and the @samp{-e} option, see the GNU 73Info entry for @file{wget}. 74@c man end 75@end ignore 76 77@page 78@vskip 0pt plus 1filll 79@insertcopying 80@end titlepage 81 82@contents 83 84@ifnottex 85@node Top, Overview, (dir), (dir) 86@top Wget @value{VERSION} 87 88@insertcopying 89@end ifnottex 90 91@menu 92* Overview:: Features of Wget. 93* Invoking:: Wget command-line arguments. 94* Recursive Download:: Downloading interlinked pages. 95* Following Links:: The available methods of chasing links. 96* Time-Stamping:: Mirroring according to time-stamps. 97* Startup File:: Wget's initialization file. 98* Examples:: Examples of usage. 99* Various:: The stuff that doesn't fit anywhere else. 100* Appendices:: Some useful references. 101* Copying this manual:: You may give out copies of this manual. 102* Concept Index:: Topics covered by this manual. 103@end menu 104 105@node Overview, Invoking, Top, Top 106@chapter Overview 107@cindex overview 108@cindex features 109 110@c man begin DESCRIPTION 111GNU Wget is a free utility for non-interactive download of files from 112the Web. It supports @sc{http}, @sc{https}, and @sc{ftp} protocols, as 113well as retrieval through @sc{http} proxies. 114 115@c man end 116This chapter is a partial overview of Wget's features. 117 118@itemize @bullet 119@item 120@c man begin DESCRIPTION 121Wget is non-interactive, meaning that it can work in the background, 122while the user is not logged on. This allows you to start a retrieval 123and disconnect from the system, letting Wget finish the work. By 124contrast, most of the Web browsers require constant user's presence, 125which can be a great hindrance when transferring a lot of data. 126@c man end 127 128@item 129@ignore 130@c man begin DESCRIPTION 131 132@c man end 133@end ignore 134@c man begin DESCRIPTION 135Wget can follow links in @sc{html}, @sc{xhtml}, and @sc{css} pages, to 136create local versions of remote web sites, fully recreating the 137directory structure of the original site. This is sometimes referred to 138as ``recursive downloading.'' While doing that, Wget respects the Robot 139Exclusion Standard (@file{/robots.txt}). Wget can be instructed to 140convert the links in downloaded files to point at the local files, for 141offline viewing. 142@c man end 143 144@item 145File name wildcard matching and recursive mirroring of directories are 146available when retrieving via @sc{ftp}. Wget can read the time-stamp 147information given by both @sc{http} and @sc{ftp} servers, and store it 148locally. Thus Wget can see if the remote file has changed since last 149retrieval, and automatically retrieve the new version if it has. This 150makes Wget suitable for mirroring of @sc{ftp} sites, as well as home 151pages. 152 153@item 154@ignore 155@c man begin DESCRIPTION 156 157@c man end 158@end ignore 159@c man begin DESCRIPTION 160Wget has been designed for robustness over slow or unstable network 161connections; if a download fails due to a network problem, it will 162keep retrying until the whole file has been retrieved. If the server 163supports regetting, it will instruct the server to continue the 164download from where it left off. 165@c man end 166 167@item 168Wget supports proxy servers, which can lighten the network load, speed 169up retrieval and provide access behind firewalls. Wget uses the passive 170@sc{ftp} downloading by default, active @sc{ftp} being an option. 171 172@item 173Wget supports IP version 6, the next generation of IP. IPv6 is 174autodetected at compile-time, and can be disabled at either build or 175run time. Binaries built with IPv6 support work well in both 176IPv4-only and dual family environments. 177 178@item 179Built-in features offer mechanisms to tune which links you wish to follow 180(@pxref{Following Links}). 181 182@item 183The progress of individual downloads is traced using a progress gauge. 184Interactive downloads are tracked using a ``thermometer''-style gauge, 185whereas non-interactive ones are traced with dots, each dot 186representing a fixed amount of data received (1KB by default). Either 187gauge can be customized to your preferences. 188 189@item 190Most of the features are fully configurable, either through command line 191options, or via the initialization file @file{.wgetrc} (@pxref{Startup 192File}). Wget allows you to define @dfn{global} startup files 193(@file{/usr/local/etc/wgetrc} by default) for site settings. 194 195@ignore 196@c man begin FILES 197@table @samp 198@item /usr/local/etc/wgetrc 199Default location of the @dfn{global} startup file. 200 201@item .wgetrc 202User startup file. 203@end table 204@c man end 205@end ignore 206 207@item 208Finally, GNU Wget is free software. This means that everyone may use 209it, redistribute it and/or modify it under the terms of the GNU General 210Public License, as published by the Free Software Foundation (see the 211file @file{COPYING} that came with GNU Wget, for details). 212@end itemize 213 214@node Invoking, Recursive Download, Overview, Top 215@chapter Invoking 216@cindex invoking 217@cindex command line 218@cindex arguments 219@cindex nohup 220 221By default, Wget is very simple to invoke. The basic syntax is: 222 223@example 224@c man begin SYNOPSIS 225wget [@var{option}]@dots{} [@var{URL}]@dots{} 226@c man end 227@end example 228 229Wget will simply download all the @sc{url}s specified on the command 230line. @var{URL} is a @dfn{Uniform Resource Locator}, as defined below. 231 232However, you may wish to change some of the default parameters of 233Wget. You can do it two ways: permanently, adding the appropriate 234command to @file{.wgetrc} (@pxref{Startup File}), or specifying it on 235the command line. 236 237@menu 238* URL Format:: 239* Option Syntax:: 240* Basic Startup Options:: 241* Logging and Input File Options:: 242* Download Options:: 243* Directory Options:: 244* HTTP Options:: 245* HTTPS (SSL/TLS) Options:: 246* FTP Options:: 247* Recursive Retrieval Options:: 248* Recursive Accept/Reject Options:: 249* Exit Status:: 250@end menu 251 252@node URL Format, Option Syntax, Invoking, Invoking 253@section URL Format 254@cindex URL 255@cindex URL syntax 256 257@dfn{URL} is an acronym for Uniform Resource Locator. A uniform 258resource locator is a compact string representation for a resource 259available via the Internet. Wget recognizes the @sc{url} syntax as per 260@sc{rfc1738}. This is the most widely used form (square brackets denote 261optional parts): 262 263@example 264http://host[:port]/directory/file 265ftp://host[:port]/directory/file 266@end example 267 268You can also encode your username and password within a @sc{url}: 269 270@example 271ftp://user:password@@host/path 272http://user:password@@host/path 273@end example 274 275Either @var{user} or @var{password}, or both, may be left out. If you 276leave out either the @sc{http} username or password, no authentication 277will be sent. If you leave out the @sc{ftp} username, @samp{anonymous} 278will be used. If you leave out the @sc{ftp} password, your email 279address will be supplied as a default password.@footnote{If you have a 280@file{.netrc} file in your home directory, password will also be 281searched for there.} 282 283@strong{Important Note}: if you specify a password-containing @sc{url} 284on the command line, the username and password will be plainly visible 285to all users on the system, by way of @code{ps}. On multi-user systems, 286this is a big security risk. To work around it, use @code{wget -i -} 287and feed the @sc{url}s to Wget's standard input, each on a separate 288line, terminated by @kbd{C-d}. 289 290You can encode unsafe characters in a @sc{url} as @samp{%xy}, @code{xy} 291being the hexadecimal representation of the character's @sc{ascii} 292value. Some common unsafe characters include @samp{%} (quoted as 293@samp{%25}), @samp{:} (quoted as @samp{%3A}), and @samp{@@} (quoted as 294@samp{%40}). Refer to @sc{rfc1738} for a comprehensive list of unsafe 295characters. 296 297Wget also supports the @code{type} feature for @sc{ftp} @sc{url}s. By 298default, @sc{ftp} documents are retrieved in the binary mode (type 299@samp{i}), which means that they are downloaded unchanged. Another 300useful mode is the @samp{a} (@dfn{ASCII}) mode, which converts the line 301delimiters between the different operating systems, and is thus useful 302for text files. Here is an example: 303 304@example 305ftp://host/directory/file;type=a 306@end example 307 308Two alternative variants of @sc{url} specification are also supported, 309because of historical (hysterical?) reasons and their widespreaded use. 310 311@sc{ftp}-only syntax (supported by @code{NcFTP}): 312@example 313host:/dir/file 314@end example 315 316@sc{http}-only syntax (introduced by @code{Netscape}): 317@example 318host[:port]/dir/file 319@end example 320 321These two alternative forms are deprecated, and may cease being 322supported in the future. 323 324If you do not understand the difference between these notations, or do 325not know which one to use, just use the plain ordinary format you use 326with your favorite browser, like @code{Lynx} or @code{Netscape}. 327 328@c man begin OPTIONS 329 330@node Option Syntax, Basic Startup Options, URL Format, Invoking 331@section Option Syntax 332@cindex option syntax 333@cindex syntax of options 334 335Since Wget uses GNU getopt to process command-line arguments, every 336option has a long form along with the short one. Long options are 337more convenient to remember, but take time to type. You may freely 338mix different option styles, or specify options after the command-line 339arguments. Thus you may write: 340 341@example 342wget -r --tries=10 http://fly.srk.fer.hr/ -o log 343@end example 344 345The space between the option accepting an argument and the argument may 346be omitted. Instead of @samp{-o log} you can write @samp{-olog}. 347 348You may put several options that do not require arguments together, 349like: 350 351@example 352wget -drc @var{URL} 353@end example 354 355This is completely equivalent to: 356 357@example 358wget -d -r -c @var{URL} 359@end example 360 361Since the options can be specified after the arguments, you may 362terminate them with @samp{--}. So the following will try to download 363@sc{url} @samp{-x}, reporting failure to @file{log}: 364 365@example 366wget -o log -- -x 367@end example 368 369The options that accept comma-separated lists all respect the convention 370that specifying an empty list clears its value. This can be useful to 371clear the @file{.wgetrc} settings. For instance, if your @file{.wgetrc} 372sets @code{exclude_directories} to @file{/cgi-bin}, the following 373example will first reset it, and then set it to exclude @file{/~nobody} 374and @file{/~somebody}. You can also clear the lists in @file{.wgetrc} 375(@pxref{Wgetrc Syntax}). 376 377@example 378wget -X '' -X /~nobody,/~somebody 379@end example 380 381Most options that do not accept arguments are @dfn{boolean} options, 382so named because their state can be captured with a yes-or-no 383(``boolean'') variable. For example, @samp{--follow-ftp} tells Wget 384to follow FTP links from HTML files and, on the other hand, 385@samp{--no-glob} tells it not to perform file globbing on FTP URLs. A 386boolean option is either @dfn{affirmative} or @dfn{negative} 387(beginning with @samp{--no}). All such options share several 388properties. 389 390Unless stated otherwise, it is assumed that the default behavior is 391the opposite of what the option accomplishes. For example, the 392documented existence of @samp{--follow-ftp} assumes that the default 393is to @emph{not} follow FTP links from HTML pages. 394 395Affirmative options can be negated by prepending the @samp{--no-} to 396the option name; negative options can be negated by omitting the 397@samp{--no-} prefix. This might seem superfluous---if the default for 398an affirmative option is to not do something, then why provide a way 399to explicitly turn it off? But the startup file may in fact change 400the default. For instance, using @code{follow_ftp = on} in 401@file{.wgetrc} makes Wget @emph{follow} FTP links by default, and 402using @samp{--no-follow-ftp} is the only way to restore the factory 403default from the command line. 404 405@node Basic Startup Options, Logging and Input File Options, Option Syntax, Invoking 406@section Basic Startup Options 407 408@table @samp 409@item -V 410@itemx --version 411Display the version of Wget. 412 413@item -h 414@itemx --help 415Print a help message describing all of Wget's command-line options. 416 417@item -b 418@itemx --background 419Go to background immediately after startup. If no output file is 420specified via the @samp{-o}, output is redirected to @file{wget-log}. 421 422@cindex execute wgetrc command 423@item -e @var{command} 424@itemx --execute @var{command} 425Execute @var{command} as if it were a part of @file{.wgetrc} 426(@pxref{Startup File}). A command thus invoked will be executed 427@emph{after} the commands in @file{.wgetrc}, thus taking precedence over 428them. If you need to specify more than one wgetrc command, use multiple 429instances of @samp{-e}. 430 431@end table 432 433@node Logging and Input File Options, Download Options, Basic Startup Options, Invoking 434@section Logging and Input File Options 435 436@table @samp 437@cindex output file 438@cindex log file 439@item -o @var{logfile} 440@itemx --output-file=@var{logfile} 441Log all messages to @var{logfile}. The messages are normally reported 442to standard error. 443 444@cindex append to log 445@item -a @var{logfile} 446@itemx --append-output=@var{logfile} 447Append to @var{logfile}. This is the same as @samp{-o}, only it appends 448to @var{logfile} instead of overwriting the old log file. If 449@var{logfile} does not exist, a new file is created. 450 451@cindex debug 452@item -d 453@itemx --debug 454Turn on debug output, meaning various information important to the 455developers of Wget if it does not work properly. Your system 456administrator may have chosen to compile Wget without debug support, in 457which case @samp{-d} will not work. Please note that compiling with 458debug support is always safe---Wget compiled with the debug support will 459@emph{not} print any debug info unless requested with @samp{-d}. 460@xref{Reporting Bugs}, for more information on how to use @samp{-d} for 461sending bug reports. 462 463@cindex quiet 464@item -q 465@itemx --quiet 466Turn off Wget's output. 467 468@cindex verbose 469@item -v 470@itemx --verbose 471Turn on verbose output, with all the available data. The default output 472is verbose. 473 474@item -nv 475@itemx --no-verbose 476Turn off verbose without being completely quiet (use @samp{-q} for 477that), which means that error messages and basic information still get 478printed. 479 480@cindex input-file 481@item -i @var{file} 482@itemx --input-file=@var{file} 483Read @sc{url}s from a local or external @var{file}. If @samp{-} is 484specified as @var{file}, @sc{url}s are read from the standard input. 485(Use @samp{./-} to read from a file literally named @samp{-}.) 486 487If this function is used, no @sc{url}s need be present on the command 488line. If there are @sc{url}s both on the command line and in an input 489file, those on the command lines will be the first ones to be 490retrieved. If @samp{--force-html} is not specified, then @var{file} 491should consist of a series of URLs, one per line. 492 493However, if you specify @samp{--force-html}, the document will be 494regarded as @samp{html}. In that case you may have problems with 495relative links, which you can solve either by adding @code{<base 496href="@var{url}">} to the documents or by specifying 497@samp{--base=@var{url}} on the command line. 498 499If the @var{file} is an external one, the document will be automatically 500treated as @samp{html} if the Content-Type matches @samp{text/html}. 501Furthermore, the @var{file}'s location will be implicitly used as base 502href if none was specified. 503 504@cindex force html 505@item -F 506@itemx --force-html 507When input is read from a file, force it to be treated as an @sc{html} 508file. This enables you to retrieve relative links from existing 509@sc{html} files on your local disk, by adding @code{<base 510href="@var{url}">} to @sc{html}, or using the @samp{--base} command-line 511option. 512 513@cindex base for relative links in input file 514@item -B @var{URL} 515@itemx --base=@var{URL} 516Resolves relative links using @var{URL} as the point of reference, 517when reading links from an HTML file specified via the 518@samp{-i}/@samp{--input-file} option (together with 519@samp{--force-html}, or when the input file was fetched remotely from 520a server describing it as @sc{html}). This is equivalent to the 521presence of a @code{BASE} tag in the @sc{html} input file, with 522@var{URL} as the value for the @code{href} attribute. 523 524For instance, if you specify @samp{http://foo/bar/a.html} for 525@var{URL}, and Wget reads @samp{../baz/b.html} from the input file, it 526would be resolved to @samp{http://foo/baz/b.html}. 527@end table 528 529@node Download Options, Directory Options, Logging and Input File Options, Invoking 530@section Download Options 531 532@table @samp 533@cindex bind address 534@cindex client IP address 535@cindex IP address, client 536@item --bind-address=@var{ADDRESS} 537When making client TCP/IP connections, bind to @var{ADDRESS} on 538the local machine. @var{ADDRESS} may be specified as a hostname or IP 539address. This option can be useful if your machine is bound to multiple 540IPs. 541 542@cindex retries 543@cindex tries 544@cindex number of retries 545@item -t @var{number} 546@itemx --tries=@var{number} 547Set number of retries to @var{number}. Specify 0 or @samp{inf} for 548infinite retrying. The default is to retry 20 times, with the exception 549of fatal errors like ``connection refused'' or ``not found'' (404), 550which are not retried. 551 552@item -O @var{file} 553@itemx --output-document=@var{file} 554The documents will not be written to the appropriate files, but all 555will be concatenated together and written to @var{file}. If @samp{-} 556is used as @var{file}, documents will be printed to standard output, 557disabling link conversion. (Use @samp{./-} to print to a file 558literally named @samp{-}.) 559 560Use of @samp{-O} is @emph{not} intended to mean simply ``use the name 561@var{file} instead of the one in the URL;'' rather, it is 562analogous to shell redirection: 563@samp{wget -O file http://foo} is intended to work like 564@samp{wget -O - http://foo > file}; @file{file} will be truncated 565immediately, and @emph{all} downloaded content will be written there. 566 567For this reason, @samp{-N} (for timestamp-checking) is not supported 568in combination with @samp{-O}: since @var{file} is always newly 569created, it will always have a very new timestamp. A warning will be 570issued if this combination is used. 571 572Similarly, using @samp{-r} or @samp{-p} with @samp{-O} may not work as 573you expect: Wget won't just download the first file to @var{file} and 574then download the rest to their normal names: @emph{all} downloaded 575content will be placed in @var{file}. This was disabled in version 5761.11, but has been reinstated (with a warning) in 1.11.2, as there are 577some cases where this behavior can actually have some use. 578 579Note that a combination with @samp{-k} is only permitted when 580downloading a single document, as in that case it will just convert 581all relative URIs to external ones; @samp{-k} makes no sense for 582multiple URIs when they're all being downloaded to a single file. 583 584@cindex clobbering, file 585@cindex downloading multiple times 586@cindex no-clobber 587@item -nc 588@itemx --no-clobber 589If a file is downloaded more than once in the same directory, Wget's 590behavior depends on a few options, including @samp{-nc}. In certain 591cases, the local file will be @dfn{clobbered}, or overwritten, upon 592repeated download. In other cases it will be preserved. 593 594When running Wget without @samp{-N}, @samp{-nc}, @samp{-r}, or 595@samp{-p}, downloading the same file in the same directory will result 596in the original copy of @var{file} being preserved and the second copy 597being named @samp{@var{file}.1}. If that file is downloaded yet 598again, the third copy will be named @samp{@var{file}.2}, and so on. 599(This is also the behavior with @samp{-nd}, even if @samp{-r} or 600@samp{-p} are in effect.) When @samp{-nc} is specified, this behavior 601is suppressed, and Wget will refuse to download newer copies of 602@samp{@var{file}}. Therefore, ``@code{no-clobber}'' is actually a 603misnomer in this mode---it's not clobbering that's prevented (as the 604numeric suffixes were already preventing clobbering), but rather the 605multiple version saving that's prevented. 606 607When running Wget with @samp{-r} or @samp{-p}, but without @samp{-N}, 608@samp{-nd}, or @samp{-nc}, re-downloading a file will result in the 609new copy simply overwriting the old. Adding @samp{-nc} will prevent 610this behavior, instead causing the original version to be preserved 611and any newer copies on the server to be ignored. 612 613When running Wget with @samp{-N}, with or without @samp{-r} or 614@samp{-p}, the decision as to whether or not to download a newer copy 615of a file depends on the local and remote timestamp and size of the 616file (@pxref{Time-Stamping}). @samp{-nc} may not be specified at the 617same time as @samp{-N}. 618 619Note that when @samp{-nc} is specified, files with the suffixes 620@samp{.html} or @samp{.htm} will be loaded from the local disk and 621parsed as if they had been retrieved from the Web. 622 623@cindex continue retrieval 624@cindex incomplete downloads 625@cindex resume download 626@item -c 627@itemx --continue 628Continue getting a partially-downloaded file. This is useful when you 629want to finish up a download started by a previous instance of Wget, or 630by another program. For instance: 631 632@example 633wget -c ftp://sunsite.doc.ic.ac.uk/ls-lR.Z 634@end example 635 636If there is a file named @file{ls-lR.Z} in the current directory, Wget 637will assume that it is the first portion of the remote file, and will 638ask the server to continue the retrieval from an offset equal to the 639length of the local file. 640 641Note that you don't need to specify this option if you just want the 642current invocation of Wget to retry downloading a file should the 643connection be lost midway through. This is the default behavior. 644@samp{-c} only affects resumption of downloads started @emph{prior} to 645this invocation of Wget, and whose local files are still sitting around. 646 647Without @samp{-c}, the previous example would just download the remote 648file to @file{ls-lR.Z.1}, leaving the truncated @file{ls-lR.Z} file 649alone. 650 651Beginning with Wget 1.7, if you use @samp{-c} on a non-empty file, and 652it turns out that the server does not support continued downloading, 653Wget will refuse to start the download from scratch, which would 654effectively ruin existing contents. If you really want the download to 655start from scratch, remove the file. 656 657Also beginning with Wget 1.7, if you use @samp{-c} on a file which is of 658equal size as the one on the server, Wget will refuse to download the 659file and print an explanatory message. The same happens when the file 660is smaller on the server than locally (presumably because it was changed 661on the server since your last download attempt)---because ``continuing'' 662is not meaningful, no download occurs. 663 664On the other side of the coin, while using @samp{-c}, any file that's 665bigger on the server than locally will be considered an incomplete 666download and only @code{(length(remote) - length(local))} bytes will be 667downloaded and tacked onto the end of the local file. This behavior can 668be desirable in certain cases---for instance, you can use @samp{wget -c} 669to download just the new portion that's been appended to a data 670collection or log file. 671 672However, if the file is bigger on the server because it's been 673@emph{changed}, as opposed to just @emph{appended} to, you'll end up 674with a garbled file. Wget has no way of verifying that the local file 675is really a valid prefix of the remote file. You need to be especially 676careful of this when using @samp{-c} in conjunction with @samp{-r}, 677since every file will be considered as an "incomplete download" candidate. 678 679Another instance where you'll get a garbled file if you try to use 680@samp{-c} is if you have a lame @sc{http} proxy that inserts a 681``transfer interrupted'' string into the local file. In the future a 682``rollback'' option may be added to deal with this case. 683 684Note that @samp{-c} only works with @sc{ftp} servers and with @sc{http} 685servers that support the @code{Range} header. 686 687@cindex progress indicator 688@cindex dot style 689@item --progress=@var{type} 690Select the type of the progress indicator you wish to use. Legal 691indicators are ``dot'' and ``bar''. 692 693The ``bar'' indicator is used by default. It draws an @sc{ascii} progress 694bar graphics (a.k.a ``thermometer'' display) indicating the status of 695retrieval. If the output is not a TTY, the ``dot'' bar will be used by 696default. 697 698Use @samp{--progress=dot} to switch to the ``dot'' display. It traces 699the retrieval by printing dots on the screen, each dot representing a 700fixed amount of downloaded data. 701 702When using the dotted retrieval, you may also set the @dfn{style} by 703specifying the type as @samp{dot:@var{style}}. Different styles assign 704different meaning to one dot. With the @code{default} style each dot 705represents 1K, there are ten dots in a cluster and 50 dots in a line. 706The @code{binary} style has a more ``computer''-like orientation---8K 707dots, 16-dots clusters and 48 dots per line (which makes for 384K 708lines). The @code{mega} style is suitable for downloading very large 709files---each dot represents 64K retrieved, there are eight dots in a 710cluster, and 48 dots on each line (so each line contains 3M). 711 712Note that you can set the default style using the @code{progress} 713command in @file{.wgetrc}. That setting may be overridden from the 714command line. The exception is that, when the output is not a TTY, the 715``dot'' progress will be favored over ``bar''. To force the bar output, 716use @samp{--progress=bar:force}. 717 718@item -N 719@itemx --timestamping 720Turn on time-stamping. @xref{Time-Stamping}, for details. 721 722@cindex server response, print 723@item -S 724@itemx --server-response 725Print the headers sent by @sc{http} servers and responses sent by 726@sc{ftp} servers. 727 728@cindex Wget as spider 729@cindex spider 730@item --spider 731When invoked with this option, Wget will behave as a Web @dfn{spider}, 732which means that it will not download the pages, just check that they 733are there. For example, you can use Wget to check your bookmarks: 734 735@example 736wget --spider --force-html -i bookmarks.html 737@end example 738 739This feature needs much more work for Wget to get close to the 740functionality of real web spiders. 741 742@cindex timeout 743@item -T seconds 744@itemx --timeout=@var{seconds} 745Set the network timeout to @var{seconds} seconds. This is equivalent 746to specifying @samp{--dns-timeout}, @samp{--connect-timeout}, and 747@samp{--read-timeout}, all at the same time. 748 749When interacting with the network, Wget can check for timeout and 750abort the operation if it takes too long. This prevents anomalies 751like hanging reads and infinite connects. The only timeout enabled by 752default is a 900-second read timeout. Setting a timeout to 0 disables 753it altogether. Unless you know what you are doing, it is best not to 754change the default timeout settings. 755 756All timeout-related options accept decimal values, as well as 757subsecond values. For example, @samp{0.1} seconds is a legal (though 758unwise) choice of timeout. Subsecond timeouts are useful for checking 759server response times or for testing network latency. 760 761@cindex DNS timeout 762@cindex timeout, DNS 763@item --dns-timeout=@var{seconds} 764Set the DNS lookup timeout to @var{seconds} seconds. DNS lookups that 765don't complete within the specified time will fail. By default, there 766is no timeout on DNS lookups, other than that implemented by system 767libraries. 768 769@cindex connect timeout 770@cindex timeout, connect 771@item --connect-timeout=@var{seconds} 772Set the connect timeout to @var{seconds} seconds. TCP connections that 773take longer to establish will be aborted. By default, there is no 774connect timeout, other than that implemented by system libraries. 775 776@cindex read timeout 777@cindex timeout, read 778@item --read-timeout=@var{seconds} 779Set the read (and write) timeout to @var{seconds} seconds. The 780``time'' of this timeout refers to @dfn{idle time}: if, at any point in 781the download, no data is received for more than the specified number 782of seconds, reading fails and the download is restarted. This option 783does not directly affect the duration of the entire download. 784 785Of course, the remote server may choose to terminate the connection 786sooner than this option requires. The default read timeout is 900 787seconds. 788 789@cindex bandwidth, limit 790@cindex rate, limit 791@cindex limit bandwidth 792@item --limit-rate=@var{amount} 793Limit the download speed to @var{amount} bytes per second. Amount may 794be expressed in bytes, kilobytes with the @samp{k} suffix, or megabytes 795with the @samp{m} suffix. For example, @samp{--limit-rate=20k} will 796limit the retrieval rate to 20KB/s. This is useful when, for whatever 797reason, you don't want Wget to consume the entire available bandwidth. 798 799This option allows the use of decimal numbers, usually in conjunction 800with power suffixes; for example, @samp{--limit-rate=2.5k} is a legal 801value. 802 803Note that Wget implements the limiting by sleeping the appropriate 804amount of time after a network read that took less time than specified 805by the rate. Eventually this strategy causes the TCP transfer to slow 806down to approximately the specified rate. However, it may take some 807time for this balance to be achieved, so don't be surprised if limiting 808the rate doesn't work well with very small files. 809 810@cindex pause 811@cindex wait 812@item -w @var{seconds} 813@itemx --wait=@var{seconds} 814Wait the specified number of seconds between the retrievals. Use of 815this option is recommended, as it lightens the server load by making the 816requests less frequent. Instead of in seconds, the time can be 817specified in minutes using the @code{m} suffix, in hours using @code{h} 818suffix, or in days using @code{d} suffix. 819 820Specifying a large value for this option is useful if the network or the 821destination host is down, so that Wget can wait long enough to 822reasonably expect the network error to be fixed before the retry. The 823waiting interval specified by this function is influenced by 824@code{--random-wait}, which see. 825 826@cindex retries, waiting between 827@cindex waiting between retries 828@item --waitretry=@var{seconds} 829If you don't want Wget to wait between @emph{every} retrieval, but only 830between retries of failed downloads, you can use this option. Wget will 831use @dfn{linear backoff}, waiting 1 second after the first failure on a 832given file, then waiting 2 seconds after the second failure on that 833file, up to the maximum number of @var{seconds} you specify. Therefore, 834a value of 10 will actually make Wget wait up to (1 + 2 + ... + 10) = 55 835seconds per file. 836 837By default, Wget will assume a value of 10 seconds. 838 839@cindex wait, random 840@cindex random wait 841@item --random-wait 842Some web sites may perform log analysis to identify retrieval programs 843such as Wget by looking for statistically significant similarities in 844the time between requests. This option causes the time between requests 845to vary between 0.5 and 1.5 * @var{wait} seconds, where @var{wait} was 846specified using the @samp{--wait} option, in order to mask Wget's 847presence from such analysis. 848 849A 2001 article in a publication devoted to development on a popular 850consumer platform provided code to perform this analysis on the fly. 851Its author suggested blocking at the class C address level to ensure 852automated retrieval programs were blocked despite changing DHCP-supplied 853addresses. 854 855The @samp{--random-wait} option was inspired by this ill-advised 856recommendation to block many unrelated users from a web site due to the 857actions of one. 858 859@cindex proxy 860@itemx --no-proxy 861Don't use proxies, even if the appropriate @code{*_proxy} environment 862variable is defined. 863 864@c man end 865For more information about the use of proxies with Wget, @xref{Proxies}. 866@c man begin OPTIONS 867 868@cindex quota 869@item -Q @var{quota} 870@itemx --quota=@var{quota} 871Specify download quota for automatic retrievals. The value can be 872specified in bytes (default), kilobytes (with @samp{k} suffix), or 873megabytes (with @samp{m} suffix). 874 875Note that quota will never affect downloading a single file. So if you 876specify @samp{wget -Q10k ftp://wuarchive.wustl.edu/ls-lR.gz}, all of the 877@file{ls-lR.gz} will be downloaded. The same goes even when several 878@sc{url}s are specified on the command-line. However, quota is 879respected when retrieving either recursively, or from an input file. 880Thus you may safely type @samp{wget -Q2m -i sites}---download will be 881aborted when the quota is exceeded. 882 883Setting quota to 0 or to @samp{inf} unlimits the download quota. 884 885@cindex DNS cache 886@cindex caching of DNS lookups 887@item --no-dns-cache 888Turn off caching of DNS lookups. Normally, Wget remembers the IP 889addresses it looked up from DNS so it doesn't have to repeatedly 890contact the DNS server for the same (typically small) set of hosts it 891retrieves from. This cache exists in memory only; a new Wget run will 892contact DNS again. 893 894However, it has been reported that in some situations it is not 895desirable to cache host names, even for the duration of a 896short-running application like Wget. With this option Wget issues a 897new DNS lookup (more precisely, a new call to @code{gethostbyname} or 898@code{getaddrinfo}) each time it makes a new connection. Please note 899that this option will @emph{not} affect caching that might be 900performed by the resolving library or by an external caching layer, 901such as NSCD. 902 903If you don't understand exactly what this option does, you probably 904won't need it. 905 906@cindex file names, restrict 907@cindex Windows file names 908@item --restrict-file-names=@var{modes} 909Change which characters found in remote URLs must be escaped during 910generation of local filenames. Characters that are @dfn{restricted} 911by this option are escaped, i.e. replaced with @samp{%HH}, where 912@samp{HH} is the hexadecimal number that corresponds to the restricted 913character. This option may also be used to force all alphabetical 914cases to be either lower- or uppercase. 915 916By default, Wget escapes the characters that are not valid or safe as 917part of file names on your operating system, as well as control 918characters that are typically unprintable. This option is useful for 919changing these defaults, perhaps because you are downloading to a 920non-native partition, or because you want to disable escaping of the 921control characters, or you want to further restrict characters to only 922those in the @sc{ascii} range of values. 923 924The @var{modes} are a comma-separated set of text values. The 925acceptable values are @samp{unix}, @samp{windows}, @samp{nocontrol}, 926@samp{ascii}, @samp{lowercase}, and @samp{uppercase}. The values 927@samp{unix} and @samp{windows} are mutually exclusive (one will 928override the other), as are @samp{lowercase} and 929@samp{uppercase}. Those last are special cases, as they do not change 930the set of characters that would be escaped, but rather force local 931file paths to be converted either to lower- or uppercase. 932 933When ``unix'' is specified, Wget escapes the character @samp{/} and 934the control characters in the ranges 0--31 and 128--159. This is the 935default on Unix-like operating systems. 936 937When ``windows'' is given, Wget escapes the characters @samp{\}, 938@samp{|}, @samp{/}, @samp{:}, @samp{?}, @samp{"}, @samp{*}, @samp{<}, 939@samp{>}, and the control characters in the ranges 0--31 and 128--159. 940In addition to this, Wget in Windows mode uses @samp{+} instead of 941@samp{:} to separate host and port in local file names, and uses 942@samp{@@} instead of @samp{?} to separate the query portion of the file 943name from the rest. Therefore, a URL that would be saved as 944@samp{www.xemacs.org:4300/search.pl?input=blah} in Unix mode would be 945saved as @samp{www.xemacs.org+4300/search.pl@@input=blah} in Windows 946mode. This mode is the default on Windows. 947 948If you specify @samp{nocontrol}, then the escaping of the control 949characters is also switched off. This option may make sense 950when you are downloading URLs whose names contain UTF-8 characters, on 951a system which can save and display filenames in UTF-8 (some possible 952byte values used in UTF-8 byte sequences fall in the range of values 953designated by Wget as ``controls''). 954 955The @samp{ascii} mode is used to specify that any bytes whose values 956are outside the range of @sc{ascii} characters (that is, greater than 957127) shall be escaped. This can be useful when saving filenames 958whose encoding does not match the one used locally. 959 960@cindex IPv6 961@itemx -4 962@itemx --inet4-only 963@itemx -6 964@itemx --inet6-only 965Force connecting to IPv4 or IPv6 addresses. With @samp{--inet4-only} 966or @samp{-4}, Wget will only connect to IPv4 hosts, ignoring AAAA 967records in DNS, and refusing to connect to IPv6 addresses specified in 968URLs. Conversely, with @samp{--inet6-only} or @samp{-6}, Wget will 969only connect to IPv6 hosts and ignore A records and IPv4 addresses. 970 971Neither options should be needed normally. By default, an IPv6-aware 972Wget will use the address family specified by the host's DNS record. 973If the DNS responds with both IPv4 and IPv6 addresses, Wget will try 974them in sequence until it finds one it can connect to. (Also see 975@code{--prefer-family} option described below.) 976 977These options can be used to deliberately force the use of IPv4 or 978IPv6 address families on dual family systems, usually to aid debugging 979or to deal with broken network configuration. Only one of 980@samp{--inet6-only} and @samp{--inet4-only} may be specified at the 981same time. Neither option is available in Wget compiled without IPv6 982support. 983 984@item --prefer-family=none/IPv4/IPv6 985When given a choice of several addresses, connect to the addresses 986with specified address family first. The address order returned by 987DNS is used without change by default. 988 989This avoids spurious errors and connect attempts when accessing hosts 990that resolve to both IPv6 and IPv4 addresses from IPv4 networks. For 991example, @samp{www.kame.net} resolves to 992@samp{2001:200:0:8002:203:47ff:fea5:3085} and to 993@samp{203.178.141.194}. When the preferred family is @code{IPv4}, the 994IPv4 address is used first; when the preferred family is @code{IPv6}, 995the IPv6 address is used first; if the specified value is @code{none}, 996the address order returned by DNS is used without change. 997 998Unlike @samp{-4} and @samp{-6}, this option doesn't inhibit access to 999any address family, it only changes the @emph{order} in which the 1000addresses are accessed. Also note that the reordering performed by 1001this option is @dfn{stable}---it doesn't affect order of addresses of 1002the same family. That is, the relative order of all IPv4 addresses 1003and of all IPv6 addresses remains intact in all cases. 1004 1005@item --retry-connrefused 1006Consider ``connection refused'' a transient error and try again. 1007Normally Wget gives up on a URL when it is unable to connect to the 1008site because failure to connect is taken as a sign that the server is 1009not running at all and that retries would not help. This option is 1010for mirroring unreliable sites whose servers tend to disappear for 1011short periods of time. 1012 1013@cindex user 1014@cindex password 1015@cindex authentication 1016@item --user=@var{user} 1017@itemx --password=@var{password} 1018Specify the username @var{user} and password @var{password} for both 1019@sc{ftp} and @sc{http} file retrieval. These parameters can be overridden 1020using the @samp{--ftp-user} and @samp{--ftp-password} options for 1021@sc{ftp} connections and the @samp{--http-user} and @samp{--http-password} 1022options for @sc{http} connections. 1023 1024@item --ask-password 1025Prompt for a password for each connection established. Cannot be specified 1026when @samp{--password} is being used, because they are mutually exclusive. 1027 1028@cindex iri support 1029@cindex idn support 1030@item --no-iri 1031 1032Turn off internationalized URI (IRI) support. Use @samp{--iri} to 1033turn it on. IRI support is activated by default. 1034 1035You can set the default state of IRI support using the @code{iri} 1036command in @file{.wgetrc}. That setting may be overridden from the 1037command line. 1038 1039@cindex local encoding 1040@item --local-encoding=@var{encoding} 1041 1042Force Wget to use @var{encoding} as the default system encoding. That affects 1043how Wget converts URLs specified as arguments from locale to @sc{utf-8} for 1044IRI support. 1045 1046Wget use the function @code{nl_langinfo()} and then the @code{CHARSET} 1047environment variable to get the locale. If it fails, @sc{ascii} is used. 1048 1049You can set the default local encoding using the @code{local_encoding} 1050command in @file{.wgetrc}. That setting may be overridden from the 1051command line. 1052 1053@cindex remote encoding 1054@item --remote-encoding=@var{encoding} 1055 1056Force Wget to use @var{encoding} as the default remote server encoding. 1057That affects how Wget converts URIs found in files from remote encoding 1058to @sc{utf-8} during a recursive fetch. This options is only useful for 1059IRI support, for the interpretation of non-@sc{ascii} characters. 1060 1061For HTTP, remote encoding can be found in HTTP @code{Content-Type} 1062header and in HTML @code{Content-Type http-equiv} meta tag. 1063 1064You can set the default encoding using the @code{remoteencoding} 1065command in @file{.wgetrc}. That setting may be overridden from the 1066command line. 1067@end table 1068 1069@node Directory Options, HTTP Options, Download Options, Invoking 1070@section Directory Options 1071 1072@table @samp 1073@item -nd 1074@itemx --no-directories 1075Do not create a hierarchy of directories when retrieving recursively. 1076With this option turned on, all files will get saved to the current 1077directory, without clobbering (if a name shows up more than once, the 1078filenames will get extensions @samp{.n}). 1079 1080@item -x 1081@itemx --force-directories 1082The opposite of @samp{-nd}---create a hierarchy of directories, even if 1083one would not have been created otherwise. E.g. @samp{wget -x 1084http://fly.srk.fer.hr/robots.txt} will save the downloaded file to 1085@file{fly.srk.fer.hr/robots.txt}. 1086 1087@item -nH 1088@itemx --no-host-directories 1089Disable generation of host-prefixed directories. By default, invoking 1090Wget with @samp{-r http://fly.srk.fer.hr/} will create a structure of 1091directories beginning with @file{fly.srk.fer.hr/}. This option disables 1092such behavior. 1093 1094@item --protocol-directories 1095Use the protocol name as a directory component of local file names. For 1096example, with this option, @samp{wget -r http://@var{host}} will save to 1097@samp{http/@var{host}/...} rather than just to @samp{@var{host}/...}. 1098 1099@cindex cut directories 1100@item --cut-dirs=@var{number} 1101Ignore @var{number} directory components. This is useful for getting a 1102fine-grained control over the directory where recursive retrieval will 1103be saved. 1104 1105Take, for example, the directory at 1106@samp{ftp://ftp.xemacs.org/pub/xemacs/}. If you retrieve it with 1107@samp{-r}, it will be saved locally under 1108@file{ftp.xemacs.org/pub/xemacs/}. While the @samp{-nH} option can 1109remove the @file{ftp.xemacs.org/} part, you are still stuck with 1110@file{pub/xemacs}. This is where @samp{--cut-dirs} comes in handy; it 1111makes Wget not ``see'' @var{number} remote directory components. Here 1112are several examples of how @samp{--cut-dirs} option works. 1113 1114@example 1115@group 1116No options -> ftp.xemacs.org/pub/xemacs/ 1117-nH -> pub/xemacs/ 1118-nH --cut-dirs=1 -> xemacs/ 1119-nH --cut-dirs=2 -> . 1120 1121--cut-dirs=1 -> ftp.xemacs.org/xemacs/ 1122... 1123@end group 1124@end example 1125 1126If you just want to get rid of the directory structure, this option is 1127similar to a combination of @samp{-nd} and @samp{-P}. However, unlike 1128@samp{-nd}, @samp{--cut-dirs} does not lose with subdirectories---for 1129instance, with @samp{-nH --cut-dirs=1}, a @file{beta/} subdirectory will 1130be placed to @file{xemacs/beta}, as one would expect. 1131 1132@cindex directory prefix 1133@item -P @var{prefix} 1134@itemx --directory-prefix=@var{prefix} 1135Set directory prefix to @var{prefix}. The @dfn{directory prefix} is the 1136directory where all other files and subdirectories will be saved to, 1137i.e. the top of the retrieval tree. The default is @samp{.} (the 1138current directory). 1139@end table 1140 1141@node HTTP Options, HTTPS (SSL/TLS) Options, Directory Options, Invoking 1142@section HTTP Options 1143 1144@table @samp 1145@cindex default page name 1146@cindex index.html 1147@item --default-page=@var{name} 1148Use @var{name} as the default file name when it isn't known (i.e., for 1149URLs that end in a slash), instead of @file{index.html}. 1150 1151@cindex .html extension 1152@cindex .css extension 1153@item -E 1154@itemx --adjust-extension 1155If a file of type @samp{application/xhtml+xml} or @samp{text/html} is 1156downloaded and the URL does not end with the regexp 1157@samp{\.[Hh][Tt][Mm][Ll]?}, this option will cause the suffix @samp{.html} 1158to be appended to the local filename. This is useful, for instance, when 1159you're mirroring a remote site that uses @samp{.asp} pages, but you want 1160the mirrored pages to be viewable on your stock Apache server. Another 1161good use for this is when you're downloading CGI-generated materials. A URL 1162like @samp{http://site.com/article.cgi?25} will be saved as 1163@file{article.cgi?25.html}. 1164 1165Note that filenames changed in this way will be re-downloaded every time 1166you re-mirror a site, because Wget can't tell that the local 1167@file{@var{X}.html} file corresponds to remote URL @samp{@var{X}} (since 1168it doesn't yet know that the URL produces output of type 1169@samp{text/html} or @samp{application/xhtml+xml}. To prevent this 1170re-downloading, you must use @samp{-k} and @samp{-K} so that the original 1171version of the file will be saved as @file{@var{X}.orig} (@pxref{Recursive 1172Retrieval Options}). 1173 1174As of version 1.12, Wget will also ensure that any downloaded files of 1175type @samp{text/css} end in the suffix @samp{.css}, and the option was 1176renamed from @samp{--html-extension}, to better reflect its new 1177behavior. The old option name is still acceptable, but should now be 1178considered deprecated. 1179 1180At some point in the future, this option may well be expanded to 1181include suffixes for other types of content, including content types 1182that are not parsed by Wget. 1183 1184@cindex http user 1185@cindex http password 1186@cindex authentication 1187@item --http-user=@var{user} 1188@itemx --http-password=@var{password} 1189Specify the username @var{user} and password @var{password} on an 1190@sc{http} server. According to the type of the challenge, Wget will 1191encode them using either the @code{basic} (insecure), 1192the @code{digest}, or the Windows @code{NTLM} authentication scheme. 1193 1194Another way to specify username and password is in the @sc{url} itself 1195(@pxref{URL Format}). Either method reveals your password to anyone who 1196bothers to run @code{ps}. To prevent the passwords from being seen, 1197store them in @file{.wgetrc} or @file{.netrc}, and make sure to protect 1198those files from other users with @code{chmod}. If the passwords are 1199really important, do not leave them lying in those files either---edit 1200the files and delete them after Wget has started the download. 1201 1202@iftex 1203For more information about security issues with Wget, @xref{Security 1204Considerations}. 1205@end iftex 1206 1207@cindex Keep-Alive, turning off 1208@cindex Persistent Connections, disabling 1209@item --no-http-keep-alive 1210Turn off the ``keep-alive'' feature for HTTP downloads. Normally, Wget 1211asks the server to keep the connection open so that, when you download 1212more than one document from the same server, they get transferred over 1213the same TCP connection. This saves time and at the same time reduces 1214the load on the server. 1215 1216This option is useful when, for some reason, persistent (keep-alive) 1217connections don't work for you, for example due to a server bug or due 1218to the inability of server-side scripts to cope with the connections. 1219 1220@cindex proxy 1221@cindex cache 1222@item --no-cache 1223Disable server-side cache. In this case, Wget will send the remote 1224server an appropriate directive (@samp{Pragma: no-cache}) to get the 1225file from the remote service, rather than returning the cached version. 1226This is especially useful for retrieving and flushing out-of-date 1227documents on proxy servers. 1228 1229Caching is allowed by default. 1230 1231@cindex cookies 1232@item --no-cookies 1233Disable the use of cookies. Cookies are a mechanism for maintaining 1234server-side state. The server sends the client a cookie using the 1235@code{Set-Cookie} header, and the client responds with the same cookie 1236upon further requests. Since cookies allow the server owners to keep 1237track of visitors and for sites to exchange this information, some 1238consider them a breach of privacy. The default is to use cookies; 1239however, @emph{storing} cookies is not on by default. 1240 1241@cindex loading cookies 1242@cindex cookies, loading 1243@item --load-cookies @var{file} 1244Load cookies from @var{file} before the first HTTP retrieval. 1245@var{file} is a textual file in the format originally used by Netscape's 1246@file{cookies.txt} file. 1247 1248You will typically use this option when mirroring sites that require 1249that you be logged in to access some or all of their content. The login 1250process typically works by the web server issuing an @sc{http} cookie 1251upon receiving and verifying your credentials. The cookie is then 1252resent by the browser when accessing that part of the site, and so 1253proves your identity. 1254 1255Mirroring such a site requires Wget to send the same cookies your 1256browser sends when communicating with the site. This is achieved by 1257@samp{--load-cookies}---simply point Wget to the location of the 1258@file{cookies.txt} file, and it will send the same cookies your browser 1259would send in the same situation. Different browsers keep textual 1260cookie files in different locations: 1261 1262@table @asis 1263@item Netscape 4.x. 1264The cookies are in @file{~/.netscape/cookies.txt}. 1265 1266@item Mozilla and Netscape 6.x. 1267Mozilla's cookie file is also named @file{cookies.txt}, located 1268somewhere under @file{~/.mozilla}, in the directory of your profile. 1269The full path usually ends up looking somewhat like 1270@file{~/.mozilla/default/@var{some-weird-string}/cookies.txt}. 1271 1272@item Internet Explorer. 1273You can produce a cookie file Wget can use by using the File menu, 1274Import and Export, Export Cookies. This has been tested with Internet 1275Explorer 5; it is not guaranteed to work with earlier versions. 1276 1277@item Other browsers. 1278If you are using a different browser to create your cookies, 1279@samp{--load-cookies} will only work if you can locate or produce a 1280cookie file in the Netscape format that Wget expects. 1281@end table 1282 1283If you cannot use @samp{--load-cookies}, there might still be an 1284alternative. If your browser supports a ``cookie manager'', you can use 1285it to view the cookies used when accessing the site you're mirroring. 1286Write down the name and value of the cookie, and manually instruct Wget 1287to send those cookies, bypassing the ``official'' cookie support: 1288 1289@example 1290wget --no-cookies --header "Cookie: @var{name}=@var{value}" 1291@end example 1292 1293@cindex saving cookies 1294@cindex cookies, saving 1295@item --save-cookies @var{file} 1296Save cookies to @var{file} before exiting. This will not save cookies 1297that have expired or that have no expiry time (so-called ``session 1298cookies''), but also see @samp{--keep-session-cookies}. 1299 1300@cindex cookies, session 1301@cindex session cookies 1302@item --keep-session-cookies 1303When specified, causes @samp{--save-cookies} to also save session 1304cookies. Session cookies are normally not saved because they are 1305meant to be kept in memory and forgotten when you exit the browser. 1306Saving them is useful on sites that require you to log in or to visit 1307the home page before you can access some pages. With this option, 1308multiple Wget runs are considered a single browser session as far as 1309the site is concerned. 1310 1311Since the cookie file format does not normally carry session cookies, 1312Wget marks them with an expiry timestamp of 0. Wget's 1313@samp{--load-cookies} recognizes those as session cookies, but it might 1314confuse other browsers. Also note that cookies so loaded will be 1315treated as other session cookies, which means that if you want 1316@samp{--save-cookies} to preserve them again, you must use 1317@samp{--keep-session-cookies} again. 1318 1319@cindex Content-Length, ignore 1320@cindex ignore length 1321@item --ignore-length 1322Unfortunately, some @sc{http} servers (@sc{cgi} programs, to be more 1323precise) send out bogus @code{Content-Length} headers, which makes Wget 1324go wild, as it thinks not all the document was retrieved. You can spot 1325this syndrome if Wget retries getting the same document again and again, 1326each time claiming that the (otherwise normal) connection has closed on 1327the very same byte. 1328 1329With this option, Wget will ignore the @code{Content-Length} header---as 1330if it never existed. 1331 1332@cindex header, add 1333@item --header=@var{header-line} 1334Send @var{header-line} along with the rest of the headers in each 1335@sc{http} request. The supplied header is sent as-is, which means it 1336must contain name and value separated by colon, and must not contain 1337newlines. 1338 1339You may define more than one additional header by specifying 1340@samp{--header} more than once. 1341 1342@example 1343@group 1344wget --header='Accept-Charset: iso-8859-2' \ 1345 --header='Accept-Language: hr' \ 1346 http://fly.srk.fer.hr/ 1347@end group 1348@end example 1349 1350Specification of an empty string as the header value will clear all 1351previous user-defined headers. 1352 1353As of Wget 1.10, this option can be used to override headers otherwise 1354generated automatically. This example instructs Wget to connect to 1355localhost, but to specify @samp{foo.bar} in the @code{Host} header: 1356 1357@example 1358wget --header="Host: foo.bar" http://localhost/ 1359@end example 1360 1361In versions of Wget prior to 1.10 such use of @samp{--header} caused 1362sending of duplicate headers. 1363 1364@cindex redirect 1365@item --max-redirect=@var{number} 1366Specifies the maximum number of redirections to follow for a resource. 1367The default is 20, which is usually far more than necessary. However, on 1368those occasions where you want to allow more (or fewer), this is the 1369option to use. 1370 1371@cindex proxy user 1372@cindex proxy password 1373@cindex proxy authentication 1374@item --proxy-user=@var{user} 1375@itemx --proxy-password=@var{password} 1376Specify the username @var{user} and password @var{password} for 1377authentication on a proxy server. Wget will encode them using the 1378@code{basic} authentication scheme. 1379 1380Security considerations similar to those with @samp{--http-password} 1381pertain here as well. 1382 1383@cindex http referer 1384@cindex referer, http 1385@item --referer=@var{url} 1386Include `Referer: @var{url}' header in HTTP request. Useful for 1387retrieving documents with server-side processing that assume they are 1388always being retrieved by interactive web browsers and only come out 1389properly when Referer is set to one of the pages that point to them. 1390 1391@cindex server response, save 1392@item --save-headers 1393Save the headers sent by the @sc{http} server to the file, preceding the 1394actual contents, with an empty line as the separator. 1395 1396@cindex user-agent 1397@item -U @var{agent-string} 1398@itemx --user-agent=@var{agent-string} 1399Identify as @var{agent-string} to the @sc{http} server. 1400 1401The @sc{http} protocol allows the clients to identify themselves using a 1402@code{User-Agent} header field. This enables distinguishing the 1403@sc{www} software, usually for statistical purposes or for tracing of 1404protocol violations. Wget normally identifies as 1405@samp{Wget/@var{version}}, @var{version} being the current version 1406number of Wget. 1407 1408However, some sites have been known to impose the policy of tailoring 1409the output according to the @code{User-Agent}-supplied information. 1410While this is not such a bad idea in theory, it has been abused by 1411servers denying information to clients other than (historically) 1412Netscape or, more frequently, Microsoft Internet Explorer. This 1413option allows you to change the @code{User-Agent} line issued by Wget. 1414Use of this option is discouraged, unless you really know what you are 1415doing. 1416 1417Specifying empty user agent with @samp{--user-agent=""} instructs Wget 1418not to send the @code{User-Agent} header in @sc{http} requests. 1419 1420@cindex POST 1421@item --post-data=@var{string} 1422@itemx --post-file=@var{file} 1423Use POST as the method for all HTTP requests and send the specified 1424data in the request body. @samp{--post-data} sends @var{string} as 1425data, whereas @samp{--post-file} sends the contents of @var{file}. 1426Other than that, they work in exactly the same way. In particular, 1427they @emph{both} expect content of the form @code{key1=value1&key2=value2}, 1428with percent-encoding for special characters; the only difference is 1429that one expects its content as a command-line paramter and the other 1430accepts its content from a file. In particular, @samp{--post-file} is 1431@emph{not} for transmitting files as form attachments: those must 1432appear as @code{key=value} data (with appropriate percent-coding) just 1433like everything else. Wget does not currently support 1434@code{multipart/form-data} for transmitting POST data; only 1435@code{application/x-www-form-urlencoded}. Only one of 1436@samp{--post-data} and @samp{--post-file} should be specified. 1437 1438Please be aware that Wget needs to know the size of the POST data in 1439advance. Therefore the argument to @code{--post-file} must be a regular 1440file; specifying a FIFO or something like @file{/dev/stdin} won't work. 1441It's not quite clear how to work around this limitation inherent in 1442HTTP/1.0. Although HTTP/1.1 introduces @dfn{chunked} transfer that 1443doesn't require knowing the request length in advance, a client can't 1444use chunked unless it knows it's talking to an HTTP/1.1 server. And it 1445can't know that until it receives a response, which in turn requires the 1446request to have been completed -- a chicken-and-egg problem. 1447 1448Note: if Wget is redirected after the POST request is completed, it 1449will not send the POST data to the redirected URL. This is because 1450URLs that process POST often respond with a redirection to a regular 1451page, which does not desire or accept POST. It is not completely 1452clear that this behavior is optimal; if it doesn't work out, it might 1453be changed in the future. 1454 1455This example shows how to log to a server using POST and then proceed to 1456download the desired pages, presumably only accessible to authorized 1457users: 1458 1459@example 1460@group 1461# @r{Log in to the server. This can be done only once.} 1462wget --save-cookies cookies.txt \ 1463 --post-data 'user=foo&password=bar' \ 1464 http://server.com/auth.php 1465 1466# @r{Now grab the page or pages we care about.} 1467wget --load-cookies cookies.txt \ 1468 -p http://server.com/interesting/article.php 1469@end group 1470@end example 1471 1472If the server is using session cookies to track user authentication, 1473the above will not work because @samp{--save-cookies} will not save 1474them (and neither will browsers) and the @file{cookies.txt} file will 1475be empty. In that case use @samp{--keep-session-cookies} along with 1476@samp{--save-cookies} to force saving of session cookies. 1477 1478@cindex Content-Disposition 1479@item --content-disposition 1480 1481If this is set to on, experimental (not fully-functional) support for 1482@code{Content-Disposition} headers is enabled. This can currently result in 1483extra round-trips to the server for a @code{HEAD} request, and is known 1484to suffer from a few bugs, which is why it is not currently enabled by default. 1485 1486This option is useful for some file-downloading CGI programs that use 1487@code{Content-Disposition} headers to describe what the name of a 1488downloaded file should be. 1489 1490@cindex authentication 1491@item --auth-no-challenge 1492 1493If this option is given, Wget will send Basic HTTP authentication 1494information (plaintext username and password) for all requests, just 1495like Wget 1.10.2 and prior did by default. 1496 1497Use of this option is not recommended, and is intended only to support 1498some few obscure servers, which never send HTTP authentication 1499challenges, but accept unsolicited auth info, say, in addition to 1500form-based authentication. 1501 1502@end table 1503 1504@node HTTPS (SSL/TLS) Options, FTP Options, HTTP Options, Invoking 1505@section HTTPS (SSL/TLS) Options 1506 1507@cindex SSL 1508To support encrypted HTTP (HTTPS) downloads, Wget must be compiled 1509with an external SSL library, currently OpenSSL. If Wget is compiled 1510without SSL support, none of these options are available. 1511 1512@table @samp 1513@cindex SSL protocol, choose 1514@item --secure-protocol=@var{protocol} 1515Choose the secure protocol to be used. Legal values are @samp{auto}, 1516@samp{SSLv2}, @samp{SSLv3}, and @samp{TLSv1}. If @samp{auto} is used, 1517the SSL library is given the liberty of choosing the appropriate 1518protocol automatically, which is achieved by sending an SSLv2 greeting 1519and announcing support for SSLv3 and TLSv1. This is the default. 1520 1521Specifying @samp{SSLv2}, @samp{SSLv3}, or @samp{TLSv1} forces the use 1522of the corresponding protocol. This is useful when talking to old and 1523buggy SSL server implementations that make it hard for OpenSSL to 1524choose the correct protocol version. Fortunately, such servers are 1525quite rare. 1526 1527@cindex SSL certificate, check 1528@item --no-check-certificate 1529Don't check the server certificate against the available certificate 1530authorities. Also don't require the URL host name to match the common 1531name presented by the certificate. 1532 1533As of Wget 1.10, the default is to verify the server's certificate 1534against the recognized certificate authorities, breaking the SSL 1535handshake and aborting the download if the verification fails. 1536Although this provides more secure downloads, it does break 1537interoperability with some sites that worked with previous Wget 1538versions, particularly those using self-signed, expired, or otherwise 1539invalid certificates. This option forces an ``insecure'' mode of 1540operation that turns the certificate verification errors into warnings 1541and allows you to proceed. 1542 1543If you encounter ``certificate verification'' errors or ones saying 1544that ``common name doesn't match requested host name'', you can use 1545this option to bypass the verification and proceed with the download. 1546@emph{Only use this option if you are otherwise convinced of the 1547site's authenticity, or if you really don't care about the validity of 1548its certificate.} It is almost always a bad idea not to check the 1549certificates when transmitting confidential or important data. 1550 1551@cindex SSL certificate 1552@item --certificate=@var{file} 1553Use the client certificate stored in @var{file}. This is needed for 1554servers that are configured to require certificates from the clients 1555that connect to them. Normally a certificate is not required and this 1556switch is optional. 1557 1558@cindex SSL certificate type, specify 1559@item --certificate-type=@var{type} 1560Specify the type of the client certificate. Legal values are 1561@samp{PEM} (assumed by default) and @samp{DER}, also known as 1562@samp{ASN1}. 1563 1564@item --private-key=@var{file} 1565Read the private key from @var{file}. This allows you to provide the 1566private key in a file separate from the certificate. 1567 1568@item --private-key-type=@var{type} 1569Specify the type of the private key. Accepted values are @samp{PEM} 1570(the default) and @samp{DER}. 1571 1572@item --ca-certificate=@var{file} 1573Use @var{file} as the file with the bundle of certificate authorities 1574(``CA'') to verify the peers. The certificates must be in PEM format. 1575 1576Without this option Wget looks for CA certificates at the 1577system-specified locations, chosen at OpenSSL installation time. 1578 1579@cindex SSL certificate authority 1580@item --ca-directory=@var{directory} 1581Specifies directory containing CA certificates in PEM format. Each 1582file contains one CA certificate, and the file name is based on a hash 1583value derived from the certificate. This is achieved by processing a 1584certificate directory with the @code{c_rehash} utility supplied with 1585OpenSSL. Using @samp{--ca-directory} is more efficient than 1586@samp{--ca-certificate} when many certificates are installed because 1587it allows Wget to fetch certificates on demand. 1588 1589Without this option Wget looks for CA certificates at the 1590system-specified locations, chosen at OpenSSL installation time. 1591 1592@cindex entropy, specifying source of 1593@cindex randomness, specifying source of 1594@item --random-file=@var{file} 1595Use @var{file} as the source of random data for seeding the 1596pseudo-random number generator on systems without @file{/dev/random}. 1597 1598On such systems the SSL library needs an external source of randomness 1599to initialize. Randomness may be provided by EGD (see 1600@samp{--egd-file} below) or read from an external source specified by 1601the user. If this option is not specified, Wget looks for random data 1602in @code{$RANDFILE} or, if that is unset, in @file{$HOME/.rnd}. If 1603none of those are available, it is likely that SSL encryption will not 1604be usable. 1605 1606If you're getting the ``Could not seed OpenSSL PRNG; disabling SSL.'' 1607error, you should provide random data using some of the methods 1608described above. 1609 1610@cindex EGD 1611@item --egd-file=@var{file} 1612Use @var{file} as the EGD socket. EGD stands for @dfn{Entropy 1613Gathering Daemon}, a user-space program that collects data from 1614various unpredictable system sources and makes it available to other 1615programs that might need it. Encryption software, such as the SSL 1616library, needs sources of non-repeating randomness to seed the random 1617number generator used to produce cryptographically strong keys. 1618 1619OpenSSL allows the user to specify his own source of entropy using the 1620@code{RAND_FILE} environment variable. If this variable is unset, or 1621if the specified file does not produce enough randomness, OpenSSL will 1622read random data from EGD socket specified using this option. 1623 1624If this option is not specified (and the equivalent startup command is 1625not used), EGD is never contacted. EGD is not needed on modern Unix 1626systems that support @file{/dev/random}. 1627@end table 1628 1629@node FTP Options, Recursive Retrieval Options, HTTPS (SSL/TLS) Options, Invoking 1630@section FTP Options 1631 1632@table @samp 1633@cindex ftp user 1634@cindex ftp password 1635@cindex ftp authentication 1636@item --ftp-user=@var{user} 1637@itemx --ftp-password=@var{password} 1638Specify the username @var{user} and password @var{password} on an 1639@sc{ftp} server. Without this, or the corresponding startup option, 1640the password defaults to @samp{-wget@@}, normally used for anonymous 1641FTP. 1642 1643Another way to specify username and password is in the @sc{url} itself 1644(@pxref{URL Format}). Either method reveals your password to anyone who 1645bothers to run @code{ps}. To prevent the passwords from being seen, 1646store them in @file{.wgetrc} or @file{.netrc}, and make sure to protect 1647those files from other users with @code{chmod}. If the passwords are 1648really important, do not leave them lying in those files either---edit 1649the files and delete them after Wget has started the download. 1650 1651@iftex 1652For more information about security issues with Wget, @xref{Security 1653Considerations}. 1654@end iftex 1655 1656@cindex .listing files, removing 1657@item --no-remove-listing 1658Don't remove the temporary @file{.listing} files generated by @sc{ftp} 1659retrievals. Normally, these files contain the raw directory listings 1660received from @sc{ftp} servers. Not removing them can be useful for 1661debugging purposes, or when you want to be able to easily check on the 1662contents of remote server directories (e.g. to verify that a mirror 1663you're running is complete). 1664 1665Note that even though Wget writes to a known filename for this file, 1666this is not a security hole in the scenario of a user making 1667@file{.listing} a symbolic link to @file{/etc/passwd} or something and 1668asking @code{root} to run Wget in his or her directory. Depending on 1669the options used, either Wget will refuse to write to @file{.listing}, 1670making the globbing/recursion/time-stamping operation fail, or the 1671symbolic link will be deleted and replaced with the actual 1672@file{.listing} file, or the listing will be written to a 1673@file{.listing.@var{number}} file. 1674 1675Even though this situation isn't a problem, though, @code{root} should 1676never run Wget in a non-trusted user's directory. A user could do 1677something as simple as linking @file{index.html} to @file{/etc/passwd} 1678and asking @code{root} to run Wget with @samp{-N} or @samp{-r} so the file 1679will be overwritten. 1680 1681@cindex globbing, toggle 1682@item --no-glob 1683Turn off @sc{ftp} globbing. Globbing refers to the use of shell-like 1684special characters (@dfn{wildcards}), like @samp{*}, @samp{?}, @samp{[} 1685and @samp{]} to retrieve more than one file from the same directory at 1686once, like: 1687 1688@example 1689wget ftp://gnjilux.srk.fer.hr/*.msg 1690@end example 1691 1692By default, globbing will be turned on if the @sc{url} contains a 1693globbing character. This option may be used to turn globbing on or off 1694permanently. 1695 1696You may have to quote the @sc{url} to protect it from being expanded by 1697your shell. Globbing makes Wget look for a directory listing, which is 1698system-specific. This is why it currently works only with Unix @sc{ftp} 1699servers (and the ones emulating Unix @code{ls} output). 1700 1701@cindex passive ftp 1702@item --no-passive-ftp 1703Disable the use of the @dfn{passive} FTP transfer mode. Passive FTP 1704mandates that the client connect to the server to establish the data 1705connection rather than the other way around. 1706 1707If the machine is connected to the Internet directly, both passive and 1708active FTP should work equally well. Behind most firewall and NAT 1709configurations passive FTP has a better chance of working. However, 1710in some rare firewall configurations, active FTP actually works when 1711passive FTP doesn't. If you suspect this to be the case, use this 1712option, or set @code{passive_ftp=off} in your init file. 1713 1714@cindex symbolic links, retrieving 1715@item --retr-symlinks 1716Usually, when retrieving @sc{ftp} directories recursively and a symbolic 1717link is encountered, the linked-to file is not downloaded. Instead, a 1718matching symbolic link is created on the local filesystem. The 1719pointed-to file will not be downloaded unless this recursive retrieval 1720would have encountered it separately and downloaded it anyway. 1721 1722When @samp{--retr-symlinks} is specified, however, symbolic links are 1723traversed and the pointed-to files are retrieved. At this time, this 1724option does not cause Wget to traverse symlinks to directories and 1725recurse through them, but in the future it should be enhanced to do 1726this. 1727 1728Note that when retrieving a file (not a directory) because it was 1729specified on the command-line, rather than because it was recursed to, 1730this option has no effect. Symbolic links are always traversed in this 1731case. 1732@end table 1733 1734@node Recursive Retrieval Options, Recursive Accept/Reject Options, FTP Options, Invoking 1735@section Recursive Retrieval Options 1736 1737@table @samp 1738@item -r 1739@itemx --recursive 1740Turn on recursive retrieving. @xref{Recursive Download}, for more 1741details. 1742 1743@item -l @var{depth} 1744@itemx --level=@var{depth} 1745Specify recursion maximum depth level @var{depth} (@pxref{Recursive 1746Download}). The default maximum depth is 5. 1747 1748@cindex proxy filling 1749@cindex delete after retrieval 1750@cindex filling proxy cache 1751@item --delete-after 1752This option tells Wget to delete every single file it downloads, 1753@emph{after} having done so. It is useful for pre-fetching popular 1754pages through a proxy, e.g.: 1755 1756@example 1757wget -r -nd --delete-after http://whatever.com/~popular/page/ 1758@end example 1759 1760The @samp{-r} option is to retrieve recursively, and @samp{-nd} to not 1761create directories. 1762 1763Note that @samp{--delete-after} deletes files on the local machine. It 1764does not issue the @samp{DELE} command to remote FTP sites, for 1765instance. Also note that when @samp{--delete-after} is specified, 1766@samp{--convert-links} is ignored, so @samp{.orig} files are simply not 1767created in the first place. 1768 1769@cindex conversion of links 1770@cindex link conversion 1771@item -k 1772@itemx --convert-links 1773After the download is complete, convert the links in the document to 1774make them suitable for local viewing. This affects not only the visible 1775hyperlinks, but any part of the document that links to external content, 1776such as embedded images, links to style sheets, hyperlinks to non-@sc{html} 1777content, etc. 1778 1779Each link will be changed in one of the two ways: 1780 1781@itemize @bullet 1782@item 1783The links to files that have been downloaded by Wget will be changed to 1784refer to the file they point to as a relative link. 1785 1786Example: if the downloaded file @file{/foo/doc.html} links to 1787@file{/bar/img.gif}, also downloaded, then the link in @file{doc.html} 1788will be modified to point to @samp{../bar/img.gif}. This kind of 1789transformation works reliably for arbitrary combinations of directories. 1790 1791@item 1792The links to files that have not been downloaded by Wget will be changed 1793to include host name and absolute path of the location they point to. 1794 1795Example: if the downloaded file @file{/foo/doc.html} links to 1796@file{/bar/img.gif} (or to @file{../bar/img.gif}), then the link in 1797@file{doc.html} will be modified to point to 1798@file{http://@var{hostname}/bar/img.gif}. 1799@end itemize 1800 1801Because of this, local browsing works reliably: if a linked file was 1802downloaded, the link will refer to its local name; if it was not 1803downloaded, the link will refer to its full Internet address rather than 1804presenting a broken link. The fact that the former links are converted 1805to relative links ensures that you can move the downloaded hierarchy to 1806another directory. 1807 1808Note that only at the end of the download can Wget know which links have 1809been downloaded. Because of that, the work done by @samp{-k} will be 1810performed at the end of all the downloads. 1811 1812@cindex backing up converted files 1813@item -K 1814@itemx --backup-converted 1815When converting a file, back up the original version with a @samp{.orig} 1816suffix. Affects the behavior of @samp{-N} (@pxref{HTTP Time-Stamping 1817Internals}). 1818 1819@item -m 1820@itemx --mirror 1821Turn on options suitable for mirroring. This option turns on recursion 1822and time-stamping, sets infinite recursion depth and keeps @sc{ftp} 1823directory listings. It is currently equivalent to 1824@samp{-r -N -l inf --no-remove-listing}. 1825 1826@cindex page requisites 1827@cindex required images, downloading 1828@item -p 1829@itemx --page-requisites 1830This option causes Wget to download all the files that are necessary to 1831properly display a given @sc{html} page. This includes such things as 1832inlined images, sounds, and referenced stylesheets. 1833 1834Ordinarily, when downloading a single @sc{html} page, any requisite documents 1835that may be needed to display it properly are not downloaded. Using 1836@samp{-r} together with @samp{-l} can help, but since Wget does not 1837ordinarily distinguish between external and inlined documents, one is 1838generally left with ``leaf documents'' that are missing their 1839requisites. 1840 1841For instance, say document @file{1.html} contains an @code{<IMG>} tag 1842referencing @file{1.gif} and an @code{<A>} tag pointing to external 1843document @file{2.html}. Say that @file{2.html} is similar but that its 1844image is @file{2.gif} and it links to @file{3.html}. Say this 1845continues up to some arbitrarily high number. 1846 1847If one executes the command: 1848 1849@example 1850wget -r -l 2 http://@var{site}/1.html 1851@end example 1852 1853then @file{1.html}, @file{1.gif}, @file{2.html}, @file{2.gif}, and 1854@file{3.html} will be downloaded. As you can see, @file{3.html} is 1855without its requisite @file{3.gif} because Wget is simply counting the 1856number of hops (up to 2) away from @file{1.html} in order to determine 1857where to stop the recursion. However, with this command: 1858 1859@example 1860wget -r -l 2 -p http://@var{site}/1.html 1861@end example 1862 1863all the above files @emph{and} @file{3.html}'s requisite @file{3.gif} 1864will be downloaded. Similarly, 1865 1866@example 1867wget -r -l 1 -p http://@var{site}/1.html 1868@end example 1869 1870will cause @file{1.html}, @file{1.gif}, @file{2.html}, and @file{2.gif} 1871to be downloaded. One might think that: 1872 1873@example 1874wget -r -l 0 -p http://@var{site}/1.html 1875@end example 1876 1877would download just @file{1.html} and @file{1.gif}, but unfortunately 1878this is not the case, because @samp{-l 0} is equivalent to 1879@samp{-l inf}---that is, infinite recursion. To download a single @sc{html} 1880page (or a handful of them, all specified on the command-line or in a 1881@samp{-i} @sc{url} input file) and its (or their) requisites, simply leave off 1882@samp{-r} and @samp{-l}: 1883 1884@example 1885wget -p http://@var{site}/1.html 1886@end example 1887 1888Note that Wget will behave as if @samp{-r} had been specified, but only 1889that single page and its requisites will be downloaded. Links from that 1890page to external documents will not be followed. Actually, to download 1891a single page and all its requisites (even if they exist on separate 1892websites), and make sure the lot displays properly locally, this author 1893likes to use a few options in addition to @samp{-p}: 1894 1895@example 1896wget -E -H -k -K -p http://@var{site}/@var{document} 1897@end example 1898 1899To finish off this topic, it's worth knowing that Wget's idea of an 1900external document link is any URL specified in an @code{<A>} tag, an 1901@code{<AREA>} tag, or a @code{<LINK>} tag other than @code{<LINK 1902REL="stylesheet">}. 1903 1904@cindex @sc{html} comments 1905@cindex comments, @sc{html} 1906@item --strict-comments 1907Turn on strict parsing of @sc{html} comments. The default is to terminate 1908comments at the first occurrence of @samp{-->}. 1909 1910According to specifications, @sc{html} comments are expressed as @sc{sgml} 1911@dfn{declarations}. Declaration is special markup that begins with 1912@samp{<!} and ends with @samp{>}, such as @samp{<!DOCTYPE ...>}, that 1913may contain comments between a pair of @samp{--} delimiters. @sc{html} 1914comments are ``empty declarations'', @sc{sgml} declarations without any 1915non-comment text. Therefore, @samp{<!--foo-->} is a valid comment, and 1916so is @samp{<!--one-- --two-->}, but @samp{<!--1--2-->} is not. 1917 1918On the other hand, most @sc{html} writers don't perceive comments as anything 1919other than text delimited with @samp{<!--} and @samp{-->}, which is not 1920quite the same. For example, something like @samp{<!------------>} 1921works as a valid comment as long as the number of dashes is a multiple 1922of four (!). If not, the comment technically lasts until the next 1923@samp{--}, which may be at the other end of the document. Because of 1924this, many popular browsers completely ignore the specification and 1925implement what users have come to expect: comments delimited with 1926@samp{<!--} and @samp{-->}. 1927 1928Until version 1.9, Wget interpreted comments strictly, which resulted in 1929missing links in many web pages that displayed fine in browsers, but had 1930the misfortune of containing non-compliant comments. Beginning with 1931version 1.9, Wget has joined the ranks of clients that implements 1932``naive'' comments, terminating each comment at the first occurrence of 1933@samp{-->}. 1934 1935If, for whatever reason, you want strict comment parsing, use this 1936option to turn it on. 1937@end table 1938 1939@node Recursive Accept/Reject Options, Exit Status, Recursive Retrieval Options, Invoking 1940@section Recursive Accept/Reject Options 1941 1942@table @samp 1943@item -A @var{acclist} --accept @var{acclist} 1944@itemx -R @var{rejlist} --reject @var{rejlist} 1945Specify comma-separated lists of file name suffixes or patterns to 1946accept or reject (@pxref{Types of Files}). Note that if 1947any of the wildcard characters, @samp{*}, @samp{?}, @samp{[} or 1948@samp{]}, appear in an element of @var{acclist} or @var{rejlist}, 1949it will be treated as a pattern, rather than a suffix. 1950 1951@item -D @var{domain-list} 1952@itemx --domains=@var{domain-list} 1953Set domains to be followed. @var{domain-list} is a comma-separated list 1954of domains. Note that it does @emph{not} turn on @samp{-H}. 1955 1956@item --exclude-domains @var{domain-list} 1957Specify the domains that are @emph{not} to be followed. 1958(@pxref{Spanning Hosts}). 1959 1960@cindex follow FTP links 1961@item --follow-ftp 1962Follow @sc{ftp} links from @sc{html} documents. Without this option, 1963Wget will ignore all the @sc{ftp} links. 1964 1965@cindex tag-based recursive pruning 1966@item --follow-tags=@var{list} 1967Wget has an internal table of @sc{html} tag / attribute pairs that it 1968considers when looking for linked documents during a recursive 1969retrieval. If a user wants only a subset of those tags to be 1970considered, however, he or she should be specify such tags in a 1971comma-separated @var{list} with this option. 1972 1973@item --ignore-tags=@var{list} 1974This is the opposite of the @samp{--follow-tags} option. To skip 1975certain @sc{html} tags when recursively looking for documents to download, 1976specify them in a comma-separated @var{list}. 1977 1978In the past, this option was the best bet for downloading a single page 1979and its requisites, using a command-line like: 1980 1981@example 1982wget --ignore-tags=a,area -H -k -K -r http://@var{site}/@var{document} 1983@end example 1984 1985However, the author of this option came across a page with tags like 1986@code{<LINK REL="home" HREF="/">} and came to the realization that 1987specifying tags to ignore was not enough. One can't just tell Wget to 1988ignore @code{<LINK>}, because then stylesheets will not be downloaded. 1989Now the best bet for downloading a single page and its requisites is the 1990dedicated @samp{--page-requisites} option. 1991 1992@cindex case fold 1993@cindex ignore case 1994@item --ignore-case 1995Ignore case when matching files and directories. This influences the 1996behavior of -R, -A, -I, and -X options, as well as globbing 1997implemented when downloading from FTP sites. For example, with this 1998option, @samp{-A *.txt} will match @samp{file1.txt}, but also 1999@samp{file2.TXT}, @samp{file3.TxT}, and so on. 2000 2001@item -H 2002@itemx --span-hosts 2003Enable spanning across hosts when doing recursive retrieving 2004(@pxref{Spanning Hosts}). 2005 2006@item -L 2007@itemx --relative 2008Follow relative links only. Useful for retrieving a specific home page 2009without any distractions, not even those from the same hosts 2010(@pxref{Relative Links}). 2011 2012@item -I @var{list} 2013@itemx --include-directories=@var{list} 2014Specify a comma-separated list of directories you wish to follow when 2015downloading (@pxref{Directory-Based Limits}). Elements 2016of @var{list} may contain wildcards. 2017 2018@item -X @var{list} 2019@itemx --exclude-directories=@var{list} 2020Specify a comma-separated list of directories you wish to exclude from 2021download (@pxref{Directory-Based Limits}). Elements of 2022@var{list} may contain wildcards. 2023 2024@item -np 2025@item --no-parent 2026Do not ever ascend to the parent directory when retrieving recursively. 2027This is a useful option, since it guarantees that only the files 2028@emph{below} a certain hierarchy will be downloaded. 2029@xref{Directory-Based Limits}, for more details. 2030@end table 2031 2032@c man end 2033 2034@node Exit Status, , Recursive Accept/Reject Options, Invoking 2035@section Exit Status 2036 2037@c man begin EXITSTATUS 2038 2039Wget may return one of several error codes if it encounters problems. 2040 2041 2042@table @asis 2043@item 0 2044No problems occurred. 2045 2046@item 1 2047Generic error code. 2048 2049@item 2 2050Parse error---for instance, when parsing command-line options, the 2051@samp{.wgetrc} or @samp{.netrc}... 2052 2053@item 3 2054File I/O error. 2055 2056@item 4 2057Network failure. 2058 2059@item 5 2060SSL verification failure. 2061 2062@item 6 2063Username/password authentication failure. 2064 2065@item 7 2066Protocol errors. 2067 2068@item 8 2069Server issued an error response. 2070@end table 2071 2072 2073With the exceptions of 0 and 1, the lower-numbered exit codes take 2074precedence over higher-numbered ones, when multiple types of errors 2075are encountered. 2076 2077In versions of Wget prior to 1.12, Wget's exit status tended to be 2078unhelpful and inconsistent. Recursive downloads would virtually always 2079return 0 (success), regardless of any issues encountered, and 2080non-recursive fetches only returned the status corresponding to the 2081most recently-attempted download. 2082 2083@c man end 2084 2085@node Recursive Download, Following Links, Invoking, Top 2086@chapter Recursive Download 2087@cindex recursion 2088@cindex retrieving 2089@cindex recursive download 2090 2091GNU Wget is capable of traversing parts of the Web (or a single 2092@sc{http} or @sc{ftp} server), following links and directory structure. 2093We refer to this as to @dfn{recursive retrieval}, or @dfn{recursion}. 2094 2095With @sc{http} @sc{url}s, Wget retrieves and parses the @sc{html} or 2096@sc{css} from the given @sc{url}, retrieving the files the document 2097refers to, through markup like @code{href} or @code{src}, or @sc{css} 2098@sc{uri} values specified using the @samp{url()} functional notation. 2099If the freshly downloaded file is also of type @code{text/html}, 2100@code{application/xhtml+xml}, or @code{text/css}, it will be parsed 2101and followed further. 2102 2103Recursive retrieval of @sc{http} and @sc{html}/@sc{css} content is 2104@dfn{breadth-first}. This means that Wget first downloads the requested 2105document, then the documents linked from that document, then the 2106documents linked by them, and so on. In other words, Wget first 2107downloads the documents at depth 1, then those at depth 2, and so on 2108until the specified maximum depth. 2109 2110The maximum @dfn{depth} to which the retrieval may descend is specified 2111with the @samp{-l} option. The default maximum depth is five layers. 2112 2113When retrieving an @sc{ftp} @sc{url} recursively, Wget will retrieve all 2114the data from the given directory tree (including the subdirectories up 2115to the specified depth) on the remote server, creating its mirror image 2116locally. @sc{ftp} retrieval is also limited by the @code{depth} 2117parameter. Unlike @sc{http} recursion, @sc{ftp} recursion is performed 2118depth-first. 2119 2120By default, Wget will create a local directory tree, corresponding to 2121the one found on the remote server. 2122 2123Recursive retrieving can find a number of applications, the most 2124important of which is mirroring. It is also useful for @sc{www} 2125presentations, and any other opportunities where slow network 2126connections should be bypassed by storing the files locally. 2127 2128You should be warned that recursive downloads can overload the remote 2129servers. Because of that, many administrators frown upon them and may 2130ban access from your site if they detect very fast downloads of big 2131amounts of content. When downloading from Internet servers, consider 2132using the @samp{-w} option to introduce a delay between accesses to the 2133server. The download will take a while longer, but the server 2134administrator will not be alarmed by your rudeness. 2135 2136Of course, recursive download may cause problems on your machine. If 2137left to run unchecked, it can easily fill up the disk. If downloading 2138from local network, it can also take bandwidth on the system, as well as 2139consume memory and CPU. 2140 2141Try to specify the criteria that match the kind of download you are 2142trying to achieve. If you want to download only one page, use 2143@samp{--page-requisites} without any additional recursion. If you want 2144to download things under one directory, use @samp{-np} to avoid 2145downloading things from other directories. If you want to download all 2146the files from one directory, use @samp{-l 1} to make sure the recursion 2147depth never exceeds one. @xref{Following Links}, for more information 2148about this. 2149 2150Recursive retrieval should be used with care. Don't say you were not 2151warned. 2152 2153@node Following Links, Time-Stamping, Recursive Download, Top 2154@chapter Following Links 2155@cindex links 2156@cindex following links 2157 2158When retrieving recursively, one does not wish to retrieve loads of 2159unnecessary data. Most of the time the users bear in mind exactly what 2160they want to download, and want Wget to follow only specific links. 2161 2162For example, if you wish to download the music archive from 2163@samp{fly.srk.fer.hr}, you will not want to download all the home pages 2164that happen to be referenced by an obscure part of the archive. 2165 2166Wget possesses several mechanisms that allows you to fine-tune which 2167links it will follow. 2168 2169@menu 2170* Spanning Hosts:: (Un)limiting retrieval based on host name. 2171* Types of Files:: Getting only certain files. 2172* Directory-Based Limits:: Getting only certain directories. 2173* Relative Links:: Follow relative links only. 2174* FTP Links:: Following FTP links. 2175@end menu 2176 2177@node Spanning Hosts, Types of Files, Following Links, Following Links 2178@section Spanning Hosts 2179@cindex spanning hosts 2180@cindex hosts, spanning 2181 2182Wget's recursive retrieval normally refuses to visit hosts different 2183than the one you specified on the command line. This is a reasonable 2184default; without it, every retrieval would have the potential to turn 2185your Wget into a small version of google. 2186 2187However, visiting different hosts, or @dfn{host spanning,} is sometimes 2188a useful option. Maybe the images are served from a different server. 2189Maybe you're mirroring a site that consists of pages interlinked between 2190three servers. Maybe the server has two equivalent names, and the @sc{html} 2191pages refer to both interchangeably. 2192 2193@table @asis 2194@item Span to any host---@samp{-H} 2195 2196The @samp{-H} option turns on host spanning, thus allowing Wget's 2197recursive run to visit any host referenced by a link. Unless sufficient 2198recursion-limiting criteria are applied depth, these foreign hosts will 2199typically link to yet more hosts, and so on until Wget ends up sucking 2200up much more data than you have intended. 2201 2202@item Limit spanning to certain domains---@samp{-D} 2203 2204The @samp{-D} option allows you to specify the domains that will be 2205followed, thus limiting the recursion only to the hosts that belong to 2206these domains. Obviously, this makes sense only in conjunction with 2207@samp{-H}. A typical example would be downloading the contents of 2208@samp{www.server.com}, but allowing downloads from 2209@samp{images.server.com}, etc.: 2210 2211@example 2212wget -rH -Dserver.com http://www.server.com/ 2213@end example 2214 2215You can specify more than one address by separating them with a comma, 2216e.g. @samp{-Ddomain1.com,domain2.com}. 2217 2218@item Keep download off certain domains---@samp{--exclude-domains} 2219 2220If there are domains you want to exclude specifically, you can do it 2221with @samp{--exclude-domains}, which accepts the same type of arguments 2222of @samp{-D}, but will @emph{exclude} all the listed domains. For 2223example, if you want to download all the hosts from @samp{foo.edu} 2224domain, with the exception of @samp{sunsite.foo.edu}, you can do it like 2225this: 2226 2227@example 2228wget -rH -Dfoo.edu --exclude-domains sunsite.foo.edu \ 2229 http://www.foo.edu/ 2230@end example 2231 2232@end table 2233 2234@node Types of Files, Directory-Based Limits, Spanning Hosts, Following Links 2235@section Types of Files 2236@cindex types of files 2237 2238When downloading material from the web, you will often want to restrict 2239the retrieval to only certain file types. For example, if you are 2240interested in downloading @sc{gif}s, you will not be overjoyed to get 2241loads of PostScript documents, and vice versa. 2242 2243Wget offers two options to deal with this problem. Each option 2244description lists a short name, a long name, and the equivalent command 2245in @file{.wgetrc}. 2246 2247@cindex accept wildcards 2248@cindex accept suffixes 2249@cindex wildcards, accept 2250@cindex suffixes, accept 2251@table @samp 2252@item -A @var{acclist} 2253@itemx --accept @var{acclist} 2254@itemx accept = @var{acclist} 2255The argument to @samp{--accept} option is a list of file suffixes or 2256patterns that Wget will download during recursive retrieval. A suffix 2257is the ending part of a file, and consists of ``normal'' letters, 2258e.g. @samp{gif} or @samp{.jpg}. A matching pattern contains shell-like 2259wildcards, e.g. @samp{books*} or @samp{zelazny*196[0-9]*}. 2260 2261So, specifying @samp{wget -A gif,jpg} will make Wget download only the 2262files ending with @samp{gif} or @samp{jpg}, i.e. @sc{gif}s and 2263@sc{jpeg}s. On the other hand, @samp{wget -A "zelazny*196[0-9]*"} will 2264download only files beginning with @samp{zelazny} and containing numbers 2265from 1960 to 1969 anywhere within. Look up the manual of your shell for 2266a description of how pattern matching works. 2267 2268Of course, any number of suffixes and patterns can be combined into a 2269comma-separated list, and given as an argument to @samp{-A}. 2270 2271@cindex reject wildcards 2272@cindex reject suffixes 2273@cindex wildcards, reject 2274@cindex suffixes, reject 2275@item -R @var{rejlist} 2276@itemx --reject @var{rejlist} 2277@itemx reject = @var{rejlist} 2278The @samp{--reject} option works the same way as @samp{--accept}, only 2279its logic is the reverse; Wget will download all files @emph{except} the 2280ones matching the suffixes (or patterns) in the list. 2281 2282So, if you want to download a whole page except for the cumbersome 2283@sc{mpeg}s and @sc{.au} files, you can use @samp{wget -R mpg,mpeg,au}. 2284Analogously, to download all files except the ones beginning with 2285@samp{bjork}, use @samp{wget -R "bjork*"}. The quotes are to prevent 2286expansion by the shell. 2287@end table 2288 2289@noindent 2290The @samp{-A} and @samp{-R} options may be combined to achieve even 2291better fine-tuning of which files to retrieve. E.g. @samp{wget -A 2292"*zelazny*" -R .ps} will download all the files having @samp{zelazny} as 2293a part of their name, but @emph{not} the PostScript files. 2294 2295Note that these two options do not affect the downloading of @sc{html} 2296files (as determined by a @samp{.htm} or @samp{.html} filename 2297prefix). This behavior may not be desirable for all users, and may be 2298changed for future versions of Wget. 2299 2300Note, too, that query strings (strings at the end of a URL beginning 2301with a question mark (@samp{?}) are not included as part of the 2302filename for accept/reject rules, even though these will actually 2303contribute to the name chosen for the local file. It is expected that 2304a future version of Wget will provide an option to allow matching 2305against query strings. 2306 2307Finally, it's worth noting that the accept/reject lists are matched 2308@emph{twice} against downloaded files: once against the URL's filename 2309portion, to determine if the file should be downloaded in the first 2310place; then, after it has been accepted and successfully downloaded, 2311the local file's name is also checked against the accept/reject lists 2312to see if it should be removed. The rationale was that, since 2313@samp{.htm} and @samp{.html} files are always downloaded regardless of 2314accept/reject rules, they should be removed @emph{after} being 2315downloaded and scanned for links, if they did match the accept/reject 2316lists. However, this can lead to unexpected results, since the local 2317filenames can differ from the original URL filenames in the following 2318ways, all of which can change whether an accept/reject rule matches: 2319 2320@itemize @bullet 2321@item 2322If the local file already exists and @samp{--no-directories} was 2323specified, a numeric suffix will be appended to the original name. 2324@item 2325If @samp{--adjust-extension} was specified, the local filename might have 2326@samp{.html} appended to it. If Wget is invoked with @samp{-E -A.php}, 2327a filename such as @samp{index.php} will match be accepted, but upon 2328download will be named @samp{index.php.html}, which no longer matches, 2329and so the file will be deleted. 2330@item 2331Query strings do not contribute to URL matching, but are included in 2332local filenames, and so @emph{do} contribute to filename matching. 2333@end itemize 2334 2335@noindent 2336This behavior, too, is considered less-than-desirable, and may change 2337in a future version of Wget. 2338 2339@node Directory-Based Limits, Relative Links, Types of Files, Following Links 2340@section Directory-Based Limits 2341@cindex directories 2342@cindex directory limits 2343 2344Regardless of other link-following facilities, it is often useful to 2345place the restriction of what files to retrieve based on the directories 2346those files are placed in. There can be many reasons for this---the 2347home pages may be organized in a reasonable directory structure; or some 2348directories may contain useless information, e.g. @file{/cgi-bin} or 2349@file{/dev} directories. 2350 2351Wget offers three different options to deal with this requirement. Each 2352option description lists a short name, a long name, and the equivalent 2353command in @file{.wgetrc}. 2354 2355@cindex directories, include 2356@cindex include directories 2357@cindex accept directories 2358@table @samp 2359@item -I @var{list} 2360@itemx --include @var{list} 2361@itemx include_directories = @var{list} 2362@samp{-I} option accepts a comma-separated list of directories included 2363in the retrieval. Any other directories will simply be ignored. The 2364directories are absolute paths. 2365 2366So, if you wish to download from @samp{http://host/people/bozo/} 2367following only links to bozo's colleagues in the @file{/people} 2368directory and the bogus scripts in @file{/cgi-bin}, you can specify: 2369 2370@example 2371wget -I /people,/cgi-bin http://host/people/bozo/ 2372@end example 2373 2374@cindex directories, exclude 2375@cindex exclude directories 2376@cindex reject directories 2377@item -X @var{list} 2378@itemx --exclude @var{list} 2379@itemx exclude_directories = @var{list} 2380@samp{-X} option is exactly the reverse of @samp{-I}---this is a list of 2381directories @emph{excluded} from the download. E.g. if you do not want 2382Wget to download things from @file{/cgi-bin} directory, specify @samp{-X 2383/cgi-bin} on the command line. 2384 2385The same as with @samp{-A}/@samp{-R}, these two options can be combined 2386to get a better fine-tuning of downloading subdirectories. E.g. if you 2387want to load all the files from @file{/pub} hierarchy except for 2388@file{/pub/worthless}, specify @samp{-I/pub -X/pub/worthless}. 2389 2390@cindex no parent 2391@item -np 2392@itemx --no-parent 2393@itemx no_parent = on 2394The simplest, and often very useful way of limiting directories is 2395disallowing retrieval of the links that refer to the hierarchy 2396@dfn{above} than the beginning directory, i.e. disallowing ascent to the 2397parent directory/directories. 2398 2399The @samp{--no-parent} option (short @samp{-np}) is useful in this case. 2400Using it guarantees that you will never leave the existing hierarchy. 2401Supposing you issue Wget with: 2402 2403@example 2404wget -r --no-parent http://somehost/~luzer/my-archive/ 2405@end example 2406 2407You may rest assured that none of the references to 2408@file{/~his-girls-homepage/} or @file{/~luzer/all-my-mpegs/} will be 2409followed. Only the archive you are interested in will be downloaded. 2410Essentially, @samp{--no-parent} is similar to 2411@samp{-I/~luzer/my-archive}, only it handles redirections in a more 2412intelligent fashion. 2413 2414@strong{Note} that, for HTTP (and HTTPS), the trailing slash is very 2415important to @samp{--no-parent}. HTTP has no concept of a ``directory''---Wget 2416relies on you to indicate what's a directory and what isn't. In 2417@samp{http://foo/bar/}, Wget will consider @samp{bar} to be a 2418directory, while in @samp{http://foo/bar} (no trailing slash), 2419@samp{bar} will be considered a filename (so @samp{--no-parent} would be 2420meaningless, as its parent is @samp{/}). 2421@end table 2422 2423@node Relative Links, FTP Links, Directory-Based Limits, Following Links 2424@section Relative Links 2425@cindex relative links 2426 2427When @samp{-L} is turned on, only the relative links are ever followed. 2428Relative links are here defined those that do not refer to the web 2429server root. For example, these links are relative: 2430 2431@example 2432<a href="foo.gif"> 2433<a href="foo/bar.gif"> 2434<a href="../foo/bar.gif"> 2435@end example 2436 2437These links are not relative: 2438 2439@example 2440<a href="/foo.gif"> 2441<a href="/foo/bar.gif"> 2442<a href="http://www.server.com/foo/bar.gif"> 2443@end example 2444 2445Using this option guarantees that recursive retrieval will not span 2446hosts, even without @samp{-H}. In simple cases it also allows downloads 2447to ``just work'' without having to convert links. 2448 2449This option is probably not very useful and might be removed in a future 2450release. 2451 2452@node FTP Links, , Relative Links, Following Links 2453@section Following FTP Links 2454@cindex following ftp links 2455 2456The rules for @sc{ftp} are somewhat specific, as it is necessary for 2457them to be. @sc{ftp} links in @sc{html} documents are often included 2458for purposes of reference, and it is often inconvenient to download them 2459by default. 2460 2461To have @sc{ftp} links followed from @sc{html} documents, you need to 2462specify the @samp{--follow-ftp} option. Having done that, @sc{ftp} 2463links will span hosts regardless of @samp{-H} setting. This is logical, 2464as @sc{ftp} links rarely point to the same host where the @sc{http} 2465server resides. For similar reasons, the @samp{-L} options has no 2466effect on such downloads. On the other hand, domain acceptance 2467(@samp{-D}) and suffix rules (@samp{-A} and @samp{-R}) apply normally. 2468 2469Also note that followed links to @sc{ftp} directories will not be 2470retrieved recursively further. 2471 2472@node Time-Stamping, Startup File, Following Links, Top 2473@chapter Time-Stamping 2474@cindex time-stamping 2475@cindex timestamping 2476@cindex updating the archives 2477@cindex incremental updating 2478 2479One of the most important aspects of mirroring information from the 2480Internet is updating your archives. 2481 2482Downloading the whole archive again and again, just to replace a few 2483changed files is expensive, both in terms of wasted bandwidth and money, 2484and the time to do the update. This is why all the mirroring tools 2485offer the option of incremental updating. 2486 2487Such an updating mechanism means that the remote server is scanned in 2488search of @dfn{new} files. Only those new files will be downloaded in 2489the place of the old ones. 2490 2491A file is considered new if one of these two conditions are met: 2492 2493@enumerate 2494@item 2495A file of that name does not already exist locally. 2496 2497@item 2498A file of that name does exist, but the remote file was modified more 2499recently than the local file. 2500@end enumerate 2501 2502To implement this, the program needs to be aware of the time of last 2503modification of both local and remote files. We call this information the 2504@dfn{time-stamp} of a file. 2505 2506The time-stamping in GNU Wget is turned on using @samp{--timestamping} 2507(@samp{-N}) option, or through @code{timestamping = on} directive in 2508@file{.wgetrc}. With this option, for each file it intends to download, 2509Wget will check whether a local file of the same name exists. If it 2510does, and the remote file is not newer, Wget will not download it. 2511 2512If the local file does not exist, or the sizes of the files do not 2513match, Wget will download the remote file no matter what the time-stamps 2514say. 2515 2516@menu 2517* Time-Stamping Usage:: 2518* HTTP Time-Stamping Internals:: 2519* FTP Time-Stamping Internals:: 2520@end menu 2521 2522@node Time-Stamping Usage, HTTP Time-Stamping Internals, Time-Stamping, Time-Stamping 2523@section Time-Stamping Usage 2524@cindex time-stamping usage 2525@cindex usage, time-stamping 2526 2527The usage of time-stamping is simple. Say you would like to download a 2528file so that it keeps its date of modification. 2529 2530@example 2531wget -S http://www.gnu.ai.mit.edu/ 2532@end example 2533 2534A simple @code{ls -l} shows that the time stamp on the local file equals 2535the state of the @code{Last-Modified} header, as returned by the server. 2536As you can see, the time-stamping info is preserved locally, even 2537without @samp{-N} (at least for @sc{http}). 2538 2539Several days later, you would like Wget to check if the remote file has 2540changed, and download it if it has. 2541 2542@example 2543wget -N http://www.gnu.ai.mit.edu/ 2544@end example 2545 2546Wget will ask the server for the last-modified date. If the local file 2547has the same timestamp as the server, or a newer one, the remote file 2548will not be re-fetched. However, if the remote file is more recent, 2549Wget will proceed to fetch it. 2550 2551The same goes for @sc{ftp}. For example: 2552 2553@example 2554wget "ftp://ftp.ifi.uio.no/pub/emacs/gnus/*" 2555@end example 2556 2557(The quotes around that URL are to prevent the shell from trying to 2558interpret the @samp{*}.) 2559 2560After download, a local directory listing will show that the timestamps 2561match those on the remote server. Reissuing the command with @samp{-N} 2562will make Wget re-fetch @emph{only} the files that have been modified 2563since the last download. 2564 2565If you wished to mirror the GNU archive every week, you would use a 2566command like the following, weekly: 2567 2568@example 2569wget --timestamping -r ftp://ftp.gnu.org/pub/gnu/ 2570@end example 2571 2572Note that time-stamping will only work for files for which the server 2573gives a timestamp. For @sc{http}, this depends on getting a 2574@code{Last-Modified} header. For @sc{ftp}, this depends on getting a 2575directory listing with dates in a format that Wget can parse 2576(@pxref{FTP Time-Stamping Internals}). 2577 2578@node HTTP Time-Stamping Internals, FTP Time-Stamping Internals, Time-Stamping Usage, Time-Stamping 2579@section HTTP Time-Stamping Internals 2580@cindex http time-stamping 2581 2582Time-stamping in @sc{http} is implemented by checking of the 2583@code{Last-Modified} header. If you wish to retrieve the file 2584@file{foo.html} through @sc{http}, Wget will check whether 2585@file{foo.html} exists locally. If it doesn't, @file{foo.html} will be 2586retrieved unconditionally. 2587 2588If the file does exist locally, Wget will first check its local 2589time-stamp (similar to the way @code{ls -l} checks it), and then send a 2590@code{HEAD} request to the remote server, demanding the information on 2591the remote file. 2592 2593The @code{Last-Modified} header is examined to find which file was 2594modified more recently (which makes it ``newer''). If the remote file 2595is newer, it will be downloaded; if it is older, Wget will give 2596up.@footnote{As an additional check, Wget will look at the 2597@code{Content-Length} header, and compare the sizes; if they are not the 2598same, the remote file will be downloaded no matter what the time-stamp 2599says.} 2600 2601When @samp{--backup-converted} (@samp{-K}) is specified in conjunction 2602with @samp{-N}, server file @samp{@var{X}} is compared to local file 2603@samp{@var{X}.orig}, if extant, rather than being compared to local file 2604@samp{@var{X}}, which will always differ if it's been converted by 2605@samp{--convert-links} (@samp{-k}). 2606 2607Arguably, @sc{http} time-stamping should be implemented using the 2608@code{If-Modified-Since} request. 2609 2610@node FTP Time-Stamping Internals, , HTTP Time-Stamping Internals, Time-Stamping 2611@section FTP Time-Stamping Internals 2612@cindex ftp time-stamping 2613 2614In theory, @sc{ftp} time-stamping works much the same as @sc{http}, only 2615@sc{ftp} has no headers---time-stamps must be ferreted out of directory 2616listings. 2617 2618If an @sc{ftp} download is recursive or uses globbing, Wget will use the 2619@sc{ftp} @code{LIST} command to get a file listing for the directory 2620containing the desired file(s). It will try to analyze the listing, 2621treating it like Unix @code{ls -l} output, extracting the time-stamps. 2622The rest is exactly the same as for @sc{http}. Note that when 2623retrieving individual files from an @sc{ftp} server without using 2624globbing or recursion, listing files will not be downloaded (and thus 2625files will not be time-stamped) unless @samp{-N} is specified. 2626 2627Assumption that every directory listing is a Unix-style listing may 2628sound extremely constraining, but in practice it is not, as many 2629non-Unix @sc{ftp} servers use the Unixoid listing format because most 2630(all?) of the clients understand it. Bear in mind that @sc{rfc959} 2631defines no standard way to get a file list, let alone the time-stamps. 2632We can only hope that a future standard will define this. 2633 2634Another non-standard solution includes the use of @code{MDTM} command 2635that is supported by some @sc{ftp} servers (including the popular 2636@code{wu-ftpd}), which returns the exact time of the specified file. 2637Wget may support this command in the future. 2638 2639@node Startup File, Examples, Time-Stamping, Top 2640@chapter Startup File 2641@cindex startup file 2642@cindex wgetrc 2643@cindex .wgetrc 2644@cindex startup 2645@cindex .netrc 2646 2647Once you know how to change default settings of Wget through command 2648line arguments, you may wish to make some of those settings permanent. 2649You can do that in a convenient way by creating the Wget startup 2650file---@file{.wgetrc}. 2651 2652Besides @file{.wgetrc} is the ``main'' initialization file, it is 2653convenient to have a special facility for storing passwords. Thus Wget 2654reads and interprets the contents of @file{$HOME/.netrc}, if it finds 2655it. You can find @file{.netrc} format in your system manuals. 2656 2657Wget reads @file{.wgetrc} upon startup, recognizing a limited set of 2658commands. 2659 2660@menu 2661* Wgetrc Location:: Location of various wgetrc files. 2662* Wgetrc Syntax:: Syntax of wgetrc. 2663* Wgetrc Commands:: List of available commands. 2664* Sample Wgetrc:: A wgetrc example. 2665@end menu 2666 2667@node Wgetrc Location, Wgetrc Syntax, Startup File, Startup File 2668@section Wgetrc Location 2669@cindex wgetrc location 2670@cindex location of wgetrc 2671 2672When initializing, Wget will look for a @dfn{global} startup file, 2673@file{/usr/local/etc/wgetrc} by default (or some prefix other than 2674@file{/usr/local}, if Wget was not installed there) and read commands 2675from there, if it exists. 2676 2677Then it will look for the user's file. If the environmental variable 2678@code{WGETRC} is set, Wget will try to load that file. Failing that, no 2679further attempts will be made. 2680 2681If @code{WGETRC} is not set, Wget will try to load @file{$HOME/.wgetrc}. 2682 2683The fact that user's settings are loaded after the system-wide ones 2684means that in case of collision user's wgetrc @emph{overrides} the 2685system-wide wgetrc (in @file{/usr/local/etc/wgetrc} by default). 2686Fascist admins, away! 2687 2688@node Wgetrc Syntax, Wgetrc Commands, Wgetrc Location, Startup File 2689@section Wgetrc Syntax 2690@cindex wgetrc syntax 2691@cindex syntax of wgetrc 2692 2693The syntax of a wgetrc command is simple: 2694 2695@example 2696variable = value 2697@end example 2698 2699The @dfn{variable} will also be called @dfn{command}. Valid 2700@dfn{values} are different for different commands. 2701 2702The commands are case-insensitive and underscore-insensitive. Thus 2703@samp{DIr__PrefiX} is the same as @samp{dirprefix}. Empty lines, lines 2704beginning with @samp{#} and lines containing white-space only are 2705discarded. 2706 2707Commands that expect a comma-separated list will clear the list on an 2708empty command. So, if you wish to reset the rejection list specified in 2709global @file{wgetrc}, you can do it with: 2710 2711@example 2712reject = 2713@end example 2714 2715@node Wgetrc Commands, Sample Wgetrc, Wgetrc Syntax, Startup File 2716@section Wgetrc Commands 2717@cindex wgetrc commands 2718 2719The complete set of commands is listed below. Legal values are listed 2720after the @samp{=}. Simple Boolean values can be set or unset using 2721@samp{on} and @samp{off} or @samp{1} and @samp{0}. 2722 2723Some commands take pseudo-arbitrary values. @var{address} values can be 2724hostnames or dotted-quad IP addresses. @var{n} can be any positive 2725integer, or @samp{inf} for infinity, where appropriate. @var{string} 2726values can be any non-empty string. 2727 2728Most of these commands have direct command-line equivalents. Also, any 2729wgetrc command can be specified on the command line using the 2730@samp{--execute} switch (@pxref{Basic Startup Options}.) 2731 2732@table @asis 2733@item accept/reject = @var{string} 2734Same as @samp{-A}/@samp{-R} (@pxref{Types of Files}). 2735 2736@item add_hostdir = on/off 2737Enable/disable host-prefixed file names. @samp{-nH} disables it. 2738 2739@item ask_password = on/off 2740Prompt for a password for each connection established. Cannot be specified 2741when @samp{--password} is being used, because they are mutually 2742exclusive. Equivalent to @samp{--ask-password}. 2743 2744@item auth_no_challenge = on/off 2745If this option is given, Wget will send Basic HTTP authentication 2746information (plaintext username and password) for all requests. See 2747@samp{--auth-no-challenge}. 2748 2749@item background = on/off 2750Enable/disable going to background---the same as @samp{-b} (which 2751enables it). 2752 2753@item backup_converted = on/off 2754Enable/disable saving pre-converted files with the suffix 2755@samp{.orig}---the same as @samp{-K} (which enables it). 2756 2757@c @item backups = @var{number} 2758@c #### Document me! 2759@c 2760@item base = @var{string} 2761Consider relative @sc{url}s in input files (specified via the 2762@samp{input} command or the @samp{--input-file}/@samp{-i} option, 2763together with @samp{force_html} or @samp{--force-html}) 2764as being relative to @var{string}---the same as @samp{--base=@var{string}}. 2765 2766@item bind_address = @var{address} 2767Bind to @var{address}, like the @samp{--bind-address=@var{address}}. 2768 2769@item ca_certificate = @var{file} 2770Set the certificate authority bundle file to @var{file}. The same 2771as @samp{--ca-certificate=@var{file}}. 2772 2773@item ca_directory = @var{directory} 2774Set the directory used for certificate authorities. The same as 2775@samp{--ca-directory=@var{directory}}. 2776 2777@item cache = on/off 2778When set to off, disallow server-caching. See the @samp{--no-cache} 2779option. 2780 2781@item certificate = @var{file} 2782Set the client certificate file name to @var{file}. The same as 2783@samp{--certificate=@var{file}}. 2784 2785@item certificate_type = @var{string} 2786Specify the type of the client certificate, legal values being 2787@samp{PEM} (the default) and @samp{DER} (aka ASN1). The same as 2788@samp{--certificate-type=@var{string}}. 2789 2790@item check_certificate = on/off 2791If this is set to off, the server certificate is not checked against 2792the specified client authorities. The default is ``on''. The same as 2793@samp{--check-certificate}. 2794 2795@item connect_timeout = @var{n} 2796Set the connect timeout---the same as @samp{--connect-timeout}. 2797 2798@item content_disposition = on/off 2799Turn on recognition of the (non-standard) @samp{Content-Disposition} 2800HTTP header---if set to @samp{on}, the same as @samp{--content-disposition}. 2801 2802@item continue = on/off 2803If set to on, force continuation of preexistent partially retrieved 2804files. See @samp{-c} before setting it. 2805 2806@item convert_links = on/off 2807Convert non-relative links locally. The same as @samp{-k}. 2808 2809@item cookies = on/off 2810When set to off, disallow cookies. See the @samp{--cookies} option. 2811 2812@item cut_dirs = @var{n} 2813Ignore @var{n} remote directory components. Equivalent to 2814@samp{--cut-dirs=@var{n}}. 2815 2816@item debug = on/off 2817Debug mode, same as @samp{-d}. 2818 2819@item default_page = @var{string} 2820Default page name---the same as @samp{--default-page=@var{string}}. 2821 2822@item delete_after = on/off 2823Delete after download---the same as @samp{--delete-after}. 2824 2825@item dir_prefix = @var{string} 2826Top of directory tree---the same as @samp{-P @var{string}}. 2827 2828@item dirstruct = on/off 2829Turning dirstruct on or off---the same as @samp{-x} or @samp{-nd}, 2830respectively. 2831 2832@item dns_cache = on/off 2833Turn DNS caching on/off. Since DNS caching is on by default, this 2834option is normally used to turn it off and is equivalent to 2835@samp{--no-dns-cache}. 2836 2837@item dns_timeout = @var{n} 2838Set the DNS timeout---the same as @samp{--dns-timeout}. 2839 2840@item domains = @var{string} 2841Same as @samp{-D} (@pxref{Spanning Hosts}). 2842 2843@item dot_bytes = @var{n} 2844Specify the number of bytes ``contained'' in a dot, as seen throughout 2845the retrieval (1024 by default). You can postfix the value with 2846@samp{k} or @samp{m}, representing kilobytes and megabytes, 2847respectively. With dot settings you can tailor the dot retrieval to 2848suit your needs, or you can use the predefined @dfn{styles} 2849(@pxref{Download Options}). 2850 2851@item dot_spacing = @var{n} 2852Specify the number of dots in a single cluster (10 by default). 2853 2854@item dots_in_line = @var{n} 2855Specify the number of dots that will be printed in each line throughout 2856the retrieval (50 by default). 2857 2858@item egd_file = @var{file} 2859Use @var{string} as the EGD socket file name. The same as 2860@samp{--egd-file=@var{file}}. 2861 2862@item exclude_directories = @var{string} 2863Specify a comma-separated list of directories you wish to exclude from 2864download---the same as @samp{-X @var{string}} (@pxref{Directory-Based 2865Limits}). 2866 2867@item exclude_domains = @var{string} 2868Same as @samp{--exclude-domains=@var{string}} (@pxref{Spanning 2869Hosts}). 2870 2871@item follow_ftp = on/off 2872Follow @sc{ftp} links from @sc{html} documents---the same as 2873@samp{--follow-ftp}. 2874 2875@item follow_tags = @var{string} 2876Only follow certain @sc{html} tags when doing a recursive retrieval, 2877just like @samp{--follow-tags=@var{string}}. 2878 2879@item force_html = on/off 2880If set to on, force the input filename to be regarded as an @sc{html} 2881document---the same as @samp{-F}. 2882 2883@item ftp_password = @var{string} 2884Set your @sc{ftp} password to @var{string}. Without this setting, the 2885password defaults to @samp{-wget@@}, which is a useful default for 2886anonymous @sc{ftp} access. 2887 2888This command used to be named @code{passwd} prior to Wget 1.10. 2889 2890@item ftp_proxy = @var{string} 2891Use @var{string} as @sc{ftp} proxy, instead of the one specified in 2892environment. 2893 2894@item ftp_user = @var{string} 2895Set @sc{ftp} user to @var{string}. 2896 2897This command used to be named @code{login} prior to Wget 1.10. 2898 2899@item glob = on/off 2900Turn globbing on/off---the same as @samp{--glob} and @samp{--no-glob}. 2901 2902@item header = @var{string} 2903Define a header for HTTP downloads, like using 2904@samp{--header=@var{string}}. 2905 2906@item adjust_extension = on/off 2907Add a @samp{.html} extension to @samp{text/html} or 2908@samp{application/xhtml+xml} files that lack one, or a @samp{.css} 2909extension to @samp{text/css} files that lack one, like 2910@samp{-E}. Previously named @samp{html_extension} (still acceptable, 2911but deprecated). 2912 2913@item http_keep_alive = on/off 2914Turn the keep-alive feature on or off (defaults to on). Turning it 2915off is equivalent to @samp{--no-http-keep-alive}. 2916 2917@item http_password = @var{string} 2918Set @sc{http} password, equivalent to 2919@samp{--http-password=@var{string}}. 2920 2921@item http_proxy = @var{string} 2922Use @var{string} as @sc{http} proxy, instead of the one specified in 2923environment. 2924 2925@item http_user = @var{string} 2926Set @sc{http} user to @var{string}, equivalent to 2927@samp{--http-user=@var{string}}. 2928 2929@item https_proxy = @var{string} 2930Use @var{string} as @sc{https} proxy, instead of the one specified in 2931environment. 2932 2933@item ignore_case = on/off 2934When set to on, match files and directories case insensitively; the 2935same as @samp{--ignore-case}. 2936 2937@item ignore_length = on/off 2938When set to on, ignore @code{Content-Length} header; the same as 2939@samp{--ignore-length}. 2940 2941@item ignore_tags = @var{string} 2942Ignore certain @sc{html} tags when doing a recursive retrieval, like 2943@samp{--ignore-tags=@var{string}}. 2944 2945@item include_directories = @var{string} 2946Specify a comma-separated list of directories you wish to follow when 2947downloading---the same as @samp{-I @var{string}}. 2948 2949@item iri = on/off 2950When set to on, enable internationalized URI (IRI) support; the same as 2951@samp{--iri}. 2952 2953@item inet4_only = on/off 2954Force connecting to IPv4 addresses, off by default. You can put this 2955in the global init file to disable Wget's attempts to resolve and 2956connect to IPv6 hosts. Available only if Wget was compiled with IPv6 2957support. The same as @samp{--inet4-only} or @samp{-4}. 2958 2959@item inet6_only = on/off 2960Force connecting to IPv6 addresses, off by default. Available only if 2961Wget was compiled with IPv6 support. The same as @samp{--inet6-only} 2962or @samp{-6}. 2963 2964@item input = @var{file} 2965Read the @sc{url}s from @var{string}, like @samp{-i @var{file}}. 2966 2967@item keep_session_cookies = on/off 2968When specified, causes @samp{save_cookies = on} to also save session 2969cookies. See @samp{--keep-session-cookies}. 2970 2971@item limit_rate = @var{rate} 2972Limit the download speed to no more than @var{rate} bytes per second. 2973The same as @samp{--limit-rate=@var{rate}}. 2974 2975@item load_cookies = @var{file} 2976Load cookies from @var{file}. See @samp{--load-cookies @var{file}}. 2977 2978@item local_encoding = @var{encoding} 2979Force Wget to use @var{encoding} as the default system encoding. See 2980@samp{--local-encoding}. 2981 2982@item logfile = @var{file} 2983Set logfile to @var{file}, the same as @samp{-o @var{file}}. 2984 2985@item max_redirect = @var{number} 2986Specifies the maximum number of redirections to follow for a resource. 2987See @samp{--max-redirect=@var{number}}. 2988 2989@item mirror = on/off 2990Turn mirroring on/off. The same as @samp{-m}. 2991 2992@item netrc = on/off 2993Turn reading netrc on or off. 2994 2995@item no_clobber = on/off 2996Same as @samp{-nc}. 2997 2998@item no_parent = on/off 2999Disallow retrieving outside the directory hierarchy, like 3000@samp{--no-parent} (@pxref{Directory-Based Limits}). 3001 3002@item no_proxy = @var{string} 3003Use @var{string} as the comma-separated list of domains to avoid in 3004proxy loading, instead of the one specified in environment. 3005 3006@item output_document = @var{file} 3007Set the output filename---the same as @samp{-O @var{file}}. 3008 3009@item page_requisites = on/off 3010Download all ancillary documents necessary for a single @sc{html} page to 3011display properly---the same as @samp{-p}. 3012 3013@item passive_ftp = on/off 3014Change setting of passive @sc{ftp}, equivalent to the 3015@samp{--passive-ftp} option. 3016 3017@itemx password = @var{string} 3018Specify password @var{string} for both @sc{ftp} and @sc{http} file retrieval. 3019This command can be overridden using the @samp{ftp_password} and 3020@samp{http_password} command for @sc{ftp} and @sc{http} respectively. 3021 3022@item post_data = @var{string} 3023Use POST as the method for all HTTP requests and send @var{string} in 3024the request body. The same as @samp{--post-data=@var{string}}. 3025 3026@item post_file = @var{file} 3027Use POST as the method for all HTTP requests and send the contents of 3028@var{file} in the request body. The same as 3029@samp{--post-file=@var{file}}. 3030 3031@item prefer_family = none/IPv4/IPv6 3032When given a choice of several addresses, connect to the addresses 3033with specified address family first. The address order returned by 3034DNS is used without change by default. The same as @samp{--prefer-family}, 3035which see for a detailed discussion of why this is useful. 3036 3037@item private_key = @var{file} 3038Set the private key file to @var{file}. The same as 3039@samp{--private-key=@var{file}}. 3040 3041@item private_key_type = @var{string} 3042Specify the type of the private key, legal values being @samp{PEM} 3043(the default) and @samp{DER} (aka ASN1). The same as 3044@samp{--private-type=@var{string}}. 3045 3046@item progress = @var{string} 3047Set the type of the progress indicator. Legal types are @samp{dot} 3048and @samp{bar}. Equivalent to @samp{--progress=@var{string}}. 3049 3050@item protocol_directories = on/off 3051When set, use the protocol name as a directory component of local file 3052names. The same as @samp{--protocol-directories}. 3053 3054@item proxy_password = @var{string} 3055Set proxy authentication password to @var{string}, like 3056@samp{--proxy-password=@var{string}}. 3057 3058@item proxy_user = @var{string} 3059Set proxy authentication user name to @var{string}, like 3060@samp{--proxy-user=@var{string}}. 3061 3062@item quiet = on/off 3063Quiet mode---the same as @samp{-q}. 3064 3065@item quota = @var{quota} 3066Specify the download quota, which is useful to put in the global 3067@file{wgetrc}. When download quota is specified, Wget will stop 3068retrieving after the download sum has become greater than quota. The 3069quota can be specified in bytes (default), kbytes @samp{k} appended) or 3070mbytes (@samp{m} appended). Thus @samp{quota = 5m} will set the quota 3071to 5 megabytes. Note that the user's startup file overrides system 3072settings. 3073 3074@item random_file = @var{file} 3075Use @var{file} as a source of randomness on systems lacking 3076@file{/dev/random}. 3077 3078@item random_wait = on/off 3079Turn random between-request wait times on or off. The same as 3080@samp{--random-wait}. 3081 3082@item read_timeout = @var{n} 3083Set the read (and write) timeout---the same as 3084@samp{--read-timeout=@var{n}}. 3085 3086@item reclevel = @var{n} 3087Recursion level (depth)---the same as @samp{-l @var{n}}. 3088 3089@item recursive = on/off 3090Recursive on/off---the same as @samp{-r}. 3091 3092@item referer = @var{string} 3093Set HTTP @samp{Referer:} header just like 3094@samp{--referer=@var{string}}. (Note that it was the folks who wrote 3095the @sc{http} spec who got the spelling of ``referrer'' wrong.) 3096 3097@item relative_only = on/off 3098Follow only relative links---the same as @samp{-L} (@pxref{Relative 3099Links}). 3100 3101@item remote_encoding = @var{encoding} 3102Force Wget to use @var{encoding} as the default remote server encoding. 3103See @samp{--remote-encoding}. 3104 3105@item remove_listing = on/off 3106If set to on, remove @sc{ftp} listings downloaded by Wget. Setting it 3107to off is the same as @samp{--no-remove-listing}. 3108 3109@item restrict_file_names = unix/windows 3110Restrict the file names generated by Wget from URLs. See 3111@samp{--restrict-file-names} for a more detailed description. 3112 3113@item retr_symlinks = on/off 3114When set to on, retrieve symbolic links as if they were plain files; the 3115same as @samp{--retr-symlinks}. 3116 3117@item retry_connrefused = on/off 3118When set to on, consider ``connection refused'' a transient 3119error---the same as @samp{--retry-connrefused}. 3120 3121@item robots = on/off 3122Specify whether the norobots convention is respected by Wget, ``on'' by 3123default. This switch controls both the @file{/robots.txt} and the 3124@samp{nofollow} aspect of the spec. @xref{Robot Exclusion}, for more 3125details about this. Be sure you know what you are doing before turning 3126this off. 3127 3128@item save_cookies = @var{file} 3129Save cookies to @var{file}. The same as @samp{--save-cookies 3130@var{file}}. 3131 3132@item save_headers = on/off 3133Same as @samp{--save-headers}. 3134 3135@item secure_protocol = @var{string} 3136Choose the secure protocol to be used. Legal values are @samp{auto} 3137(the default), @samp{SSLv2}, @samp{SSLv3}, and @samp{TLSv1}. The same 3138as @samp{--secure-protocol=@var{string}}. 3139 3140@item server_response = on/off 3141Choose whether or not to print the @sc{http} and @sc{ftp} server 3142responses---the same as @samp{-S}. 3143 3144@item span_hosts = on/off 3145Same as @samp{-H}. 3146 3147@item spider = on/off 3148Same as @samp{--spider}. 3149 3150@item strict_comments = on/off 3151Same as @samp{--strict-comments}. 3152 3153@item timeout = @var{n} 3154Set all applicable timeout values to @var{n}, the same as @samp{-T 3155@var{n}}. 3156 3157@item timestamping = on/off 3158Turn timestamping on/off. The same as @samp{-N} (@pxref{Time-Stamping}). 3159 3160@item tries = @var{n} 3161Set number of retries per @sc{url}---the same as @samp{-t @var{n}}. 3162 3163@item use_proxy = on/off 3164When set to off, don't use proxy even when proxy-related environment 3165variables are set. In that case it is the same as using 3166@samp{--no-proxy}. 3167 3168@item user = @var{string} 3169Specify username @var{string} for both @sc{ftp} and @sc{http} file retrieval. 3170This command can be overridden using the @samp{ftp_user} and 3171@samp{http_user} command for @sc{ftp} and @sc{http} respectively. 3172 3173@item user_agent = @var{string} 3174User agent identification sent to the HTTP Server---the same as 3175@samp{--user-agent=@var{string}}. 3176 3177@item verbose = on/off 3178Turn verbose on/off---the same as @samp{-v}/@samp{-nv}. 3179 3180@item wait = @var{n} 3181Wait @var{n} seconds between retrievals---the same as @samp{-w 3182@var{n}}. 3183 3184@item wait_retry = @var{n} 3185Wait up to @var{n} seconds between retries of failed retrievals 3186only---the same as @samp{--waitretry=@var{n}}. Note that this is 3187turned on by default in the global @file{wgetrc}. 3188@end table 3189 3190@node Sample Wgetrc, , Wgetrc Commands, Startup File 3191@section Sample Wgetrc 3192@cindex sample wgetrc 3193 3194This is the sample initialization file, as given in the distribution. 3195It is divided in two section---one for global usage (suitable for global 3196startup file), and one for local usage (suitable for 3197@file{$HOME/.wgetrc}). Be careful about the things you change. 3198 3199Note that almost all the lines are commented out. For a command to have 3200any effect, you must remove the @samp{#} character at the beginning of 3201its line. 3202 3203@example 3204@include sample.wgetrc.munged_for_texi_inclusion 3205@end example 3206 3207@node Examples, Various, Startup File, Top 3208@chapter Examples 3209@cindex examples 3210 3211@c man begin EXAMPLES 3212The examples are divided into three sections loosely based on their 3213complexity. 3214 3215@menu 3216* Simple Usage:: Simple, basic usage of the program. 3217* Advanced Usage:: Advanced tips. 3218* Very Advanced Usage:: The hairy stuff. 3219@end menu 3220 3221@node Simple Usage, Advanced Usage, Examples, Examples 3222@section Simple Usage 3223 3224@itemize @bullet 3225@item 3226Say you want to download a @sc{url}. Just type: 3227 3228@example 3229wget http://fly.srk.fer.hr/ 3230@end example 3231 3232@item 3233But what will happen if the connection is slow, and the file is lengthy? 3234The connection will probably fail before the whole file is retrieved, 3235more than once. In this case, Wget will try getting the file until it 3236either gets the whole of it, or exceeds the default number of retries 3237(this being 20). It is easy to change the number of tries to 45, to 3238insure that the whole file will arrive safely: 3239 3240@example 3241wget --tries=45 http://fly.srk.fer.hr/jpg/flyweb.jpg 3242@end example 3243 3244@item 3245Now let's leave Wget to work in the background, and write its progress 3246to log file @file{log}. It is tiring to type @samp{--tries}, so we 3247shall use @samp{-t}. 3248 3249@example 3250wget -t 45 -o log http://fly.srk.fer.hr/jpg/flyweb.jpg & 3251@end example 3252 3253The ampersand at the end of the line makes sure that Wget works in the 3254background. To unlimit the number of retries, use @samp{-t inf}. 3255 3256@item 3257The usage of @sc{ftp} is as simple. Wget will take care of login and 3258password. 3259 3260@example 3261wget ftp://gnjilux.srk.fer.hr/welcome.msg 3262@end example 3263 3264@item 3265If you specify a directory, Wget will retrieve the directory listing, 3266parse it and convert it to @sc{html}. Try: 3267 3268@example 3269wget ftp://ftp.gnu.org/pub/gnu/ 3270links index.html 3271@end example 3272@end itemize 3273 3274@node Advanced Usage, Very Advanced Usage, Simple Usage, Examples 3275@section Advanced Usage 3276 3277@itemize @bullet 3278@item 3279You have a file that contains the URLs you want to download? Use the 3280@samp{-i} switch: 3281 3282@example 3283wget -i @var{file} 3284@end example 3285 3286If you specify @samp{-} as file name, the @sc{url}s will be read from 3287standard input. 3288 3289@item 3290Create a five levels deep mirror image of the GNU web site, with the 3291same directory structure the original has, with only one try per 3292document, saving the log of the activities to @file{gnulog}: 3293 3294@example 3295wget -r http://www.gnu.org/ -o gnulog 3296@end example 3297 3298@item 3299The same as the above, but convert the links in the downloaded files to 3300point to local files, so you can view the documents off-line: 3301 3302@example 3303wget --convert-links -r http://www.gnu.org/ -o gnulog 3304@end example 3305 3306@item 3307Retrieve only one @sc{html} page, but make sure that all the elements needed 3308for the page to be displayed, such as inline images and external style 3309sheets, are also downloaded. Also make sure the downloaded page 3310references the downloaded links. 3311 3312@example 3313wget -p --convert-links http://www.server.com/dir/page.html 3314@end example 3315 3316The @sc{html} page will be saved to @file{www.server.com/dir/page.html}, and 3317the images, stylesheets, etc., somewhere under @file{www.server.com/}, 3318depending on where they were on the remote server. 3319 3320@item 3321The same as the above, but without the @file{www.server.com/} directory. 3322In fact, I don't want to have all those random server directories 3323anyway---just save @emph{all} those files under a @file{download/} 3324subdirectory of the current directory. 3325 3326@example 3327wget -p --convert-links -nH -nd -Pdownload \ 3328 http://www.server.com/dir/page.html 3329@end example 3330 3331@item 3332Retrieve the index.html of @samp{www.lycos.com}, showing the original 3333server headers: 3334 3335@example 3336wget -S http://www.lycos.com/ 3337@end example 3338 3339@item 3340Save the server headers with the file, perhaps for post-processing. 3341 3342@example 3343wget --save-headers http://www.lycos.com/ 3344more index.html 3345@end example 3346 3347@item 3348Retrieve the first two levels of @samp{wuarchive.wustl.edu}, saving them 3349to @file{/tmp}. 3350 3351@example 3352wget -r -l2 -P/tmp ftp://wuarchive.wustl.edu/ 3353@end example 3354 3355@item 3356You want to download all the @sc{gif}s from a directory on an @sc{http} 3357server. You tried @samp{wget http://www.server.com/dir/*.gif}, but that 3358didn't work because @sc{http} retrieval does not support globbing. In 3359that case, use: 3360 3361@example 3362wget -r -l1 --no-parent -A.gif http://www.server.com/dir/ 3363@end example 3364 3365More verbose, but the effect is the same. @samp{-r -l1} means to 3366retrieve recursively (@pxref{Recursive Download}), with maximum depth 3367of 1. @samp{--no-parent} means that references to the parent directory 3368are ignored (@pxref{Directory-Based Limits}), and @samp{-A.gif} means to 3369download only the @sc{gif} files. @samp{-A "*.gif"} would have worked 3370too. 3371 3372@item 3373Suppose you were in the middle of downloading, when Wget was 3374interrupted. Now you do not want to clobber the files already present. 3375It would be: 3376 3377@example 3378wget -nc -r http://www.gnu.org/ 3379@end example 3380 3381@item 3382If you want to encode your own username and password to @sc{http} or 3383@sc{ftp}, use the appropriate @sc{url} syntax (@pxref{URL Format}). 3384 3385@example 3386wget ftp://hniksic:mypassword@@unix.server.com/.emacs 3387@end example 3388 3389Note, however, that this usage is not advisable on multi-user systems 3390because it reveals your password to anyone who looks at the output of 3391@code{ps}. 3392 3393@cindex redirecting output 3394@item 3395You would like the output documents to go to standard output instead of 3396to files? 3397 3398@example 3399wget -O - http://jagor.srce.hr/ http://www.srce.hr/ 3400@end example 3401 3402You can also combine the two options and make pipelines to retrieve the 3403documents from remote hotlists: 3404 3405@example 3406wget -O - http://cool.list.com/ | wget --force-html -i - 3407@end example 3408@end itemize 3409 3410@node Very Advanced Usage, , Advanced Usage, Examples 3411@section Very Advanced Usage 3412 3413@cindex mirroring 3414@itemize @bullet 3415@item 3416If you wish Wget to keep a mirror of a page (or @sc{ftp} 3417subdirectories), use @samp{--mirror} (@samp{-m}), which is the shorthand 3418for @samp{-r -l inf -N}. You can put Wget in the crontab file asking it 3419to recheck a site each Sunday: 3420 3421@example 3422crontab 34230 0 * * 0 wget --mirror http://www.gnu.org/ -o /home/me/weeklog 3424@end example 3425 3426@item 3427In addition to the above, you want the links to be converted for local 3428viewing. But, after having read this manual, you know that link 3429conversion doesn't play well with timestamping, so you also want Wget to 3430back up the original @sc{html} files before the conversion. Wget invocation 3431would look like this: 3432 3433@example 3434wget --mirror --convert-links --backup-converted \ 3435 http://www.gnu.org/ -o /home/me/weeklog 3436@end example 3437 3438@item 3439But you've also noticed that local viewing doesn't work all that well 3440when @sc{html} files are saved under extensions other than @samp{.html}, 3441perhaps because they were served as @file{index.cgi}. So you'd like 3442Wget to rename all the files served with content-type @samp{text/html} 3443or @samp{application/xhtml+xml} to @file{@var{name}.html}. 3444 3445@example 3446wget --mirror --convert-links --backup-converted \ 3447 --html-extension -o /home/me/weeklog \ 3448 http://www.gnu.org/ 3449@end example 3450 3451Or, with less typing: 3452 3453@example 3454wget -m -k -K -E http://www.gnu.org/ -o /home/me/weeklog 3455@end example 3456@end itemize 3457@c man end 3458 3459@node Various, Appendices, Examples, Top 3460@chapter Various 3461@cindex various 3462 3463This chapter contains all the stuff that could not fit anywhere else. 3464 3465@menu 3466* Proxies:: Support for proxy servers. 3467* Distribution:: Getting the latest version. 3468* Web Site:: GNU Wget's presence on the World Wide Web. 3469* Mailing Lists:: Wget mailing list for announcements and discussion. 3470* Internet Relay Chat:: Wget's presence on IRC. 3471* Reporting Bugs:: How and where to report bugs. 3472* Portability:: The systems Wget works on. 3473* Signals:: Signal-handling performed by Wget. 3474@end menu 3475 3476@node Proxies, Distribution, Various, Various 3477@section Proxies 3478@cindex proxies 3479 3480@dfn{Proxies} are special-purpose @sc{http} servers designed to transfer 3481data from remote servers to local clients. One typical use of proxies 3482is lightening network load for users behind a slow connection. This is 3483achieved by channeling all @sc{http} and @sc{ftp} requests through the 3484proxy which caches the transferred data. When a cached resource is 3485requested again, proxy will return the data from cache. Another use for 3486proxies is for companies that separate (for security reasons) their 3487internal networks from the rest of Internet. In order to obtain 3488information from the Web, their users connect and retrieve remote data 3489using an authorized proxy. 3490 3491Wget supports proxies for both @sc{http} and @sc{ftp} retrievals. The 3492standard way to specify proxy location, which Wget recognizes, is using 3493the following environment variables: 3494 3495@table @code 3496@item http_proxy 3497@itemx https_proxy 3498If set, the @code{http_proxy} and @code{https_proxy} variables should 3499contain the @sc{url}s of the proxies for @sc{http} and @sc{https} 3500connections respectively. 3501 3502@item ftp_proxy 3503This variable should contain the @sc{url} of the proxy for @sc{ftp} 3504connections. It is quite common that @code{http_proxy} and 3505@code{ftp_proxy} are set to the same @sc{url}. 3506 3507@item no_proxy 3508This variable should contain a comma-separated list of domain extensions 3509proxy should @emph{not} be used for. For instance, if the value of 3510@code{no_proxy} is @samp{.mit.edu}, proxy will not be used to retrieve 3511documents from MIT. 3512@end table 3513 3514In addition to the environment variables, proxy location and settings 3515may be specified from within Wget itself. 3516 3517@table @samp 3518@itemx --no-proxy 3519@itemx proxy = on/off 3520This option and the corresponding command may be used to suppress the 3521use of proxy, even if the appropriate environment variables are set. 3522 3523@item http_proxy = @var{URL} 3524@itemx https_proxy = @var{URL} 3525@itemx ftp_proxy = @var{URL} 3526@itemx no_proxy = @var{string} 3527These startup file variables allow you to override the proxy settings 3528specified by the environment. 3529@end table 3530 3531Some proxy servers require authorization to enable you to use them. The 3532authorization consists of @dfn{username} and @dfn{password}, which must 3533be sent by Wget. As with @sc{http} authorization, several 3534authentication schemes exist. For proxy authorization only the 3535@code{Basic} authentication scheme is currently implemented. 3536 3537You may specify your username and password either through the proxy 3538@sc{url} or through the command-line options. Assuming that the 3539company's proxy is located at @samp{proxy.company.com} at port 8001, a 3540proxy @sc{url} location containing authorization data might look like 3541this: 3542 3543@example 3544http://hniksic:mypassword@@proxy.company.com:8001/ 3545@end example 3546 3547Alternatively, you may use the @samp{proxy-user} and 3548@samp{proxy-password} options, and the equivalent @file{.wgetrc} 3549settings @code{proxy_user} and @code{proxy_password} to set the proxy 3550username and password. 3551 3552@node Distribution, Web Site, Proxies, Various 3553@section Distribution 3554@cindex latest version 3555 3556Like all GNU utilities, the latest version of Wget can be found at the 3557master GNU archive site ftp.gnu.org, and its mirrors. For example, 3558Wget @value{VERSION} can be found at 3559@url{ftp://ftp.gnu.org/pub/gnu/wget/wget-@value{VERSION}.tar.gz} 3560 3561@node Web Site, Mailing Lists, Distribution, Various 3562@section Web Site 3563@cindex web site 3564 3565The official web site for GNU Wget is at 3566@url{http://www.gnu.org/software/wget/}. However, most useful 3567information resides at ``The Wget Wgiki'', 3568@url{http://wget.addictivecode.org/}. 3569 3570@node Mailing Lists, Internet Relay Chat, Web Site, Various 3571@section Mailing Lists 3572@cindex mailing list 3573@cindex list 3574 3575@unnumberedsubsec Primary List 3576 3577The primary mailinglist for discussion, bug-reports, or questions 3578about GNU Wget is at @email{bug-wget@@gnu.org}. To subscribe, send an 3579email to @email{bug-wget-join@@gnu.org}, or visit 3580@url{http://lists.gnu.org/mailman/listinfo/bug-wget}. 3581 3582You do not need to subscribe to send a message to the list; however, 3583please note that unsubscribed messages are moderated, and may take a 3584while before they hit the list---@strong{usually around a day}. If 3585you want your message to show up immediately, please subscribe to the 3586list before posting. Archives for the list may be found at 3587@url{http://lists.gnu.org/pipermail/bug-wget/}. 3588 3589An NNTP/Usenettish gateway is also available via 3590@uref{http://gmane.org/about.php,Gmane}. You can see the Gmane 3591archives at 3592@url{http://news.gmane.org/gmane.comp.web.wget.general}. Note that the 3593Gmane archives conveniently include messages from both the current 3594list, and the previous one. Messages also show up in the Gmane 3595archives sooner than they do at @url{lists.gnu.org}. 3596 3597@unnumberedsubsec Bug Notices List 3598 3599Additionally, there is the @email{wget-notify@@addictivecode.org} mailing 3600list. This is a non-discussion list that receives bug report 3601notifications from the bug-tracker. To subscribe to this list, 3602send an email to @email{wget-notify-join@@addictivecode.org}, 3603or visit @url{http://addictivecode.org/mailman/listinfo/wget-notify}. 3604 3605@unnumberedsubsec Obsolete Lists 3606 3607Previously, the mailing list @email{wget@@sunsite.dk} was used as the 3608main discussion list, and another list, 3609@email{wget-patches@@sunsite.dk} was used for submitting and 3610discussing patches to GNU Wget. 3611 3612Messages from @email{wget@@sunsite.dk} are archived at 3613@itemize @tie{} 3614@item 3615@url{http://www.mail-archive.com/wget%40sunsite.dk/} and at 3616@item 3617@url{http://news.gmane.org/gmane.comp.web.wget.general} (which also 3618continues to archive the current list, @email{bug-wget@@gnu.org}). 3619@end itemize 3620 3621Messages from @email{wget-patches@@sunsite.dk} are archived at 3622@itemize @tie{} 3623@item 3624@url{http://news.gmane.org/gmane.comp.web.wget.patches}. 3625@end itemize 3626 3627@node Internet Relay Chat, Reporting Bugs, Mailing Lists, Various 3628@section Internet Relay Chat 3629@cindex Internet Relay Chat 3630@cindex IRC 3631@cindex #wget 3632 3633In addition to the mailinglists, we also have a support channel set up 3634via IRC at @code{irc.freenode.org}, @code{#wget}. Come check it out! 3635 3636@node Reporting Bugs, Portability, Internet Relay Chat, Various 3637@section Reporting Bugs 3638@cindex bugs 3639@cindex reporting bugs 3640@cindex bug reports 3641 3642@c man begin BUGS 3643You are welcome to submit bug reports via the GNU Wget bug tracker (see 3644@url{http://wget.addictivecode.org/BugTracker}). 3645 3646Before actually submitting a bug report, please try to follow a few 3647simple guidelines. 3648 3649@enumerate 3650@item 3651Please try to ascertain that the behavior you see really is a bug. If 3652Wget crashes, it's a bug. If Wget does not behave as documented, 3653it's a bug. If things work strange, but you are not sure about the way 3654they are supposed to work, it might well be a bug, but you might want to 3655double-check the documentation and the mailing lists (@pxref{Mailing 3656Lists}). 3657 3658@item 3659Try to repeat the bug in as simple circumstances as possible. E.g. if 3660Wget crashes while downloading @samp{wget -rl0 -kKE -t5 --no-proxy 3661http://yoyodyne.com -o /tmp/log}, you should try to see if the crash is 3662repeatable, and if will occur with a simpler set of options. You might 3663even try to start the download at the page where the crash occurred to 3664see if that page somehow triggered the crash. 3665 3666Also, while I will probably be interested to know the contents of your 3667@file{.wgetrc} file, just dumping it into the debug message is probably 3668a bad idea. Instead, you should first try to see if the bug repeats 3669with @file{.wgetrc} moved out of the way. Only if it turns out that 3670@file{.wgetrc} settings affect the bug, mail me the relevant parts of 3671the file. 3672 3673@item 3674Please start Wget with @samp{-d} option and send us the resulting 3675output (or relevant parts thereof). If Wget was compiled without 3676debug support, recompile it---it is @emph{much} easier to trace bugs 3677with debug support on. 3678 3679Note: please make sure to remove any potentially sensitive information 3680from the debug log before sending it to the bug address. The 3681@code{-d} won't go out of its way to collect sensitive information, 3682but the log @emph{will} contain a fairly complete transcript of Wget's 3683communication with the server, which may include passwords and pieces 3684of downloaded data. Since the bug address is publically archived, you 3685may assume that all bug reports are visible to the public. 3686 3687@item 3688If Wget has crashed, try to run it in a debugger, e.g. @code{gdb `which 3689wget` core} and type @code{where} to get the backtrace. This may not 3690work if the system administrator has disabled core files, but it is 3691safe to try. 3692@end enumerate 3693@c man end 3694 3695@node Portability, Signals, Reporting Bugs, Various 3696@section Portability 3697@cindex portability 3698@cindex operating systems 3699 3700Like all GNU software, Wget works on the GNU system. However, since it 3701uses GNU Autoconf for building and configuring, and mostly avoids using 3702``special'' features of any particular Unix, it should compile (and 3703work) on all common Unix flavors. 3704 3705Various Wget versions have been compiled and tested under many kinds of 3706Unix systems, including GNU/Linux, Solaris, SunOS 4.x, Mac OS X, OSF 3707(aka Digital Unix or Tru64), Ultrix, *BSD, IRIX, AIX, and others. Some 3708of those systems are no longer in widespread use and may not be able to 3709support recent versions of Wget. If Wget fails to compile on your 3710system, we would like to know about it. 3711 3712Thanks to kind contributors, this version of Wget compiles and works 3713on 32-bit Microsoft Windows platforms. It has been compiled 3714successfully using MS Visual C++ 6.0, Watcom, Borland C, and GCC 3715compilers. Naturally, it is crippled of some features available on 3716Unix, but it should work as a substitute for people stuck with 3717Windows. Note that Windows-specific portions of Wget are not 3718guaranteed to be supported in the future, although this has been the 3719case in practice for many years now. All questions and problems in 3720Windows usage should be reported to Wget mailing list at 3721@email{wget@@sunsite.dk} where the volunteers who maintain the 3722Windows-related features might look at them. 3723 3724Support for building on MS-DOS via DJGPP has been contributed by Gisle 3725Vanem; a port to VMS is maintained by Steven Schweda, and is available 3726at @url{http://antinode.org/}. 3727 3728@node Signals, , Portability, Various 3729@section Signals 3730@cindex signal handling 3731@cindex hangup 3732 3733Since the purpose of Wget is background work, it catches the hangup 3734signal (@code{SIGHUP}) and ignores it. If the output was on standard 3735output, it will be redirected to a file named @file{wget-log}. 3736Otherwise, @code{SIGHUP} is ignored. This is convenient when you wish 3737to redirect the output of Wget after having started it. 3738 3739@example 3740$ wget http://www.gnus.org/dist/gnus.tar.gz & 3741... 3742$ kill -HUP %% 3743SIGHUP received, redirecting output to `wget-log'. 3744@end example 3745 3746Other than that, Wget will not try to interfere with signals in any way. 3747@kbd{C-c}, @code{kill -TERM} and @code{kill -KILL} should kill it alike. 3748 3749@node Appendices, Copying this manual, Various, Top 3750@chapter Appendices 3751 3752This chapter contains some references I consider useful. 3753 3754@menu 3755* Robot Exclusion:: Wget's support for RES. 3756* Security Considerations:: Security with Wget. 3757* Contributors:: People who helped. 3758@end menu 3759 3760@node Robot Exclusion, Security Considerations, Appendices, Appendices 3761@section Robot Exclusion 3762@cindex robot exclusion 3763@cindex robots.txt 3764@cindex server maintenance 3765 3766It is extremely easy to make Wget wander aimlessly around a web site, 3767sucking all the available data in progress. @samp{wget -r @var{site}}, 3768and you're set. Great? Not for the server admin. 3769 3770As long as Wget is only retrieving static pages, and doing it at a 3771reasonable rate (see the @samp{--wait} option), there's not much of a 3772problem. The trouble is that Wget can't tell the difference between the 3773smallest static page and the most demanding CGI. A site I know has a 3774section handled by a CGI Perl script that converts Info files to @sc{html} on 3775the fly. The script is slow, but works well enough for human users 3776viewing an occasional Info file. However, when someone's recursive Wget 3777download stumbles upon the index page that links to all the Info files 3778through the script, the system is brought to its knees without providing 3779anything useful to the user (This task of converting Info files could be 3780done locally and access to Info documentation for all installed GNU 3781software on a system is available from the @code{info} command). 3782 3783To avoid this kind of accident, as well as to preserve privacy for 3784documents that need to be protected from well-behaved robots, the 3785concept of @dfn{robot exclusion} was invented. The idea is that 3786the server administrators and document authors can specify which 3787portions of the site they wish to protect from robots and those 3788they will permit access. 3789 3790The most popular mechanism, and the @i{de facto} standard supported by 3791all the major robots, is the ``Robots Exclusion Standard'' (RES) written 3792by Martijn Koster et al. in 1994. It specifies the format of a text 3793file containing directives that instruct the robots which URL paths to 3794avoid. To be found by the robots, the specifications must be placed in 3795@file{/robots.txt} in the server root, which the robots are expected to 3796download and parse. 3797 3798Although Wget is not a web robot in the strictest sense of the word, it 3799can download large parts of the site without the user's intervention to 3800download an individual page. Because of that, Wget honors RES when 3801downloading recursively. For instance, when you issue: 3802 3803@example 3804wget -r http://www.server.com/ 3805@end example 3806 3807First the index of @samp{www.server.com} will be downloaded. If Wget 3808finds that it wants to download more documents from that server, it will 3809request @samp{http://www.server.com/robots.txt} and, if found, use it 3810for further downloads. @file{robots.txt} is loaded only once per each 3811server. 3812 3813Until version 1.8, Wget supported the first version of the standard, 3814written by Martijn Koster in 1994 and available at 3815@url{http://www.robotstxt.org/wc/norobots.html}. As of version 1.8, 3816Wget has supported the additional directives specified in the internet 3817draft @samp{<draft-koster-robots-00.txt>} titled ``A Method for Web 3818Robots Control''. The draft, which has as far as I know never made to 3819an @sc{rfc}, is available at 3820@url{http://www.robotstxt.org/wc/norobots-rfc.txt}. 3821 3822This manual no longer includes the text of the Robot Exclusion Standard. 3823 3824The second, less known mechanism, enables the author of an individual 3825document to specify whether they want the links from the file to be 3826followed by a robot. This is achieved using the @code{META} tag, like 3827this: 3828 3829@example 3830<meta name="robots" content="nofollow"> 3831@end example 3832 3833This is explained in some detail at 3834@url{http://www.robotstxt.org/wc/meta-user.html}. Wget supports this 3835method of robot exclusion in addition to the usual @file{/robots.txt} 3836exclusion. 3837 3838If you know what you are doing and really really wish to turn off the 3839robot exclusion, set the @code{robots} variable to @samp{off} in your 3840@file{.wgetrc}. You can achieve the same effect from the command line 3841using the @code{-e} switch, e.g. @samp{wget -e robots=off @var{url}...}. 3842 3843@node Security Considerations, Contributors, Robot Exclusion, Appendices 3844@section Security Considerations 3845@cindex security 3846 3847When using Wget, you must be aware that it sends unencrypted passwords 3848through the network, which may present a security problem. Here are the 3849main issues, and some solutions. 3850 3851@enumerate 3852@item 3853The passwords on the command line are visible using @code{ps}. The best 3854way around it is to use @code{wget -i -} and feed the @sc{url}s to 3855Wget's standard input, each on a separate line, terminated by @kbd{C-d}. 3856Another workaround is to use @file{.netrc} to store passwords; however, 3857storing unencrypted passwords is also considered a security risk. 3858 3859@item 3860Using the insecure @dfn{basic} authentication scheme, unencrypted 3861passwords are transmitted through the network routers and gateways. 3862 3863@item 3864The @sc{ftp} passwords are also in no way encrypted. There is no good 3865solution for this at the moment. 3866 3867@item 3868Although the ``normal'' output of Wget tries to hide the passwords, 3869debugging logs show them, in all forms. This problem is avoided by 3870being careful when you send debug logs (yes, even when you send them to 3871me). 3872@end enumerate 3873 3874@node Contributors, , Security Considerations, Appendices 3875@section Contributors 3876@cindex contributors 3877 3878@iftex 3879GNU Wget was written by Hrvoje Nik@v{s}i@'{c} @email{hniksic@@xemacs.org}, 3880@end iftex 3881@ifnottex 3882GNU Wget was written by Hrvoje Niksic @email{hniksic@@xemacs.org}, 3883@end ifnottex 3884and it is currently maintained by Micah Cowan @email{micah@@cowan.name}. 3885 3886However, the development of Wget could never have gone as far as it has, were 3887it not for the help of many people, either with bug reports, feature proposals, 3888patches, or letters saying ``Thanks!''. 3889 3890Special thanks goes to the following people (no particular order): 3891 3892@itemize @bullet 3893@item Dan Harkless---contributed a lot of code and documentation of 3894extremely high quality, as well as the @code{--page-requisites} and 3895related options. He was the principal maintainer for some time and 3896released Wget 1.6. 3897 3898@item Ian Abbott---contributed bug fixes, Windows-related fixes, and 3899provided a prototype implementation of the breadth-first recursive 3900download. Co-maintained Wget during the 1.8 release cycle. 3901 3902@item 3903The dotsrc.org crew, in particular Karsten Thygesen---donated system 3904resources such as the mailing list, web space, @sc{ftp} space, and 3905version control repositories, along with a lot of time to make these 3906actually work. Christian Reiniger was of invaluable help with setting 3907up Subversion. 3908 3909@item 3910Heiko Herold---provided high-quality Windows builds and contributed 3911bug and build reports for many years. 3912 3913@item 3914Shawn McHorse---bug reports and patches. 3915 3916@item 3917Kaveh R. Ghazi---on-the-fly @code{ansi2knr}-ization. Lots of 3918portability fixes. 3919 3920@item 3921Gordon Matzigkeit---@file{.netrc} support. 3922 3923@item 3924@iftex 3925Zlatko @v{C}alu@v{s}i@'{c}, Tomislav Vujec and Dra@v{z}en 3926Ka@v{c}ar---feature suggestions and ``philosophical'' discussions. 3927@end iftex 3928@ifnottex 3929Zlatko Calusic, Tomislav Vujec and Drazen Kacar---feature suggestions 3930and ``philosophical'' discussions. 3931@end ifnottex 3932 3933@item 3934Darko Budor---initial port to Windows. 3935 3936@item 3937Antonio Rosella---help and suggestions, plus the initial Italian 3938translation. 3939 3940@item 3941@iftex 3942Tomislav Petrovi@'{c}, Mario Miko@v{c}evi@'{c}---many bug reports and 3943suggestions. 3944@end iftex 3945@ifnottex 3946Tomislav Petrovic, Mario Mikocevic---many bug reports and suggestions. 3947@end ifnottex 3948 3949@item 3950@iftex 3951Fran@,{c}ois Pinard---many thorough bug reports and discussions. 3952@end iftex 3953@ifnottex 3954Francois Pinard---many thorough bug reports and discussions. 3955@end ifnottex 3956 3957@item 3958Karl Eichwalder---lots of help with internationalization, Makefile 3959layout and many other things. 3960 3961@item 3962Junio Hamano---donated support for Opie and @sc{http} @code{Digest} 3963authentication. 3964 3965@item 3966Mauro Tortonesi---improved IPv6 support, adding support for dual 3967family systems. Refactored and enhanced FTP IPv6 code. Maintained GNU 3968Wget from 2004--2007. 3969 3970@item 3971Christopher G.@: Lewis---maintenance of the Windows version of GNU WGet. 3972 3973@item 3974Gisle Vanem---many helpful patches and improvements, especially for 3975Windows and MS-DOS support. 3976 3977@item 3978Ralf Wildenhues---contributed patches to convert Wget to use Automake as 3979part of its build process, and various bugfixes. 3980 3981@item 3982Steven Schubiger---Many helpful patches, bugfixes and improvements. 3983Notably, conversion of Wget to use the Gnulib quotes and quoteargs 3984modules, and the addition of password prompts at the console, via the 3985Gnulib getpasswd-gnu module. 3986 3987@item 3988Ted Mielczarek---donated support for CSS. 3989 3990@item 3991Saint Xavier---Support for IRIs (RFC 3987). 3992 3993@item 3994People who provided donations for development---including Brian Gough. 3995@end itemize 3996 3997The following people have provided patches, bug/build reports, useful 3998suggestions, beta testing services, fan mail and all the other things 3999that make maintenance so much fun: 4000 4001Tim Adam, 4002Adrian Aichner, 4003Martin Baehr, 4004Dieter Baron, 4005Roger Beeman, 4006Dan Berger, 4007T.@: Bharath, 4008Christian Biere, 4009Paul Bludov, 4010Daniel Bodea, 4011Mark Boyns, 4012John Burden, 4013Julien Buty, 4014Wanderlei Cavassin, 4015Gilles Cedoc, 4016Tim Charron, 4017Noel Cragg, 4018@iftex 4019Kristijan @v{C}onka@v{s}, 4020@end iftex 4021@ifnottex 4022Kristijan Conkas, 4023@end ifnottex 4024John Daily, 4025Andreas Damm, 4026Ahmon Dancy, 4027Andrew Davison, 4028Bertrand Demiddelaer, 4029Alexander Dergachev, 4030Andrew Deryabin, 4031Ulrich Drepper, 4032Marc Duponcheel, 4033@iftex 4034Damir D@v{z}eko, 4035@end iftex 4036@ifnottex 4037Damir Dzeko, 4038@end ifnottex 4039Alan Eldridge, 4040Hans-Andreas Engel, 4041@iftex 4042Aleksandar Erkalovi@'{c}, 4043@end iftex 4044@ifnottex 4045Aleksandar Erkalovic, 4046@end ifnottex 4047Andy Eskilsson, 4048@iftex 4049Jo@~{a}o Ferreira, 4050@end iftex 4051@ifnottex 4052Joao Ferreira, 4053@end ifnottex 4054Christian Fraenkel, 4055David Fritz, 4056Mike Frysinger, 4057Charles C.@: Fu, 4058FUJISHIMA Satsuki, 4059Masashi Fujita, 4060Howard Gayle, 4061Marcel Gerrits, 4062Lemble Gregory, 4063Hans Grobler, 4064Alain Guibert, 4065Mathieu Guillaume, 4066Aaron Hawley, 4067Jochen Hein, 4068Karl Heuer, 4069Madhusudan Hosaagrahara, 4070HIROSE Masaaki, 4071Ulf Harnhammar, 4072Gregor Hoffleit, 4073Erik Magnus Hulthen, 4074Richard Huveneers, 4075Jonas Jensen, 4076Larry Jones, 4077Simon Josefsson, 4078@iftex 4079Mario Juri@'{c}, 4080@end iftex 4081@ifnottex 4082Mario Juric, 4083@end ifnottex 4084@iftex 4085Hack Kampbj@o rn, 4086@end iftex 4087@ifnottex 4088Hack Kampbjorn, 4089@end ifnottex 4090Const Kaplinsky, 4091@iftex 4092Goran Kezunovi@'{c}, 4093@end iftex 4094@ifnottex 4095Goran Kezunovic, 4096@end ifnottex 4097Igor Khristophorov, 4098Robert Kleine, 4099KOJIMA Haime, 4100Fila Kolodny, 4101Alexander Kourakos, 4102Martin Kraemer, 4103Sami Krank, 4104Jay Krell, 4105@tex 4106$\Sigma\acute{\iota}\mu o\varsigma\; 4107\Xi\varepsilon\nu\iota\tau\acute{\epsilon}\lambda\lambda\eta\varsigma$ 4108(Simos KSenitellis), 4109@end tex 4110@ifnottex 4111Simos KSenitellis, 4112@end ifnottex 4113Christian Lackas, 4114Hrvoje Lacko, 4115Daniel S.@: Lewart, 4116@iftex 4117Nicol@'{a}s Lichtmeier, 4118@end iftex 4119@ifnottex 4120Nicolas Lichtmeier, 4121@end ifnottex 4122Dave Love, 4123Alexander V.@: Lukyanov, 4124@iftex 4125Thomas Lu@ss{}nig, 4126@end iftex 4127@ifnottex 4128Thomas Lussnig, 4129@end ifnottex 4130Andre Majorel, 4131Aurelien Marchand, 4132Matthew J.@: Mellon, 4133Jordan Mendelson, 4134Ted Mielczarek, 4135Robert Millan, 4136Lin Zhe Min, 4137Jan Minar, 4138Tim Mooney, 4139Keith Moore, 4140Adam D.@: Moss, 4141Simon Munton, 4142Charlie Negyesi, 4143R.@: K.@: Owen, 4144Jim Paris, 4145Kenny Parnell, 4146Leonid Petrov, 4147Simone Piunno, 4148Andrew Pollock, 4149Steve Pothier, 4150@iftex 4151Jan P@v{r}ikryl, 4152@end iftex 4153@ifnottex 4154Jan Prikryl, 4155@end ifnottex 4156Marin Purgar, 4157@iftex 4158Csaba R@'{a}duly, 4159@end iftex 4160@ifnottex 4161Csaba Raduly, 4162@end ifnottex 4163Keith Refson, 4164Bill Richardson, 4165Tyler Riddle, 4166Tobias Ringstrom, 4167Jochen Roderburg, 4168@c Texinfo doesn't grok @'{@i}, so we have to use TeX itself. 4169@tex 4170Juan Jos\'{e} Rodr\'{\i}guez, 4171@end tex 4172@ifnottex 4173Juan Jose Rodriguez, 4174@end ifnottex 4175Maciej W.@: Rozycki, 4176Edward J.@: Sabol, 4177Heinz Salzmann, 4178Robert Schmidt, 4179Nicolas Schodet, 4180Benno Schulenberg, 4181Andreas Schwab, 4182Steven M.@: Schweda, 4183Chris Seawood, 4184Pranab Shenoy, 4185Dennis Smit, 4186Toomas Soome, 4187Tage Stabell-Kulo, 4188Philip Stadermann, 4189Daniel Stenberg, 4190Sven Sternberger, 4191Markus Strasser, 4192John Summerfield, 4193Szakacsits Szabolcs, 4194Mike Thomas, 4195Philipp Thomas, 4196Mauro Tortonesi, 4197Dave Turner, 4198Gisle Vanem, 4199Rabin Vincent, 4200Russell Vincent, 4201@iftex 4202@v{Z}eljko Vrba, 4203@end iftex 4204@ifnottex 4205Zeljko Vrba, 4206@end ifnottex 4207Charles G Waldman, 4208Douglas E.@: Wegscheid, 4209Ralf Wildenhues, 4210Joshua David Williams, 4211Benjamin Wolsey, 4212Saint Xavier, 4213YAMAZAKI Makoto, 4214Jasmin Zainul, 4215@iftex 4216Bojan @v{Z}drnja, 4217@end iftex 4218@ifnottex 4219Bojan Zdrnja, 4220@end ifnottex 4221Kristijan Zimmer, 4222Xin Zou. 4223 4224Apologies to all who I accidentally left out, and many thanks to all the 4225subscribers of the Wget mailing list. 4226 4227@node Copying this manual, Concept Index, Appendices, Top 4228@appendix Copying this manual 4229 4230@menu 4231* GNU Free Documentation License:: Licnse for copying this manual. 4232@end menu 4233 4234@node GNU Free Documentation License, , Copying this manual, Copying this manual 4235@appendixsec GNU Free Documentation License 4236@cindex FDL, GNU Free Documentation License 4237 4238@include fdl.texi 4239 4240 4241@node Concept Index, , Copying this manual, Top 4242@unnumbered Concept Index 4243@printindex cp 4244 4245@contents 4246 4247@bye 4248