EditPageHistoryDiffInfoLikePages

Annotated edit history of wget(1) version 1, including all changes. View license author blame.

Rev	Author	#	Line
1	perry	1	WGET
		2	!!!WGET
		3	NAME
		4	SYNOPSIS
		5	DESCRIPTION
		6	OPTIONS
		7	EXAMPLES
		8	FILES
		9	BUGS
		10	SEE ALSO
		11	AUTHOR
		12	COPYRIGHT
		13	----
		14	!!NAME
		15
		16
		17	wget - GNU Wget Manual
		18	!!SYNOPSIS
		19
		20
		21	wget [[''option'']... [[ ''URL''
		22	]...
		23	!!DESCRIPTION
		24
		25
		26	GNU Wget is a free utility for
		27	non-interactive download of files from the Web. It supports
		28	HTTP , HTTPS , and
		29	FTP protocols, as well as retrieval through
		30	HTTP proxies.
		31
		32
		33	Wget is non-interactive, meaning that it can work in the
		34	background, while the user is not logged on. This allows you
		35	to start a retrieval and disconnect from the system, letting
		36	Wget finish the work. By contrast, most of the Web browsers
		37	require constant user's presence, which can be a great
		38	hindrance when transferring a lot of data.
		39
		40
		41	Wget can follow links in HTML pages and
		42	create local versions of remote web sites, fully recreating
		43	the directory structure of the original site. This is
		44	sometimes referred to as ``recursive downloading.'' While
		45	doing that, Wget respects the Robot Exclusion Standard
		46	(''/robots.txt''). Wget can be instructed to convert the
		47	links in downloaded HTML files to the local
		48	files for offline viewing.
		49
		50
		51	Wget has been designed for robustness over slow or unstable
		52	network connections; if a download fails due to a network
		53	problem, it will keep retrying until the whole file has been
		54	retrieved. If the server supports regetting, it will
		55	instruct the server to continue the download from where it
		56	left off.
		57	!!OPTIONS
		58
		59
		60	__Basic Startup Options__
		61
		62
		63	__-V__
		64
		65
		66	__--version__
		67
		68
		69	Display the version of Wget.
		70
		71
		72	__-h__
		73
		74
		75	__--help__
		76
		77
		78	Print a help message describing all of Wget's command-line
		79	options.
		80
		81
		82	__-b__
		83
		84
		85	__--background__
		86
		87
		88	Go to background immediately after startup. If no output
		89	file is specified via the __-o__, output is redirected to
		90	''wget-log''.
		91
		92
		93	__-e__ ''command''
		94
		95
		96	__--execute__ ''command''
		97
		98
		99	Execute ''command'' as if it were a part of
		100	''.wgetrc''. A command thus invoked will be executed
		101	''after'' the commands in ''.wgetrc'', thus taking
		102	precedence over them.
		103
		104
		105	__Logging and Input File Options__
		106
		107
		108	__-o__ ''logfile''
		109
		110
		111	__--output-file=__''logfile''
		112
		113
		114	Log all messages to ''logfile''. The messages are
		115	normally reported to standard error.
		116
		117
		118	__-a__ ''logfile''
		119
		120
		121	__--append-output=__''logfile''
		122
		123
		124	Append to ''logfile''. This is the same as __-o__,
		125	only it appends to ''logfile'' instead of overwriting the
		126	old log file. If ''logfile'' does not exist, a new file
		127	is created.
		128
		129
		130	__-d__
		131
		132
		133	__--debug__
		134
		135
		136	Turn on debug output, meaning various information important
		137	to the developers of Wget if it does not work properly. Your
		138	system administrator may have chosen to compile Wget without
		139	debug support, in which case __-d__ will not work. Please
		140	note that compiling with debug support is always safe---Wget
		141	compiled with the debug support will ''not'' print any
		142	debug info unless requested with __-d__.
		143
		144
		145	__-q__
		146
		147
		148	__--quiet__
		149
		150
		151	Turn off Wget's output.
		152
		153
		154	__-v__
		155
		156
		157	__--verbose__
		158
		159
		160	Turn on verbose output, with all the available data. The
		161	default output is verbose.
		162
		163
		164	__-nv__
		165
		166
		167	__--non-verbose__
		168
		169
		170	Non-verbose output---turn off verbose without being
		171	completely quiet (use __-q__ for that), which means that
		172	error messages and basic information still get
		173	printed.
		174
		175
		176	__-i__ ''file''
		177
		178
		179	__--input-file=__''file''
		180
		181
		182	Read URLs from ''file'', in which case no URLs need to be
		183	on the command line. If there are URLs both on the command
		184	line and in an input file, those on the command lines will
		185	be the first ones to be retrieved. The ''file'' need not
		186	be an HTML document (but no harm if it
		187	is)---it is enough if the URLs are just listed
		188	sequentially.
		189
		190
		191	However, if you specify __--force-html__, the document
		192	will be regarded as __html__. In that case you may have
		193	problems with relative links, which you can solve either by
		194	adding to the
		195	documents or by specifying __--base=__''url'' on the
		196	command line.
		197
		198
		199	__-F__
		200
		201
		202	__--force-html__
		203
		204
		205	When input is read from a file, force it to be treated as an
		206	HTML file. This enables you to retrieve
		207	relative links from existing HTML files on
		208	your local disk, by adding
		209	to HTML , or
		210	using the __--base__ command-line option.
		211
		212
		213	__-B__ ''URL''
		214
		215
		216	__--base=__ ''URL''
		217
		218
		219	When used in conjunction with __-F__, prepends
		220	''URL'' to relative links in the file
		221	specified by __-i__.
		222
		223
		224	__Download Options__
		225
		226
		227	__--bind-address=__
		228	''ADDRESS''
		229
		230
		231	When making client TCP/IP connections,
		232	bind() to ''ADDRESS'' on the
		233	local machine. ''ADDRESS'' may be
		234	specified as a hostname or IP address. This
		235	option can be useful if your machine is bound to multiple
		236	IPs.
		237
		238
		239	__-t__ ''number''
		240
		241
		242	__--tries=__''number''
		243
		244
		245	Set number of retries to ''number''. Specify 0 or
		246	__inf__ for infinite retrying.
		247
		248
		249	__-O__ ''file''
		250
		251
		252	__--output-document=__''file''
		253
		254
		255	The documents will not be written to the appropriate files,
		256	but all will be concatenated together and written to
		257	''file''. If ''file'' already exists, it will be
		258	overwritten. If the ''file'' is __-__, the documents
		259	will be written to standard output. Including this option
		260	automatically sets the number of tries to 1.
		261
		262
		263	__-nc__
		264
		265
		266	__--no-clobber__
		267
		268
		269	If a file is downloaded more than once in the same
		270	directory, Wget's behavior depends on a few options,
		271	including __-nc__. In certain cases, the local file will
		272	be ''clobbered'', or overwritten, upon repeated download.
		273	In other cases it will be preserved.
		274
		275
		276	When running Wget without __-N__, __-nc__, or
		277	__-r__, downloading the same file in the same directory
		278	will result in the original copy of ''file'' being
		279	preserved and the second copy being named
		280	''file''__.1__. If that file is downloaded yet again,
		281	the third copy will be named ''file''__.2__, and so
		282	on. When __-nc__ is specified, this behavior is
		283	suppressed, and Wget will refuse to download newer copies of
		284	''file''. Therefore, ``no-clobber'' is actually
		285	a misnomer in this mode---it's not clobbering that's
		286	prevented (as the numeric suffixes were already preventing
		287	clobbering), but rather the multiple version saving that's
		288	prevented.
		289
		290
		291	When running Wget with __-r__, but without __-N__ or
		292	__-nc__, re-downloading a file will result in the new
		293	copy simply overwriting the old. Adding __-nc__ will
		294	prevent this behavior, instead causing the original version
		295	to be preserved and any newer copies on the server to be
		296	ignored.
		297
		298
		299	When running Wget with __-N__, with or without __-r__,
		300	the decision as to whether or not to download a newer copy
		301	of a file depends on the local and remote timestamp and size
		302	of the file. __-nc__ may not be specified at the same
		303	time as __-N__.
		304
		305
		306	Note that when __-nc__ is specified, files with the
		307	suffixes __.html__ or (yuck) __.htm__ will be loaded
		308	from the local disk and parsed as if they had been retrieved
		309	from the Web.
		310
		311
		312	__-c__
		313
		314
		315	__--continue__
		316
		317
		318	Continue getting a partially-downloaded file. This is useful
		319	when you want to finish up a download started by a previous
		320	instance of Wget, or by another program. For
		321	instance:
		322
		323
		324	wget -c ftp://sunsite.doc.ic.ac.uk/ls-lR.Z
		325	If there is a file named ''ls-lR.Z'' in the current directory, Wget will assume that it is the first portion of the remote file, and will ask the server to continue the retrieval from an offset equal to the length of the local file.
		326
		327
		328	Note that you don't need to specify this option if you just
		329	want the current invocation of Wget to retry downloading a
		330	file should the connection be lost midway through. This is
		331	the default behavior. __-c__ only affects resumption of
		332	downloads started ''prior'' to this invocation of Wget,
		333	and whose local files are still sitting around.
		334
		335
		336	Without __-c__, the previous example would just download
		337	the remote file to ''ls-lR.Z.1'', leaving the truncated
		338	''ls-lR.Z'' file alone.
		339
		340
		341	Beginning with Wget 1.7, if you use __-c__ on a non-empty
		342	file, and it turns out that the server does not support
		343	continued downloading, Wget will refuse to start the
		344	download from scratch, which would effectively ruin existing
		345	contents. If you really want the download to start from
		346	scratch, remove the file.
		347
		348
		349	Also beginning with Wget 1.7, if you use __-c__ on a file
		350	which is of equal size as the one on the server, Wget will
		351	refuse to download the file and print an explanatory
		352	message. The same happens when the file is smaller on the
		353	server than locally (presumably because it was changed on
		354	the server since your last download attempt)---because
		355	``continuing'' is not meaningful, no download
		356	occurs.
		357
		358
		359	On the other side of the coin, while using __-c__, any
		360	file that's bigger on the server than locally will be
		361	considered an incomplete download and only
		362	(length(remote) - length(local)) bytes will be
		363	downloaded and tacked onto the end of the local file. This
		364	behavior can be desirable in certain cases---for instance,
		365	you can use __wget -c__ to download just the new portion
		366	that's been appended to a data collection or log
		367	file.
		368
		369
		370	However, if the file is bigger on the server because it's
		371	been ''changed'', as opposed to just ''appended'' to,
		372	you'll end up with a garbled file. Wget has no way of
		373	verifying that the local file is really a valid prefix of
		374	the remote file. You need to be especially careful of this
		375	when using __-c__ in conjunction with __-r__, since
		376	every file will be considered as an ``incomplete download''
		377	candidate.
		378
		379
		380	Another instance where you'll get a garbled file if you try
		381	to use __-c__ is if you have a lame HTTP
		382	proxy that inserts a ``transfer interrupted'' string into
		383	the local file. In the future a ``rollback'' option may be
		384	added to deal with this case.
		385
		386
		387	Note that __-c__ only works with FTP
		388	servers and with HTTP servers that support
		389	the Range header.
		390
		391
		392	__--progress=__''type''
		393
		394
		395	Select the type of the progress indicator you wish to use.
		396	Legal indicators are ``dot'' and ``bar''.
		397
		398
		399	The ``dot'' indicator is used by default. It traces the
		400	retrieval by printing dots on the screen, each dot
		401	representing a fixed amount of downloaded data.
		402
		403
		404	When using the dotted retrieval, you may also set the
		405	''style'' by specifying the type as
		406	__dot:__''style''. Different styles assign different
		407	meaning to one dot. With the default style each dot
		408	represents 1K, there are ten dots in a cluster and 50 dots
		409	in a line. The binary style has a more
		410	``computer''-like orientation---8K dots, 16-dots clusters
		411	and 48 dots per line (which makes for 384K lines). The
		412	mega style is suitable for downloading very large
		413	files---each dot represents 64K retrieved, there are eight
		414	dots in a cluster, and 48 dots on each line (so each line
		415	contains 3M).
		416
		417
		418	Specifying __--progress=bar__ will draw a nice
		419	ASCII progress bar graphics (a.k.a
		420	``thermometer'' display) to indicate retrieval. If the
		421	output is not a TTY , this option will be
		422	ignored, and Wget will revert to the dot indicator. If you
		423	want to force the bar indicator, use
		424	__--progress=bar:force__.
		425
		426
		427	__-N__
		428
		429
		430	__--timestamping__
		431
		432
		433	Turn on time-stamping.
		434
		435
		436	__-S__
		437
		438
		439	__--server-response__
		440
		441
		442	Print the headers sent by HTTP servers and
		443	responses sent by FTP servers.
		444
		445
		446	__--spider__
		447
		448
		449	When invoked with this option, Wget will behave as a Web
		450	''spider'', which means that it will not download the
		451	pages, just check that they are there. You can use it to
		452	check your bookmarks, e.g. with:
		453
		454
		455	wget --spider --force-html -i bookmarks.html
		456	This feature needs much more work for Wget to get close to the functionality of real WWW spiders.
		457
		458
		459	__-T seconds__
		460
		461
		462	__--timeout=__''seconds''
		463
		464
		465	Set the read timeout to ''seconds'' seconds. Whenever a
		466	network read is issued, the file descriptor is checked for a
		467	timeout, which could otherwise leave a pending connection
		468	(uninterrupted read). The default timeout is 900 seconds
		469	(fifteen minutes). Setting timeout to 0 will disable
		470	checking for timeouts.
		471
		472
		473	Please do not lower the default timeout value with this
		474	option unless you know what you are doing.
		475
		476
		477	__-w__ ''seconds''
		478
		479
		480	__--wait=__''seconds''
		481
		482
		483	Wait the specified number of seconds between the retrievals.
		484	Use of this option is recommended, as it lightens the server
		485	load by making the requests less frequent. Instead of in
		486	seconds, the time can be specified in minutes using the
		487	m suffix, in hours using h suffix, or in
		488	days using d suffix.
		489
		490
		491	Specifying a large value for this option is useful if the
		492	network or the destination host is down, so that Wget can
		493	wait long enough to reasonably expect the network error to
		494	be fixed before the retry.
		495
		496
		497	__--waitretry=__''seconds''
		498
		499
		500	If you don't want Wget to wait between ''every''
		501	retrieval, but only between retries of failed downloads, you
		502	can use this option. Wget will use ''linear backoff'',
		503	waiting 1 second after the first failure on a given file,
		504	then waiting 2 seconds after the second failure on that
		505	file, up to the maximum number of ''seconds'' you
		506	specify. Therefore, a value of 10 will actually make Wget
		507	wait up to (1 + 2 + ... + 10) = 55 seconds per
		508	file.
		509
		510
		511	Note that this option is turned on by default in the global
		512	''wgetrc'' file.
		513
		514
		515	__--random-wait__
		516
		517
		518	Some web sites may perform log analysis to identify
		519	retrieval programs such as Wget by looking for statistically
		520	significant similarities in the time between requests. This
		521	option causes the time between requests to vary between 0
		522	and 2 * ''wait'' seconds, where ''wait'' was specified
		523	using the __-w__ or __--wait__ options, in order to
		524	mask Wget's presence from such analysis.
		525
		526
		527	A recent article in a publication devoted to development on
		528	a popular consumer platform provided code to perform this
		529	analysis on the fly. Its author suggested blocking at the
		530	class C address level to ensure automated retrieval programs
		531	were blocked despite changing DHCP-supplied
		532	addresses.
		533
		534
		535	The __--random-wait__ option was inspired by this
		536	ill-advised recommendation to block many unrelated users
		537	from a web site due to the actions of one.
		538
		539
		540	__-Y on/off__
		541
		542
		543	__--proxy=on/off__
		544
		545
		546	Turn proxy support on or off. The proxy is on by default if
		547	the appropriate environmental variable is
		548	defined.
		549
		550
		551	__-Q__ ''quota''
		552
		553
		554	__--quota=__''quota''
		555
		556
		557	Specify download quota for automatic retrievals. The value
		558	can be specified in bytes (default), kilobytes (with
		559	__k__ suffix), or megabytes (with __m__
		560	suffix).
		561
		562
		563	Note that quota will never affect downloading a single file.
		564	So if you specify __wget -Q10k
		565	ftp://wuarchive.wustl.edu/ls-lR.gz__, all of the
		566	''ls-lR.gz'' will be downloaded. The same goes even when
		567	several URLs are specified on the command-line. However,
		568	quota is respected when retrieving either recursively, or
		569	from an input file. Thus you may safely type __wget -Q2m -i
		570	sites__---download will be aborted when the quota is
		571	exceeded.
		572
		573
		574	Setting quota to 0 or to __inf__ unlimits the download
		575	quota.
		576
		577
		578	__Directory Options__
		579
		580
		581	__-nd__
		582
		583
		584	__--no-directories__
		585
		586
		587	Do not create a hierarchy of directories when retrieving
		588	recursively. With this option turned on, all files will get
		589	saved to the current directory, without clobbering (if a
		590	name shows up more than once, the filenames will get
		591	extensions __.n__).
		592
		593
		594	__-x__
		595
		596
		597	__--force-directories__
		598
		599
		600	The opposite of __-nd__---create a hierarchy of
		601	directories, even if one would not have been created
		602	otherwise. E.g. __wget -x
		603	http://fly.srk.fer.hr/robots.txt__ will save the
		604	downloaded file to
		605	''fly.srk.fer.hr/robots.txt''.
		606
		607
		608	__-nH__
		609
		610
		611	__--no-host-directories__
		612
		613
		614	Disable generation of host-prefixed directories. By default,
		615	invoking Wget with __-r http://fly.srk.fer.hr/__ will
		616	create a structure of directories beginning with
		617	''fly.srk.fer.hr/''. This option disables such
		618	behavior.
		619
		620
		621	__--cut-dirs=__''number''
		622
		623
		624	Ignore ''number'' directory components. This is useful
		625	for getting a fine-grained control over the directory where
		626	recursive retrieval will be saved.
		627
		628
		629	Take, for example, the directory at
		630	__ftp://ftp.xemacs.org/pub/xemacs/__. If you retrieve it
		631	with __-r__, it will be saved locally under
		632	''ftp.xemacs.org/pub/xemacs/''. While the __-nH__
		633	option can remove the ''ftp.xemacs.org/'' part, you are
		634	still stuck with ''pub/xemacs''. This is where
		635	__--cut-dirs__ comes in handy; it makes Wget not ``see''
		636	''number'' remote directory components. Here are several
		637	examples of how __--cut-dirs__ option works.
		638
		639
		640	No options -
		641	--cut-dirs=1 -
		642	If you just want to get rid of the directory structure, this option is similar to a combination of __-nd__ and __-P__. However, unlike __-nd__, __--cut-dirs__ does not lose with subdirectories---for instance, with __-nH --cut-dirs=1__, a ''beta/'' subdirectory will be placed to ''xemacs/beta'', as one would expect.
		643
		644
		645	__-P__ ''prefix''
		646
		647
		648	__--directory-prefix=__''prefix''
		649
		650
		651	Set directory prefix to ''prefix''. The ''directory
		652	prefix'' is the directory where all other files and
		653	subdirectories will be saved to, i.e. the top of the
		654	retrieval tree. The default is __.__ (the current
		655	directory).
		656
		657
		658	__HTTP Options__
		659
		660
		661	__-E__
		662
		663
		664	__--html-extension__
		665
		666
		667	If a file of type __text/html__ is downloaded and the
		668	URL does not end with the regexp
		669	__.[[Hh][[Tt][[Mm][[Ll]?__, this option will cause the suffix
		670	__.html__ to be appended to the local filename. This is
		671	useful, for instance, when you're mirroring a remote site
		672	that uses __.asp__ pages, but you want the mirrored pages
		673	to be viewable on your stock Apache server. Another good use
		674	for this is when you're downloading the output of CGIs. A
		675	URL like
		676	__http://site.com/article.cgi?25__ will be saved as
		677	''article.cgi?25.html''.
		678
		679
		680	Note that filenames changed in this way will be
		681	re-downloaded every time you re-mirror a site, because Wget
		682	can't tell that the local ''X.html'' file corresponds to
		683	remote URL ''X'' (since it doesn't yet
		684	know that the URL produces output of type
		685	__text/html__. To prevent this re-downloading, you must
		686	use __-k__ and __-K__ so that the original version of
		687	the file will be saved as ''X.orig''.
		688
		689
		690	__--http-user=__''user''
		691
		692
		693	__--http-passwd=__''password''
		694
		695
		696	Specify the username ''user'' and password
		697	''password'' on an HTTP server. According
		698	to the type of the challenge, Wget will encode them using
		699	either the basic (insecure) or the digest
		700	authentication scheme.
		701
		702
		703	Another way to specify username and password is in the
		704	URL itself. Either method reveals your
		705	password to anyone who bothers to run ps. To
		706	prevent the passwords from being seen, store them in
		707	''.wgetrc'' or ''.netrc'', and make sure to protect
		708	those files from other users with chmod. If the
		709	passwords are really important, do not leave them lying in
		710	those files either---edit the files and delete them after
		711	Wget has started the download.
		712
		713
		714	For more information about security issues with
		715	Wget,
		716
		717
		718	__-C on/off__
		719
		720
		721	__--cache=on/off__
		722
		723
		724	When set to off, disable server-side cache. In this case,
		725	Wget will send the remote server an appropriate directive
		726	(__Pragma: no-cache__) to get the file from the remote
		727	service, rather than returning the cached version. This is
		728	especially useful for retrieving and flushing out-of-date
		729	documents on proxy servers.
		730
		731
		732	Caching is allowed by default.
		733
		734
		735	__--cookies=on/off__
		736
		737
		738	When set to off, disable the use of cookies. Cookies are a
		739	mechanism for maintaining server-side state. The server
		740	sends the client a cookie using the Set-Cookie
		741	header, and the client responds with the same cookie upon
		742	further requests. Since cookies allow the server owners to
		743	keep track of visitors and for sites to exchange this
		744	information, some consider them a breach of privacy. The
		745	default is to use cookies; however, ''storing'' cookies
		746	is not on by default.
		747
		748
		749	__--load-cookies__ ''file''
		750
		751
		752	Load cookies from ''file'' before the first
		753	HTTP retrieval. ''file'' is a textual file
		754	in the format originally used by Netscape's
		755	''cookies.txt'' file.
		756
		757
		758	You will typically use this option when mirroring sites that
		759	require that you be logged in to access some or all of their
		760	content. The login process typically works by the web server
		761	issuing an HTTP cookie upon receiving and
		762	verifying your credentials. The cookie is then resent by the
		763	browser when accessing that part of the site, and so proves
		764	your identity.
		765
		766
		767	Mirroring such a site requires Wget to send the same cookies
		768	your browser sends when communicating with the site. This is
		769	achieved by __--load-cookies__---simply point Wget to the
		770	location of the ''cookies.txt'' file, and it will send
		771	the same cookies your browser would send in the same
		772	situation. Different browsers keep textual cookie files in
		773	different locations:
		774
		775
		776	Netscape 4.x.
		777
		778
		779	The cookies are in
		780	''~/.netscape/cookies.txt''.
		781
		782
		783	Mozilla and Netscape 6.x.
		784
		785
		786	Mozilla's cookie file is also named ''cookies.txt'',
		787	located somewhere under ''~/.mozilla'', in the directory
		788	of your profile. The full path usually ends up looking
		789	somewhat like
		790	''~/.mozilla/default/some-weird-string/cookies.txt''.
		791
		792
		793	Internet Explorer.
		794
		795
		796	You can produce a cookie file Wget can use by using the File
		797	menu, Import and Export, Export Cookies. This has been
		798	tested with Internet Explorer 5; it is not guaranteed to
		799	work with earlier versions.
		800
		801
		802	Other browsers.
		803
		804
		805	If you are using a different browser to create your cookies,
		806	__--load-cookies__ will only work if you can locate or
		807	produce a cookie file in the Netscape format that Wget
		808	expects.
		809
		810
		811	If you cannot use __--load-cookies__, there might still
		812	be an alternative. If your browser supports a ``cookie
		813	manager'', you can use it to view the cookies used when
		814	accessing the site you're mirroring. Write down the name and
		815	value of the cookie, and manually instruct Wget to send
		816	those cookies, bypassing the ``official'' cookie
		817	support:
		818
		819
		820	wget --cookies=off --header
		821
		822
		823	__--save-cookies__ ''file''
		824
		825
		826	Save cookies from ''file'' at the end of session. Cookies
		827	whose expiry time is not specified, or those that have
		828	already expired, are not saved.
		829
		830
		831	__--ignore-length__
		832
		833
		834	Unfortunately, some HTTP servers (
		835	CGI programs, to be more precise) send out
		836	bogus Content-Length headers, which makes Wget go
		837	wild, as it thinks not all the document was retrieved. You
		838	can spot this syndrome if Wget retries getting the same
		839	document again and again, each time claiming that the
		840	(otherwise normal) connection has closed on the very same
		841	byte.
		842
		843
		844	With this option, Wget will ignore the
		845	Content-Length header---as if it never
		846	existed.
		847
		848
		849	__--header=__''additional-header''
		850
		851
		852	Define an ''additional-header'' to be passed to the
		853	HTTP servers. Headers must contain a __:__
		854	preceded by one or more non-blank characters, and must not
		855	contain newlines.
		856
		857
		858	You may define more than one additional header by specifying
		859	__--header__ more than once.
		860
		861
		862	wget --header='Accept-Charset: iso-8859-2' \
		863	--header='Accept-Language: hr' \
		864	http://fly.srk.fer.hr/
		865	Specification of an empty string as the header value will clear all previous user-defined headers.
		866
		867
		868	__--proxy-user=__''user''
		869
		870
		871	__--proxy-passwd=__''password''
		872
		873
		874	Specify the username ''user'' and password
		875	''password'' for authentication on a proxy server. Wget
		876	will encode them using the basic authentication
		877	scheme.
		878
		879
		880	Security considerations similar to those with
		881	__--http-passwd__ pertain here as well.
		882
		883
		884	__--referer=__''url''
		885
		886
		887	Include `Referer: ''url''' header in HTTP
		888	request. Useful for retrieving documents with server-side
		889	processing that assume they are always being retrieved by
		890	interactive web browsers and only come out properly when
		891	Referer is set to one of the pages that point to
		892	them.
		893
		894
		895	__-s__
		896
		897
		898	__--save-headers__
		899
		900
		901	Save the headers sent by the HTTP server to
		902	the file, preceding the actual contents, with an empty line
		903	as the separator.
		904
		905
		906	__-U__ ''agent-string''
		907
		908
		909	__--user-agent=__''agent-string''
		910
		911
		912	Identify as ''agent-string'' to the HTTP
		913	server.
		914
		915
		916	The HTTP protocol allows the clients to
		917	identify themselves using a User-Agent header
		918	field. This enables distinguishing the WWW
		919	software, usually for statistical purposes or for tracing of
		920	protocol violations. Wget normally identifies as
		921	__Wget/__''version'', ''version'' being the current
		922	version number of Wget.
		923
		924
		925	However, some sites have been known to impose the policy of
		926	tailoring the output according to the
		927	User-Agent-supplied information. While conceptually
		928	this is not such a bad idea, it has been abused by servers
		929	denying information to clients other than Mozilla
		930	or Microsoft Internet Explorer. This option allows
		931	you to change the User-Agent line issued by Wget.
		932	Use of this option is discouraged, unless you really know
		933	what you are doing.
		934
		935
		936	__FTP Options__
		937
		938
		939	__-nr__
		940
		941
		942	__--dont-remove-listing__
		943
		944
		945	Don't remove the temporary ''.listing'' files generated
		946	by FTP retrievals. Normally, these files
		947	contain the raw directory listings received from
		948	FTP servers. Not removing them can be useful
		949	for debugging purposes, or when you want to be able to
		950	easily check on the contents of remote server directories
		951	(e.g. to verify that a mirror you're running is
		952	complete).
		953
		954
		955	Note that even though Wget writes to a known filename for
		956	this file, this is not a security hole in the scenario of a
		957	user making ''.listing'' a symbolic link to
		958	''/etc/passwd'' or something and asking root to
		959	run Wget in his or her directory. Depending on the options
		960	used, either Wget will refuse to write to ''.listing'',
		961	making the globbing/recursion/time-stamping operation fail,
		962	or the symbolic link will be deleted and replaced with the
		963	actual ''.listing'' file, or the listing will be written
		964	to a ''.listing.number'' file.
		965
		966
		967	Even though this situation isn't a problem, though,
		968	root should never run Wget in a non-trusted user's
		969	directory. A user could do something as simple as linking
		970	''index.html'' to ''/etc/passwd'' and asking
		971	root to run Wget with __-N__ or __-r__ so the
		972	file will be overwritten.
		973
		974
		975	__-g on/off__
		976
		977
		978	__--glob=on/off__
		979
		980
		981	Turn FTP globbing on or off. Globbing means
		982	you may use the shell-like special characters
		983	(''wildcards''), like __*__, __?__, __[[__ and
		984	__]__ to retrieve more than one file from the same
		985	directory at once, like:
		986
		987
		988	wget ftp://gnjilux.srk.fer.hr/*.msg
		989	By default, globbing will be turned on if the URL contains a globbing character. This option may be used to turn globbing on or off permanently.
		990
		991
		992	You may have to quote the URL to protect it
		993	from being expanded by your shell. Globbing makes Wget look
		994	for a directory listing, which is system-specific. This is
		995	why it currently works only with Unix FTP
		996	servers (and the ones emulating Unix ls
		997	output).
		998
		999
		1000	__--passive-ftp__
		1001
		1002
		1003	Use the ''passive'' FTP retrieval scheme,
		1004	in which the client initiates the data connection. This is
		1005	sometimes required for FTP to work behind
		1006	firewalls.
		1007
		1008
		1009	__--retr-symlinks__
		1010
		1011
		1012	Usually, when retrieving FTP directories
		1013	recursively and a symbolic link is encountered, the
		1014	linked-to file is not downloaded. Instead, a matching
		1015	symbolic link is created on the local filesystem. The
		1016	pointed-to file will not be downloaded unless this recursive
		1017	retrieval would have encountered it separately and
		1018	downloaded it anyway.
		1019
		1020
		1021	When __--retr-symlinks__ is specified, however, symbolic
		1022	links are traversed and the pointed-to files are retrieved.
		1023	At this time, this option does not cause Wget to traverse
		1024	symlinks to directories and recurse through them, but in the
		1025	future it should be enhanced to do this.
		1026
		1027
		1028	Note that when retrieving a file (not a directory) because
		1029	it was specified on the commandline, rather than because it
		1030	was recursed to, this option has no effect. Symbolic links
		1031	are always traversed in this case.
		1032
		1033
		1034	__Recursive Retrieval Options__
		1035
		1036
		1037	__-r__
		1038
		1039
		1040	__--recursive__
		1041
		1042
		1043	Turn on recursive retrieving.
		1044
		1045
		1046	__-l__ ''depth''
		1047
		1048
		1049	__--level=__''depth''
		1050
		1051
		1052	Specify recursion maximum depth level ''depth''. The
		1053	default maximum depth is 5.
		1054
		1055
		1056	__--delete-after__
		1057
		1058
		1059	This option tells Wget to delete every single file it
		1060	downloads, ''after'' having done so. It is useful for
		1061	pre-fetching popular pages through a proxy,
		1062	e.g.:
		1063
		1064
		1065	wget -r -nd --delete-after http://whatever.com/~popular/page/
		1066	The __-r__ option is to retrieve recursively, and __-nd__ to not create directories.
		1067
		1068
		1069	Note that __--delete-after__ deletes files on the local
		1070	machine. It does not issue the __DELE__
		1071	command to remote FTP sites, for instance.
		1072	Also note that when __--delete-after__ is specified,
		1073	__--convert-links__ is ignored, so __.orig__ files are
		1074	simply not created in the first place.
		1075
		1076
		1077	__-k__
		1078
		1079
		1080	__--convert-links__
		1081
		1082
		1083	After the download is complete, convert the links in the
		1084	document to make them suitable for local viewing. This
		1085	affects not only the visible hyperlinks, but any part of the
		1086	document that links to external content, such as embedded
		1087	images, links to style sheets, hyperlinks to non-HTML
		1088	content, etc.
		1089
		1090
		1091	Each link will be changed in one of the two
		1092	ways:
		1093
		1094
		1095	The links to files that have been downloaded by Wget will be
		1096	changed to refer to the file they point to as a relative
		1097	link.
		1098
		1099
		1100	Example: if the downloaded file ''/foo/doc.html'' links
		1101	to ''/bar/img.gif'', also downloaded, then the link in
		1102	''doc.html'' will be modified to point to
		1103	__../bar/img.gif__. This kind of transformation works
		1104	reliably for arbitrary combinations of
		1105	directories.
		1106
		1107
		1108	The links to files that have not been downloaded by Wget
		1109	will be changed to include host name and absolute path of
		1110	the location they point to.
		1111
		1112
		1113	Example: if the downloaded file ''/foo/doc.html'' links
		1114	to ''/bar/img.gif'' (or to ''../bar/img.gif''), then
		1115	the link in ''doc.html'' will be modified to point to
		1116	''http://hostname/bar/img.gif''.
		1117
		1118
		1119	Because of this, local browsing works reliably: if a linked
		1120	file was downloaded, the link will refer to its local name;
		1121	if it was not downloaded, the link will refer to its full
		1122	Internet address rather than presenting a broken link. The
		1123	fact that the former links are converted to relative links
		1124	ensures that you can move the downloaded hierarchy to
		1125	another directory.
		1126
		1127
		1128	Note that only at the end of the download can Wget know
		1129	which links have been downloaded. Because of that, the work
		1130	done by __-k__ will be performed at the end of all the
		1131	downloads.
		1132
		1133
		1134	__-K__
		1135
		1136
		1137	__--backup-converted__
		1138
		1139
		1140	When converting a file, back up the original version with a
		1141	__.orig__ suffix. Affects the behavior of
		1142	__-N__.
		1143
		1144
		1145	__-m__
		1146
		1147
		1148	__--mirror__
		1149
		1150
		1151	Turn on options suitable for mirroring. This option turns on
		1152	recursion and time-stamping, sets infinite recursion depth
		1153	and keeps FTP directory listings. It is
		1154	currently equivalent to __-r -N -l inf
		1155	-nr__.
		1156
		1157
		1158	__-p__
		1159
		1160
		1161	__--page-requisites__
		1162
		1163
		1164	This option causes Wget to download all the files that are
		1165	necessary to properly display a given HTML
		1166	page. This includes such things as inlined images, sounds,
		1167	and referenced stylesheets.
		1168
		1169
		1170	Ordinarily, when downloading a single HTML
		1171	page, any requisite documents that may be needed to display
		1172	it properly are not downloaded. Using __-r__ together
		1173	with __-l__ can help, but since Wget does not ordinarily
		1174	distinguish between external and inlined documents, one is
		1175	generally left with ``leaf documents'' that are missing
		1176	their requisites.
		1177
		1178
		1179	For instance, say document ''1.html'' contains an
		1180	tag referencing ''1.gif'' and an
		1181	tag pointing to external document
		1182	''2.html''. Say that ''2.html'' is similar but that
		1183	its image is ''2.gif'' and it links to ''3.html''. Say
		1184	this continues up to some arbitrarily high
		1185	number.
		1186
		1187
		1188	If one executes the command:
		1189
		1190
		1191	wget -r -l 2 http://I
		1192	then ''1.html'', ''1.gif'', ''2.html'', ''2.gif'', and ''3.html'' will be downloaded. As you can see, ''3.html'' is without its requisite ''3.gif'' because Wget is simply counting the number of hops (up to 2) away from ''1.html'' in order to determine where to stop the recursion. However, with this command:
		1193
		1194
		1195	wget -r -l 2 -p http://I
		1196	all the above files ''and 3.html'''s requisite ''3.gif'' will be downloaded. Similarly,
		1197
		1198
		1199	wget -r -l 1 -p http://I
		1200	will cause ''1.html'', ''1.gif'', ''2.html'', and ''2.gif'' to be downloaded. One might think that:
		1201
		1202
		1203	wget -r -l 0 -p http://I
		1204	would download just ''1.html'' and ''1.gif'', but unfortunately this is not the case, because __-l 0__ is equivalent to __-l inf__---that is, infinite recursion. To download a single HTML page (or a handful of them, all specified on the commandline or in a __-i__ URL input file) and its (or their) requisites, simply leave off __-r__ and __-l__:
		1205
		1206
		1207	wget -p http://I
		1208	Note that Wget will behave as if __-r__ had been specified, but only that single page and its requisites will be downloaded. Links from that page to external documents will not be followed. Actually, to download a single page and all its requisites (even if they exist on separate websites), and make sure the lot displays properly locally, this author likes to use a few options in addition to __-p__:
		1209
		1210
		1211	wget -E -H -k -K -p http://I
		1212	To finish off this topic, it's worth knowing that Wget's idea of an external document link is any URL specified in an tag, an tag, or a tag other than .
		1213
		1214
		1215	__Recursive Accept/Reject Options__
		1216
		1217
		1218	__-A__ ''acclist'' __--accept__
		1219	''acclist''
		1220
		1221
		1222	__-R__ ''rejlist'' __--reject__
		1223	''rejlist''
		1224
		1225
		1226	Specify comma-separated lists of file name suffixes or
		1227	patterns to accept or reject.
		1228
		1229
		1230	__-D__ ''domain-list''
		1231
		1232
		1233	__--domains=__''domain-list''
		1234
		1235
		1236	Set domains to be followed. ''domain-list'' is a
		1237	comma-separated list of domains. Note that it does
		1238	''not'' turn on __-H__.
		1239
		1240
		1241	__--exclude-domains__ ''domain-list''
		1242
		1243
		1244	Specify the domains that are ''not'' to be
		1245	followed..
		1246
		1247
		1248	__--follow-ftp__
		1249
		1250
		1251	Follow FTP links from HTML
		1252	documents. Without this option, Wget will ignore all the
		1253	FTP links.
		1254
		1255
		1256	__--follow-tags=__''list''
		1257
		1258
		1259	Wget has an internal table of HTML tag /
		1260	attribute pairs that it considers when looking for linked
		1261	documents during a recursive retrieval. If a user wants only
		1262	a subset of those tags to be considered, however, he or she
		1263	should be specify such tags in a comma-separated ''list''
		1264	with this option.
		1265
		1266
		1267	__-G__ ''list''
		1268
		1269
		1270	__--ignore-tags=__''list''
		1271
		1272
		1273	This is the opposite of the __--follow-tags__ option. To
		1274	skip certain HTML tags when recursively
		1275	looking for documents to download, specify them in a
		1276	comma-separated ''list''.
		1277
		1278
		1279	In the past, the __-G__ option was the best bet for
		1280	downloading a single page and its requisites, using a
		1281	commandline like:
		1282
		1283
		1284	wget -Ga,area -H -k -K -r http://I
		1285	However, the author of this option came across a page with tags like and came to the realization that __-G__ was not enough. One can't just tell Wget to ignore , because then stylesheets will not be downloaded. Now the best bet for downloading a single page and its requisites is the dedicated __--page-requisites__ option.
		1286
		1287
		1288	__-H__
		1289
		1290
		1291	__--span-hosts__
		1292
		1293
		1294	Enable spanning across hosts when doing recursive
		1295	retrieving.
		1296
		1297
		1298	__-L__
		1299
		1300
		1301	__--relative__
		1302
		1303
		1304	Follow relative links only. Useful for retrieving a specific
		1305	home page without any distractions, not even those from the
		1306	same hosts.
		1307
		1308
		1309	__-I__ ''list''
		1310
		1311
		1312	__--include-directories=__''list''
		1313
		1314
		1315	Specify a comma-separated list of directories you wish to
		1316	follow when downloading Elements of ''list'' may contain
		1317	wildcards.
		1318
		1319
		1320	__-X__ ''list''
		1321
		1322
		1323	__--exclude-directories=__''list''
		1324
		1325
		1326	Specify a comma-separated list of directories you wish to
		1327	exclude from download Elements of ''list'' may contain
		1328	wildcards.
		1329
		1330
		1331	__-np__
		1332
		1333
		1334	__--no-parent__
		1335
		1336
		1337	Do not ever ascend to the parent directory when retrieving
		1338	recursively. This is a useful option, since it guarantees
		1339	that only the files ''below'' a certain hierarchy will be
		1340	downloaded.
		1341	!!EXAMPLES
		1342
		1343
		1344	The examples are divided into three sections loosely based
		1345	on their complexity.
		1346
		1347
		1348	__Simple Usage__
		1349
		1350
		1351	Say you want to download a URL . Just
		1352	type:
		1353
		1354
		1355	wget http://fly.srk.fer.hr/
		1356
		1357
		1358	But what will happen if the connection is slow, and the file
		1359	is lengthy? The connection will probably fail before the
		1360	whole file is retrieved, more than once. In this case, Wget
		1361	will try getting the file until it either gets the whole of
		1362	it, or exceeds the default number of retries (this being
		1363	20). It is easy to change the number of tries to 45, to
		1364	insure that the whole file will arrive safely:
		1365
		1366
		1367	wget --tries=45 http://fly.srk.fer.hr/jpg/flyweb.jpg
		1368
		1369
		1370	Now let's leave Wget to work in the background, and write
		1371	its progress to log file ''log''. It is tiring to type
		1372	__--tries__, so we shall use __-t__.
		1373
		1374
		1375	wget -t 45 -o log http://fly.srk.fer.hr/jpg/flyweb.jpg
		1376	The ampersand at the end of the line makes sure that Wget works in the background. To unlimit the number of retries, use __-t inf__.
		1377
		1378
		1379	The usage of FTP is as simple. Wget will take
		1380	care of login and password.
		1381
		1382
		1383	wget ftp://gnjilux.srk.fer.hr/welcome.msg
		1384
		1385
		1386	If you specify a directory, Wget will retrieve the directory
		1387	listing, parse it and convert it to HTML .
		1388	Try:
		1389
		1390
		1391	wget ftp://prep.ai.mit.edu/pub/gnu/
		1392	links index.html
		1393
		1394
		1395	__Advanced Usage__
		1396
		1397
		1398	You have a file that contains the URLs you want to download?
		1399	Use the __-i__ switch:
		1400
		1401
		1402	wget -i I
		1403	If you specify __-__ as file name, the URLs will be read from standard input.
		1404
		1405
		1406	Create a five levels deep mirror image of the
		1407	GNU web site, with the same directory
		1408	structure the original has, with only one try per document,
		1409	saving the log of the activities to
		1410	''gnulog'':
		1411
		1412
		1413	wget -r http://www.gnu.org/ -o gnulog
		1414
		1415
		1416	The same as the above, but convert the links in the
		1417	HTML files to point to local files, so you
		1418	can view the documents off-line:
		1419
		1420
		1421	wget --convert-links -r http://www.gnu.org/ -o gnulog
		1422
		1423
		1424	Retrieve only one HTML page, but make sure
		1425	that all the elements needed for the page to be displayed,
		1426	such as inline images and external style sheets, are also
		1427	downloaded. Also make sure the downloaded page references
		1428	the downloaded links.
		1429
		1430
		1431	wget -p --convert-links http://www.server.com/dir/page.html
		1432	The HTML page will be saved to ''www.server.com/dir/page.html'', and the images, stylesheets, etc., somewhere under ''www.server.com/'', depending on where they were on the remote server.
		1433
		1434
		1435	The same as the above, but without the
		1436	''www.server.com/'' directory. In fact, I don't want to
		1437	have all those random server directories anyway---just save
		1438	''all'' those files under a ''download/'' subdirectory
		1439	of the current directory.
		1440
		1441
		1442	wget -p --convert-links -nH -nd -Pdownload \
		1443	http://www.server.com/dir/page.html
		1444
		1445
		1446	Retrieve the index.html of __www.lycos.com__, showing the
		1447	original server headers:
		1448
		1449
		1450	wget -S http://www.lycos.com/
		1451
		1452
		1453	Save the server headers with the file, perhaps for
		1454	post-processing.
		1455
		1456
		1457	wget -s http://www.lycos.com/
		1458	more index.html
		1459
		1460
		1461	Retrieve the first two levels of __wuarchive.wustl.edu__,
		1462	saving them to ''/tmp''.
		1463
		1464
		1465	wget -r -l2 -P/tmp ftp://wuarchive.wustl.edu/
		1466
		1467
		1468	You want to download all the GIFs from a directory on an
		1469	HTTP server. You tried __wget
		1470	http://www.server.com/dir/*.gif__, but that didn't work
		1471	because HTTP retrieval does not support
		1472	globbing. In that case, use:
		1473
		1474
		1475	wget -r -l1 --no-parent -A.gif http://www.server.com/dir/
		1476	More verbose, but the effect is the same. __-r -l1__ means to retrieve recursively, with maximum depth of 1. __--no-parent__ means that references to the parent directory are ignored, and __-A.gif__ means to download only the GIF files. __-A ``*.gif''__ would have worked too.
		1477
		1478
		1479	Suppose you were in the middle of downloading, when Wget was
		1480	interrupted. Now you do not want to clobber the files
		1481	already present. It would be:
		1482
		1483
		1484	wget -nc -r http://www.gnu.org/
		1485
		1486
		1487	If you want to encode your own username and password to
		1488	HTTP or FTP , use the
		1489	appropriate URL syntax.
		1490
		1491
		1492	wget ftp://hniksic:mypassword@unix.server.com/.emacs
		1493	Note, however, that this usage is not advisable on multi-user systems because it reveals your password to anyone who looks at the output of ps.
		1494
		1495
		1496	You would like the output documents to go to standard output
		1497	instead of to files?
		1498
		1499
		1500	wget -O - http://jagor.srce.hr/ http://www.srce.hr/
		1501	You can also combine the two options and make pipelines to retrieve the documents from remote hotlists:
		1502
		1503
		1504	wget -O - http://cool.list.com/ wget --force-html -i -
		1505
		1506
		1507	__Very Advanced Usage__
		1508
		1509
		1510	If you wish Wget to keep a mirror of a page (or
		1511	FTP subdirectories), use __--mirror__
		1512	(__-m__), which is the shorthand for __-r -l inf -N__.
		1513	You can put Wget in the crontab file asking it to recheck a
		1514	site each Sunday:
		1515
		1516
		1517	crontab
		1518	0 0 * * 0 wget --mirror http://www.gnu.org/ -o /home/me/weeklog
		1519
		1520
		1521	In addition to the above, you want the links to be converted
		1522	for local viewing. But, after having read this manual, you
		1523	know that link conversion doesn't play well with
		1524	timestamping, so you also want Wget to back up the original
		1525	HTML files before the conversion. Wget
		1526	invocation would look like this:
		1527
		1528
		1529	wget --mirror --convert-links --backup-converted \
		1530	http://www.gnu.org/ -o /home/me/weeklog
		1531
		1532
		1533	But you've also noticed that local viewing doesn't work all
		1534	that well when HTML files are saved under
		1535	extensions other than __.html__, perhaps because they
		1536	were served as ''index.cgi''. So you'd like Wget to
		1537	rename all the files served with content-type
		1538	__text/html__ to ''name.html''.
		1539
		1540
		1541	wget --mirror --convert-links --backup-converted \
		1542	--html-extension -o /home/me/weeklog \
		1543	http://www.gnu.org/
		1544	Or, with less typing:
		1545
		1546
		1547	wget -m -k -K -E http://www.gnu.org/ -o /home/me/weeklog
		1548	!!FILES
		1549
		1550
		1551	__/etc/wgetrc__
		1552
		1553
		1554	Default location of the ''global'' startup
		1555	file.
		1556
		1557
		1558	__.wgetrc__
		1559
		1560
		1561	User startup file.
		1562	!!BUGS
		1563
		1564
		1565	You are welcome to send bug reports about GNU
		1566	Wget to bug-wget@gnu.org
		1567
		1568
		1569	Before actually submitting a bug report, please try to
		1570	follow a few simple guidelines.
		1571
		1572
		1573	1.
		1574
		1575
		1576	Please try to ascertain that the behaviour you see really is
		1577	a bug. If Wget crashes, it's a bug. If Wget does not behave
		1578	as documented, it's a bug. If things work strange, but you
		1579	are not sure about the way they are supposed to work, it
		1580	might well be a bug.
		1581
		1582
		1583	2.
		1584
		1585
		1586	Try to repeat the bug in as simple circumstances as
		1587	possible. E.g. if Wget crashes while downloading __wget
		1588	-rl0 -kKE -t5 -Y0 http://yoyodyne.com -o /tmp/log__, you
		1589	should try to see if the crash is repeatable, and if will
		1590	occur with a simpler set of options. You might even try to
		1591	start the download at the page where the crash occurred to
		1592	see if that page somehow triggered the crash.
		1593
		1594
		1595	Also, while I will probably be interested to know the
		1596	contents of your ''.wgetrc'' file, just dumping it into
		1597	the debug message is probably a bad idea. Instead, you
		1598	should first try to see if the bug repeats with
		1599	''.wgetrc'' moved out of the way. Only if it turns out
		1600	that ''.wgetrc'' settings affect the bug, mail me the
		1601	relevant parts of the file.
		1602
		1603
		1604	3.
		1605
		1606
		1607	Please start Wget with __-d__ option and send the log (or
		1608	the relevant parts of it). If Wget was compiled without
		1609	debug support, recompile it. It is ''much'' easier to
		1610	trace bugs with debug support on.
		1611
		1612
		1613	4.
		1614
		1615
		1616	If Wget has crashed, try to run it in a debugger, e.g.
		1617	gdb `which wget` core and type where to
		1618	get the backtrace.
		1619	!!SEE ALSO
		1620
		1621
		1622	GNU Info entry for ''wget''.
		1623	!!AUTHOR
		1624
		1625
		1626	Originally written by Hrvoje Niksic
		1627	!!COPYRIGHT
		1628
		1629
		1630	Copyright (c) 1996, 1997, 1998, 2000, 2001 Free Software
		1631	Foundation, Inc.
		1632
		1633
		1634	Permission is granted to make and distribute verbatim copies
		1635	of this manual provided the copyright notice and this
		1636	permission notice are preserved on all copies.
		1637
		1638
		1639	Permission is granted to copy, distribute and/or modify this
		1640	document under the terms of the GNU Free
		1641	Documentation License, Version 1.1 or any later version
		1642	published by the Free Software Foundation; with the
		1643	Invariant Sections being `` GNU General
		1644	Public License'' and `` GNU Free
		1645	Documentation License'', with no Front-Cover Texts, and with
		1646	no Back-Cover Texts. A copy of the license is included in
		1647	the section entitled `` GNU Free
		1648	Documentation License''.
		1649	----