[ViewVC] Diff of: cvs/AnyEvent-HTTP/HTTP.pm

Comparing AnyEvent-HTTP/HTTP.pm (file contents):
Revision 1.11 by root, Thu Jun 5 15:34:00 2008 UTC vs.
Revision 1.83 by root, Sun Jan 2 05:02:28 2011 UTC

…		…
3	AnyEvent::HTTP - simple but non-blocking HTTP/HTTPS client	3	AnyEvent::HTTP - simple but non-blocking HTTP/HTTPS client
4		4
5	=head1 SYNOPSIS	5	=head1 SYNOPSIS
6		6
7	use AnyEvent::HTTP;	7	use AnyEvent::HTTP;
		8
		9	http_get "http://www.nethype.de/", sub { print $_[1] };
		10
		11	# ... do something else here
8		12
9	=head1 DESCRIPTION	13	=head1 DESCRIPTION
10		14
11	This module is an L<AnyEvent> user, you need to make sure that you use and	15	This module is an L<AnyEvent> user, you need to make sure that you use and
12	run a supported event loop.	16	run a supported event loop.
…		…
35	package AnyEvent::HTTP;	39	package AnyEvent::HTTP;
36		40
37	use strict;	41	use strict;
38	no warnings;	42	no warnings;
39		43
40	use Carp;	44	use Errno ();
41		45
42	use AnyEvent ();	46	use AnyEvent 5.0 ();
43	use AnyEvent::Util ();	47	use AnyEvent::Util ();
44	use AnyEvent::Socket ();
45	use AnyEvent::Handle ();	48	use AnyEvent::Handle ();
46		49
47	use base Exporter::;	50	use base Exporter::;
48		51
49	our $VERSION = '1.0';	52	our $VERSION = '1.5';
50		53
51	our @EXPORT = qw(http_get http_request);	54	our @EXPORT = qw(http_get http_post http_head http_request);
52		55
53	our $USERAGENT = "Mozilla/5.0 (compatible; AnyEvent::HTTP/$VERSION; +http://software.schmorp.de/pkg/AnyEvent)";	56	our $USERAGENT = "Mozilla/5.0 (compatible; U; AnyEvent-HTTP/$VERSION; +http://software.schmorp.de/pkg/AnyEvent)";
54	our $MAX_RECURSE = 10;	57	our $MAX_RECURSE = 10;
55	our $MAX_PERSISTENT = 8;	58	our $MAX_PERSISTENT = 8;
56	our $PERSISTENT_TIMEOUT = 2;	59	our $PERSISTENT_TIMEOUT = 2;
57	our $TIMEOUT = 300;	60	our $TIMEOUT = 300;
58		61
59	# changing these is evil	62	# changing these is evil
60	our $MAX_PERSISTENT_PER_HOST = 2;	63	our $MAX_PERSISTENT_PER_HOST = 0;
61	our $MAX_PER_HOST = 4;	64	our $MAX_PER_HOST = 4;
62		65
63	our $PROXY;	66	our $PROXY;
		67	our $ACTIVE = 0;
64		68
65	my %KA_COUNT; # number of open keep-alive connections per host	69	my %KA_COUNT; # number of open keep-alive connections per host
66	my %CO_SLOT; # number of open connections, and wait queue, per host	70	my %CO_SLOT; # number of open connections, and wait queue, per host
67		71
68	=item http_get $url, key => value..., $cb->($data, $headers)	72	=item http_get $url, key => value..., $cb->($data, $headers)
69		73
70	Executes an HTTP-GET request. See the http_request function for details on	74	Executes an HTTP-GET request. See the http_request function for details on
71	additional parameters.	75	additional parameters and the return value.
72		76
73	=item http_head $url, key => value..., $cb->($data, $headers)	77	=item http_head $url, key => value..., $cb->($data, $headers)
74		78
75	Executes an HTTP-HEAD request. See the http_request function for details on	79	Executes an HTTP-HEAD request. See the http_request function for details
76	additional parameters.	80	on additional parameters and the return value.
77		81
78	=item http_post $url, $body, key => value..., $cb->($data, $headers)	82	=item http_post $url, $body, key => value..., $cb->($data, $headers)
79		83
80	Executes an HTTP-POST request with a request body of C<$bod>. See the	84	Executes an HTTP-POST request with a request body of C<$body>. See the
81	http_request function for details on additional parameters.	85	http_request function for details on additional parameters and the return
		86	value.
82		87
83	=item http_request $method => $url, key => value..., $cb->($data, $headers)	88	=item http_request $method => $url, key => value..., $cb->($data, $headers)
84		89
85	Executes a HTTP request of type C<$method> (e.g. C<GET>, C<POST>). The URL	90	Executes a HTTP request of type C<$method> (e.g. C<GET>, C<POST>). The URL
86	must be an absolute http or https URL.	91	must be an absolute http or https URL.
87		92
		93	When called in void context, nothing is returned. In other contexts,
		94	C<http_request> returns a "cancellation guard" - you have to keep the
		95	object at least alive until the callback get called. If the object gets
		96	destroyed before the callback is called, the request will be cancelled.
		97
88	The callback will be called with the response data as first argument	98	The callback will be called with the response body data as first argument
89	(or C<undef> if it wasn't available due to errors), and a hash-ref with	99	(or C<undef> if an error occured), and a hash-ref with response headers
90	response headers as second argument.	100	(and trailers) as second argument.
91		101
92	All the headers in that hash are lowercased. In addition to the response	102	All the headers in that hash are lowercased. In addition to the response
93	headers, the three "pseudo-headers" C<HTTPVersion>, C<Status> and	103	headers, the "pseudo-headers" (uppercase to avoid clashing with possible
94	C<Reason> contain the three parts of the HTTP Status-Line of the same	104	response headers) C<HTTPVersion>, C<Status> and C<Reason> contain the
		105	three parts of the HTTP Status-Line of the same name. If an error occurs
		106	during the body phase of a request, then the original C<Status> and
		107	C<Reason> values from the header are available as C<OrigStatus> and
		108	C<OrigReason>.
		109
		110	The pseudo-header C<URL> contains the actual URL (which can differ from
		111	the requested URL when following redirects - for example, you might get
		112	an error that your URL scheme is not supported even though your URL is a
		113	valid http URL because it redirected to an ftp URL, in which case you can
		114	look at the URL pseudo header).
		115
		116	The pseudo-header C<Redirect> only exists when the request was a result
		117	of an internal redirect. In that case it is an array reference with
		118	the C<($data, $headers)> from the redirect response. Note that this
		119	response could in turn be the result of a redirect itself, and C<<
		120	$headers->{Redirect}[1]{Redirect} >> will then contain the original
		121	response, and so on.
		122
95	name. If the server sends a header multiple lines, then their contents	123	If the server sends a header multiple times, then their contents will be
96	will be joined together with C<\x00>.	124	joined together with a comma (C<,>), as per the HTTP spec.
97		125
98	If an internal error occurs, such as not being able to resolve a hostname,	126	If an internal error occurs, such as not being able to resolve a hostname,
99	then C<$data> will be C<undef>, C<< $headers->{Status} >> will be C<599>	127	then C<$data> will be C<undef>, C<< $headers->{Status} >> will be
100	and the C<Reason> pseudo-header will contain an error message.	128	C<590>-C<599> and the C<Reason> pseudo-header will contain an error
		129	message. Currently the following status codes are used:
		130
		131	=over 4
		132
		133	=item 595 - errors during connection etsbalishment, proxy handshake.
		134
		135	=item 596 - errors during TLS negotiation, request sending and header processing.
		136
		137	=item 597 - errors during body receiving or processing.
		138
		139	=item 598 - user aborted request via C<on_header> or C<on_body>.
		140
		141	=item 599 - other, usually nonretryable, errors (garbled URL etc.).
		142
		143	=back
101		144
102	A typical callback might look like this:	145	A typical callback might look like this:
103		146
104	sub {	147	sub {
105	my ($body, $hdr) = @_;	148	my ($body, $hdr) = @_;
…		…
121	Whether to recurse requests or not, e.g. on redirects, authentication	164	Whether to recurse requests or not, e.g. on redirects, authentication
122	retries and so on, and how often to do so.	165	retries and so on, and how often to do so.
123		166
124	=item headers => hashref	167	=item headers => hashref
125		168
126	The request headers to use.	169	The request headers to use. Currently, C<http_request> may provide its own
		170	C<Host:>, C<Content-Length:>, C<Connection:> and C<Cookie:> headers and
		171	will provide defaults at least for C<TE:>, C<Referer:> and C<User-Agent:>
		172	(this can be suppressed by using C<undef> for these headers in which case
		173	they won't be sent at all).
127		174
128	=item timeout => $seconds	175	=item timeout => $seconds
129		176
130	The time-out to use for various stages - each connect attempt will reset	177	The time-out to use for various stages - each connect attempt will reset
131	the timeout, as will read or write activity. Default timeout is 5 minutes.	178	the timeout, as will read or write activity, i.e. this is not an overall
		179	timeout.
		180
		181	Default timeout is 5 minutes.
132		182
133	=item proxy => [$host, $port[, $scheme]] or undef	183	=item proxy => [$host, $port[, $scheme]] or undef
134		184
135	Use the given http proxy for all requests. If not specified, then the	185	Use the given http proxy for all requests. If not specified, then the
136	default proxy (as specified by C<$ENV{http_proxy}>) is used.	186	default proxy (as specified by C<$ENV{http_proxy}>) is used.
137		187
138	C<$scheme> must be either missing or C<http> for HTTP, or C<https> for	188	C<$scheme> must be either missing, C<http> for HTTP or C<https> for
139	HTTPS.	189	HTTPS.
140		190
141	=item body => $string	191	=item body => $string
142		192
143	The request body, usually empty. Will be-sent as-is (future versions of	193	The request body, usually empty. Will be sent as-is (future versions of
144	this module might offer more options).	194	this module might offer more options).
145		195
146	=item cookie_jar => $hash_ref	196	=item cookie_jar => $hash_ref
147		197
148	Passing this parameter enables (simplified) cookie-processing, loosely	198	Passing this parameter enables (simplified) cookie-processing, loosely
149	based on the original netscape specification.	199	based on the original netscape specification.
150		200
151	The C<$hash_ref> must be an (initially empty) hash reference which will	201	The C<$hash_ref> must be an (initially empty) hash reference which
152	get updated automatically. It is possible to save the cookie_jar to	202	will get updated automatically. It is possible to save the cookie jar
153	persistent storage with something like JSON or Storable, but this is not	203	to persistent storage with something like JSON or Storable - see the
154	recommended, as expire times are currently being ignored.	204	C<AnyEvent::HTTP::cookie_jar_expire> function if you wish to remove
		205	expired or session-only cookies, and also for documentation on the format
		206	of the cookie jar.
155		207
156	Note that this cookie implementation is not of very high quality, nor	208	Note that this cookie implementation is not meant to be complete. If
157	meant to be complete. If you want complete cookie management you have to	209	you want complete cookie management you have to do that on your
158	do that on your own. C<cookie_jar> is meant as a quick fix to get some	210	own. C<cookie_jar> is meant as a quick fix to get most cookie-using sites
159	cookie-using sites working. Cookies are a privacy disaster, do not use	211	working. Cookies are a privacy disaster, do not use them unless required
160	them unless required to.	212	to.
		213
		214	When cookie processing is enabled, the C<Cookie:> and C<Set-Cookie:>
		215	headers will be set and handled by this module, otherwise they will be
		216	left untouched.
		217
		218	=item tls_ctx => $scheme \| $tls_ctx
		219
		220	Specifies the AnyEvent::TLS context to be used for https connections. This
		221	parameter follows the same rules as the C<tls_ctx> parameter to
		222	L<AnyEvent::Handle>, but additionally, the two strings C<low> or
		223	C<high> can be specified, which give you a predefined low-security (no
		224	verification, highest compatibility) and high-security (CA and common-name
		225	verification) TLS context.
		226
		227	The default for this option is C<low>, which could be interpreted as "give
		228	me the page, no matter what".
		229
		230	=item on_prepare => $callback->($fh)
		231
		232	In rare cases you need to "tune" the socket before it is used to
		233	connect (for exmaple, to bind it on a given IP address). This parameter
		234	overrides the prepare callback passed to C<AnyEvent::Socket::tcp_connect>
		235	and behaves exactly the same way (e.g. it has to provide a
		236	timeout). See the description for the C<$prepare_cb> argument of
		237	C<AnyEvent::Socket::tcp_connect> for details.
		238
		239	=item tcp_connect => $callback->($host, $service, $connect_cb, $prepare_cb)
		240
		241	In even rarer cases you want total control over how AnyEvent::HTTP
		242	establishes connections. Normally it uses L<AnyEvent::Socket::tcp_connect>
		243	to do this, but you can provide your own C<tcp_connect> function -
		244	obviously, it has to follow the same calling conventions, except that it
		245	may always return a connection guard object.
		246
		247	There are probably lots of weird uses for this function, starting from
		248	tracing the hosts C<http_request> actually tries to connect, to (inexact
		249	but fast) host => IP address caching or even socks protocol support.
		250
		251	=item on_header => $callback->($headers)
		252
		253	When specified, this callback will be called with the header hash as soon
		254	as headers have been successfully received from the remote server (not on
		255	locally-generated errors).
		256
		257	It has to return either true (in which case AnyEvent::HTTP will continue),
		258	or false, in which case AnyEvent::HTTP will cancel the download (and call
		259	the finish callback with an error code of C<598>).
		260
		261	This callback is useful, among other things, to quickly reject unwanted
		262	content, which, if it is supposed to be rare, can be faster than first
		263	doing a C<HEAD> request.
		264
		265	The downside is that cancelling the request makes it impossible to re-use
		266	the connection. Also, the C<on_header> callback will not receive any
		267	trailer (headers sent after the response body).
		268
		269	Example: cancel the request unless the content-type is "text/html".
		270
		271	on_header => sub {
		272	$_[0]{"content-type"} =~ /^text\/html\s*(?:;\|$)/
		273	},
		274
		275	=item on_body => $callback->($partial_body, $headers)
		276
		277	When specified, all body data will be passed to this callback instead of
		278	to the completion callback. The completion callback will get the empty
		279	string instead of the body data.
		280
		281	It has to return either true (in which case AnyEvent::HTTP will continue),
		282	or false, in which case AnyEvent::HTTP will cancel the download (and call
		283	the completion callback with an error code of C<598>).
		284
		285	The downside to cancelling the request is that it makes it impossible to
		286	re-use the connection.
		287
		288	This callback is useful when the data is too large to be held in memory
		289	(so the callback writes it to a file) or when only some information should
		290	be extracted, or when the body should be processed incrementally.
		291
		292	It is usually preferred over doing your own body handling via
		293	C<want_body_handle>, but in case of streaming APIs, where HTTP is
		294	only used to create a connection, C<want_body_handle> is the better
		295	alternative, as it allows you to install your own event handler, reducing
		296	resource usage.
		297
		298	=item want_body_handle => $enable
		299
		300	When enabled (default is disabled), the behaviour of AnyEvent::HTTP
		301	changes considerably: after parsing the headers, and instead of
		302	downloading the body (if any), the completion callback will be
		303	called. Instead of the C<$body> argument containing the body data, the
		304	callback will receive the L<AnyEvent::Handle> object associated with the
		305	connection. In error cases, C<undef> will be passed. When there is no body
		306	(e.g. status C<304>), the empty string will be passed.
		307
		308	The handle object might or might not be in TLS mode, might be connected to
		309	a proxy, be a persistent connection etc., and configured in unspecified
		310	ways. The user is responsible for this handle (it will not be used by this
		311	module anymore).
		312
		313	This is useful with some push-type services, where, after the initial
		314	headers, an interactive protocol is used (typical example would be the
		315	push-style twitter API which starts a JSON/XML stream).
		316
		317	If you think you need this, first have a look at C<on_body>, to see if
		318	that doesn't solve your problem in a better way.
161		319
162	=back	320	=back
163		321
164	Example: make a simple HTTP GET request for http://www.nethype.de/	322	Example: do a simple HTTP GET request for http://www.nethype.de/ and print
		323	the response body.
165		324
166	http_request GET => "http://www.nethype.de/", sub {	325	http_request GET => "http://www.nethype.de/", sub {
167	my ($body, $hdr) = @_;	326	my ($body, $hdr) = @_;
168	print "$body\n";	327	print "$body\n";
169	};	328	};
170		329
171	Example: make a HTTP HEAD request on https://www.google.com/, use a	330	Example: do a HTTP HEAD request on https://www.google.com/, use a
172	timeout of 30 seconds.	331	timeout of 30 seconds.
173		332
174	http_request	333	http_request
175	GET => "https://www.google.com",	334	GET => "https://www.google.com",
176	timeout => 30,	335	timeout => 30,
…		…
179	use Data::Dumper;	338	use Data::Dumper;
180	print Dumper $hdr;	339	print Dumper $hdr;
181	}	340	}
182	;	341	;
183		342
		343	Example: do another simple HTTP GET request, but immediately try to
		344	cancel it.
		345
		346	my $request = http_request GET => "http://www.nethype.de/", sub {
		347	my ($body, $hdr) = @_;
		348	print "$body\n";
		349	};
		350
		351	undef $request;
		352
184	=cut	353	=cut
185		354
		355	sub _slot_schedule;
186	sub _slot_schedule($) {	356	sub _slot_schedule($) {
187	my $host = shift;	357	my $host = shift;
188		358
189	while ($CO_SLOT{$host}[0] < $MAX_PER_HOST) {	359	while ($CO_SLOT{$host}[0] < $MAX_PER_HOST) {
190	if (my $cb = shift @{ $CO_SLOT{$host}[1] }) {	360	if (my $cb = shift @{ $CO_SLOT{$host}[1] }) {
191	# somebody wnats that slot	361	# somebody wants that slot
192	++$CO_SLOT{$host}[0];	362	++$CO_SLOT{$host}[0];
		363	++$ACTIVE;
193		364
194	$cb->(AnyEvent::Util::guard {	365	$cb->(AnyEvent::Util::guard {
		366	--$ACTIVE;
195	--$CO_SLOT{$host}[0];	367	--$CO_SLOT{$host}[0];
196	_slot_schedule $host;	368	_slot_schedule $host;
197	});	369	});
198	} else {	370	} else {
199	# nobody wants the slot, maybe we can forget about it	371	# nobody wants the slot, maybe we can forget about it
200	delete $CO_SLOT{$host} unless $CO_SLOT{$host}[0];	372	delete $CO_SLOT{$host} unless $CO_SLOT{$host}[0];
201	warn "$host deleted" unless $CO_SLOT{$host}[0];#d#
202	last;	373	last;
203	}	374	}
204	}	375	}
205	}	376	}
206		377
…		…
209	push @{ $CO_SLOT{$_[0]}[1] }, $_[1];	380	push @{ $CO_SLOT{$_[0]}[1] }, $_[1];
210		381
211	_slot_schedule $_[0];	382	_slot_schedule $_[0];
212	}	383	}
213		384
		385	#############################################################################
		386
		387	# expire cookies
		388	sub cookie_jar_expire($;$) {
		389	my ($jar, $session_end) = @_;
		390
		391	%$jar = () if $jar->{version} != 1;
		392
		393	my $anow = AE::now;
		394
		395	while (my ($chost, $paths) = each %$jar) {
		396	next unless ref $paths;
		397
		398	while (my ($cpath, $cookies) = each %$paths) {
		399	while (my ($cookie, $kv) = each %$cookies) {
		400	if (exists $kv->{_expires}) {
		401	delete $cookies->{$cookie}
		402	if $anow > $kv->{_expires};
		403	} elsif ($session_end) {
		404	delete $cookies->{$cookie};
		405	}
		406	}
		407
		408	delete $paths->{$cpath}
		409	unless %$cookies;
		410	}
		411
		412	delete $jar->{$chost}
		413	unless %$paths;
		414	}
		415	}
		416
		417	# extract cookies from jar
		418	sub cookie_jar_extract($$$$) {
		419	my ($jar, $uscheme, $uhost, $upath) = @_;
		420
		421	$uhost = lc $uhost;
		422
		423	%$jar = () if $jar->{version} != 1;
		424
		425	my @cookies;
		426
		427	while (my ($chost, $paths) = each %$jar) {
		428	next unless ref $paths;
		429
		430	if ($chost =~ /^\./) {
		431	next unless $chost eq substr $uhost, -length $chost;
		432	} elsif ($chost =~ /\./) {
		433	next unless $chost eq $uhost;
		434	} else {
		435	next;
		436	}
		437
		438	while (my ($cpath, $cookies) = each %$paths) {
		439	next unless $cpath eq substr $upath, 0, length $cpath;
		440
		441	while (my ($cookie, $kv) = each %$cookies) {
		442	next if $uscheme ne "https" && exists $kv->{secure};
		443
		444	if (exists $kv->{_expires} and AE::now > $kv->{_expires}) {
		445	delete $cookies->{$cookie};
		446	next;
		447	}
		448
		449	my $value = $kv->{value};
		450
		451	if ($value =~ /[=;,[:space:]]/) {
		452	$value =~ s/([\\"])/\\$1/g;
		453	$value = "\"$value\"";
		454	}
		455
		456	push @cookies, "$cookie=$value";
		457	}
		458	}
		459	}
		460
		461	\@cookies
		462	}
		463
		464	# parse set_cookie header into jar
		465	sub cookie_jar_set_cookie($$$$) {
		466	my ($jar, $set_cookie, $uhost, $date) = @_;
		467
		468	my $anow = int AE::now;
		469	my $snow; # server-now
		470
		471	for ($set_cookie) {
		472	# parse NAME=VALUE
		473	my @kv;
		474
		475	# expires is not http-compliant in the original cookie-spec,
		476	# we support the official date format and some extensions
		477	while (
		478	m{
		479	\G\s*
		480	(?:
		481	expires \s=\s ([A-Z][a-z][a-z]+,\ [^,;]+)
		482	\| ([^=;,[:space:]]+) (?: \s=\s (?: "((?:[^\\"]+\|\\.))" \| ([^=;,[:space:]]) ) )?
		483	)
		484	}gcxsi
		485	) {
		486	my $name = $2;
		487	my $value = $4;
		488
		489	if (defined $1) {
		490	# expires
		491	$name = "expires";
		492	$value = $1;
		493	} elsif (defined $3) {
		494	# quoted
		495	$value = $3;
		496	$value =~ s/\\(.)/$1/gs;
		497	}
		498
		499	push @kv, lc $name, $value;
		500
		501	last unless /\G\s*;/gc;
		502	}
		503
		504	last unless @kv;
		505
		506	my $name = shift @kv;
		507	my %kv = (value => shift @kv, @kv);
		508
		509	if (exists $kv{"max-age"}) {
		510	$kv{_expires} = $anow + delete $kv{"max-age"};
		511	} elsif (exists $kv{expires}) {
		512	$snow \|\|= parse_date ($date) \|\| $anow;
		513	$kv{_expires} = $anow + (parse_date (delete $kv{expires}) - $snow);
		514	} else {
		515	delete $kv{_expires};
		516	}
		517
		518	my $cdom;
		519	my $cpath = (delete $kv{path}) \|\| "/";
		520
		521	if (exists $kv{domain}) {
		522	$cdom = delete $kv{domain};
		523
		524	$cdom =~ s/^\.?/./; # make sure it starts with a "."
		525
		526	next if $cdom =~ /\.$/;
		527
		528	# this is not rfc-like and not netscape-like. go figure.
		529	my $ndots = $cdom =~ y/.//;
		530	next if $ndots < ($cdom =~ /\.[^.][^.]\.[^.][^.]$/ ? 3 : 2);
		531	} else {
		532	$cdom = $uhost;
		533	}
		534
		535	# store it
		536	$jar->{version} = 1;
		537	$jar->{lc $cdom}{$cpath}{$name} = \%kv;
		538
		539	redo if /\G\s*,/gc;
		540	}
		541	}
		542
		543	# continue to parse $_ for headers and place them into the arg
		544	sub parse_hdr() {
		545	my %hdr;
		546
		547	# things seen, not parsed:
		548	# p3pP="NON CUR OTPi OUR NOR UNI"
		549
		550	$hdr{lc $1} .= ",$2"
		551	while /\G
		552	([^:\000-\037]*):
		553	[\011\040]*
		554	((?: [^\012]+ \| \012[\011\040] )*)
		555	\012
		556	/gxc;
		557
		558	/\G$/
		559	or return;
		560
		561	# remove the "," prefix we added to all headers above
		562	substr $_, 0, 1, ""
		563	for values %hdr;
		564
		565	\%hdr
		566	}
		567
		568	our $qr_nlnl = qr{(?<![^\012])\015?\012};
		569
		570	our $TLS_CTX_LOW = { cache => 1, sslv2 => 1 };
		571	our $TLS_CTX_HIGH = { cache => 1, verify => 1, verify_peername => "https" };
		572
214	sub http_request($$$;@) {	573	sub http_request($$@) {
215	my $cb = pop;	574	my $cb = pop;
216	my ($method, $url, %arg) = @_;	575	my ($method, $url, %arg) = @_;
217		576
218	my %hdr;	577	my %hdr;
		578
		579	$arg{tls_ctx} = $TLS_CTX_LOW if $arg{tls_ctx} eq "low" \|\| !exists $arg{tls_ctx};
		580	$arg{tls_ctx} = $TLS_CTX_HIGH if $arg{tls_ctx} eq "high";
219		581
220	$method = uc $method;	582	$method = uc $method;
221		583
222	if (my $hdr = $arg{headers}) {	584	if (my $hdr = $arg{headers}) {
223	while (my ($k, $v) = each %$hdr) {	585	while (my ($k, $v) = each %$hdr) {
224	$hdr{lc $k} = $v;	586	$hdr{lc $k} = $v;
225	}	587	}
226	}	588	}
227		589
		590	# pseudo headers for all subsequent responses
		591	my @pseudo = (URL => $url);
		592	push @pseudo, Redirect => delete $arg{Redirect} if exists $arg{Redirect};
		593
228	my $recurse = exists $arg{recurse} ? $arg{recurse} : $MAX_RECURSE;	594	my $recurse = exists $arg{recurse} ? delete $arg{recurse} : $MAX_RECURSE;
229		595
230	return $cb->(undef, { Status => 599, Reason => "recursion limit reached" })	596	return $cb->(undef, { @pseudo, Status => 599, Reason => "Too many redirections" })
231	if $recurse < 0;	597	if $recurse < 0;
232		598
233	my $proxy = $arg{proxy} \|\| $PROXY;	599	my $proxy = $arg{proxy} \|\| $PROXY;
234	my $timeout = $arg{timeout} \|\| $TIMEOUT;	600	my $timeout = $arg{timeout} \|\| $TIMEOUT;
235		601
236	$hdr{"user-agent"} \|\|= $USERAGENT;
237
238	my ($scheme, $authority, $upath, $query, $fragment) =	602	my ($uscheme, $uauthority, $upath, $query, $fragment) =
239	$url =~ m\|(?:([^:/?#]+):)?(?://([^/?#]))?([^?#])(?:\?([^#]))?(?:#(.))?\|;	603	$url =~ m\|(?:([^:/?#]+):)?(?://([^/?#]))?([^?#])(?:(\?[^#]))?(?:#(.))?\|;
240		604
241	$scheme = lc $scheme;	605	$uscheme = lc $uscheme;
242		606
243	my $uport = $scheme eq "http" ? 80	607	my $uport = $uscheme eq "http" ? 80
244	: $scheme eq "https" ? 443	608	: $uscheme eq "https" ? 443
245	: return $cb->(undef, { Status => 599, Reason => "only http and https URL schemes supported" });	609	: return $cb->(undef, { @pseudo, Status => 599, Reason => "Only http and https URL schemes supported" });
246		610
247	$authority =~ /^(?: .*\@ )? ([^\@:]+) (?: : (\d+) )?$/x	611	$uauthority =~ /^(?: .*\@ )? ([^\@:]+) (?: : (\d+) )?$/x
248	or return $cb->(undef, { Status => 599, Reason => "unparsable URL" });	612	or return $cb->(undef, { @pseudo, Status => 599, Reason => "Unparsable URL" });
249		613
250	my $uhost = $1;	614	my $uhost = $1;
251	$uport = $2 if defined $2;	615	$uport = $2 if defined $2;
252		616
		617	$hdr{host} = defined $2 ? "$uhost:$2" : "$uhost"
		618	unless exists $hdr{host};
		619
253	$uhost =~ s/^\[(.*)\]$/$1/;	620	$uhost =~ s/^\[(.*)\]$/$1/;
254	$upath .= "?$query" if length $query;	621	$upath .= $query if length $query;
255		622
256	$upath =~ s%^/?%/%;	623	$upath =~ s%^/?%/%;
257		624
258	# cookie processing	625	# cookie processing
259	if (my $jar = $arg{cookie_jar}) {	626	if (my $jar = $arg{cookie_jar}) {
260	%$jar = () if $jar->{version} < 1;	627	my $cookies = cookie_jar_extract $jar, $uscheme, $uhost, $upath;
261		628
262	my @cookie;
263
264	while (my ($chost, $v) = each %$jar) {
265	next unless $chost eq substr $uhost, -length $chost;
266	next unless $chost =~ /^\./;
267
268	while (my ($cpath, $v) = each %$v) {
269	next unless $cpath eq substr $upath, 0, length $cpath;
270
271	while (my ($k, $v) = each %$v) {
272	next if $scheme ne "https" && exists $v->{secure};
273	push @cookie, "$k=$v->{value}";
274	}
275	}
276	}
277
278	$hdr{cookie} = join "; ", @cookie	629	$hdr{cookie} = join "; ", @$cookies
279	if @cookie;	630	if @$cookies;
280	}	631	}
281		632
282	my ($rhost, $rport, $rpath); # request host, port, path	633	my ($rhost, $rport, $rscheme, $rpath); # request host, port, path
283		634
284	if ($proxy) {	635	if ($proxy) {
285	($rhost, $rport, $scheme) = @$proxy;	636	($rpath, $rhost, $rport, $rscheme) = ($url, @$proxy);
286	$rpath = $url;	637
		638	$rscheme = "http" unless defined $rscheme;
		639
		640	# don't support https requests over https-proxy transport,
		641	# can't be done with tls as spec'ed, unless you double-encrypt.
		642	$rscheme = "http" if $uscheme eq "https" && $rscheme eq "https";
287	} else {	643	} else {
288	($rhost, $rport, $rpath) = ($uhost, $uport, $upath);	644	($rhost, $rport, $rscheme, $rpath) = ($uhost, $uport, $uscheme, $upath);
289	$hdr{host} = $uhost;
290	}	645	}
291		646
		647	# leave out fragment and query string, just a heuristic
		648	$hdr{referer} = "$uscheme://$uauthority$upath" unless exists $hdr{referer};
		649	$hdr{"user-agent"} = $USERAGENT unless exists $hdr{"user-agent"};
		650
292	$hdr{"content-length"} = length $arg{body};	651	$hdr{"content-length"} = length $arg{body}
		652	if length $arg{body} \|\| $method ne "GET";
		653
		654	$hdr{connection} = "close TE"; #1.1
		655	$hdr{te} = "trailers" unless exists $hdr{te}; #1.1
293		656
294	my %state = (connect_guard => 1);	657	my %state = (connect_guard => 1);
295		658
296	_get_slot $uhost, sub {	659	_get_slot $uhost, sub {
297	$state{slot_guard} = shift;	660	$state{slot_guard} = shift;
298		661
299	return unless $state{connect_guard};	662	return unless $state{connect_guard};
300		663
301	$state{connect_guard} = AnyEvent::Socket::tcp_connect $rhost, $rport, sub {	664	my $ae_error = 595; # connecting
302	$state{fh} = shift
303	or return $cb->(undef, { Status => 599, Reason => "$!" });
304		665
305	delete $state{connect_guard}; # reduce memory usage, save a tree	666	# handle actual, non-tunneled, request
		667	my $handle_actual_request = sub {
		668	$ae_error = 596; # request phase
306		669
307	# get handle	670	$state{handle}->starttls ("connect") if $uscheme eq "https" && !exists $state{handle}{tls};
308	$state{handle} = new AnyEvent::Handle
309	fh => $state{fh},
310	($scheme eq "https" ? (tls => "connect") : ());
311
312	# limit the number of persistent connections
313	if ($KA_COUNT{$_[1]} < $MAX_PERSISTENT_PER_HOST) {
314	++$KA_COUNT{$_[1]};
315	$state{handle}{ka_count_guard} = AnyEvent::Util::guard { --$KA_COUNT{$_[1]} };
316	$hdr{connection} = "keep-alive";
317	delete $hdr{connection}; # keep-alive not yet supported
318	} else {
319	delete $hdr{connection};
320	}
321
322	# (re-)configure handle
323	$state{handle}->timeout ($timeout);
324	$state{handle}->on_error (sub {
325	%state = ();
326	$cb->(undef, { Status => 599, Reason => "$!" });
327	});
328	$state{handle}->on_eof (sub {
329	%state = ();
330	$cb->(undef, { Status => 599, Reason => "unexpected end-of-file" });
331	});
332		671
333	# send request	672	# send request
334	$state{handle}->push_write (	673	$state{handle}->push_write (
335	"$method $rpath HTTP/1.0\015\012"	674	"$method $rpath HTTP/1.1\015\012"
336	. (join "", map "$_: $hdr{$_}\015\012", keys %hdr)	675	. (join "", map "\u$_: $hdr{$_}\015\012", grep defined $hdr{$_}, keys %hdr)
337	. "\015\012"	676	. "\015\012"
338	. (delete $arg{body})	677	. (delete $arg{body})
339	);	678	);
340		679
		680	# return if error occured during push_write()
		681	return unless %state;
		682
341	%hdr = (); # reduce memory usage, save a kitten	683	%hdr = (); # reduce memory usage, save a kitten, also make it possible to re-use
342		684
343	# status line	685	# status line and headers
344	$state{handle}->push_read (line => qr/\015?\012/, sub {	686	$state{read_response} = sub {
		687	for ("$_[1]") {
		688	y/\015//d; # weed out any \015, as they show up in the weirdest of places.
		689
345	$_[1] =~ /^HTTP\/([0-9\.]+) \s+ ([0-9]{3}) \s+ ([^\015\012]+)/ix	690	/^HTTP\/0([0-9\.]+) \s+ ([0-9]{3}) (?: \s+ ([^\012]) )? \012/gxci
346	or return (%state = (), $cb->(undef, { Status => 599, Reason => "invalid server response ($_[1])" }));	691	or return (%state = (), $cb->(undef, { @pseudo, Status => 599, Reason => "Invalid server response" }));
347		692
348	my %hdr = ( # response headers	693	# 100 Continue handling
		694	# should not happen as we don't send expect: 100-continue,
		695	# but we handle it just in case.
		696	# since we send the request body regardless, if we get an error
		697	# we are out of-sync, which we currently do NOT handle correctly.
		698	return $state{handle}->push_read (line => $qr_nlnl, $state{read_response})
		699	if $2 eq 100;
		700
		701	push @pseudo,
349	HTTPVersion => "\x00$1",	702	HTTPVersion => $1,
350	Status => "\x00$2",	703	Status => $2,
351	Reason => "\x00$3",	704	Reason => $3,
352	);	705	;
353		706
354	# headers, could be optimized a bit	707	my $hdr = parse_hdr
355	$state{handle}->unshift_read (line => qr/\015?\012\015?\012/, sub {
356	for ("$_[1]\012") {
357	# we support spaces in field names, as lotus domino
358	# creates them.
359	$hdr{lc $1} .= "\x00$2"
360	while /\G
361	([^:\000-\037]+):
362	[\011\040]*
363	((?: [^\015\012]+ \| \015?\012[\011\040] )*)
364	\015?\012
365	/gxc;
366
367	/\G$/
368	or return (%state = (), $cb->(undef, { Status => 599, Reason => "garbled response headers" }));	708	or return (%state = (), $cb->(undef, { @pseudo, Status => 599, Reason => "Garbled response headers" }));
		709
		710	%hdr = (%$hdr, @pseudo);
		711	}
		712
		713	# redirect handling
		714	# microsoft and other shitheads don't give a shit for following standards,
		715	# try to support some common forms of broken Location headers.
		716	if ($hdr{location} !~ /^(?: $ \| [^:\/?\#]+ : )/x) {
		717	$hdr{location} =~ s/^\.\/+//;
		718
		719	my $url = "$rscheme://$uhost:$uport";
		720
		721	unless ($hdr{location} =~ s/^\///) {
		722	$url .= $upath;
		723	$url =~ s/\/[^\/]*$//;
369	}	724	}
370		725
371	substr $_, 0, 1, ""	726	$hdr{location} = "$url/$hdr{location}";
372	for values %hdr;	727	}
373		728
374	my $finish = sub {	729	my $redirect;
		730
		731	if ($recurse) {
		732	my $status = $hdr{Status};
		733
		734	# industry standard is to redirect POST as GET for
		735	# 301, 302 and 303, in contrast to HTTP/1.0 and 1.1.
		736	# also, the UA should ask the user for 301 and 307 and POST,
		737	# industry standard seems to be to simply follow.
		738	# we go with the industry standard.
		739	if ($status == 301 or $status == 302 or $status == 303) {
		740	# HTTP/1.1 is unclear on how to mutate the method
		741	$method = "GET" unless $method eq "HEAD";
		742	$redirect = 1;
		743	} elsif ($status == 307) {
		744	$redirect = 1;
		745	}
		746	}
		747
		748	my $finish = sub { # ($data, $err_status, $err_reason[, $keepalive])
		749	my $may_keep_alive = $_[3];
		750
		751	$state{handle}->destroy if $state{handle};
375	%state = ();	752	%state = ();
376		753
		754	if (defined $_[1]) {
		755	$hdr{OrigStatus} = $hdr{Status}; $hdr{Status} = $_[1];
		756	$hdr{OrigReason} = $hdr{Reason}; $hdr{Reason} = $_[2];
		757	}
		758
377	# set-cookie processing	759	# set-cookie processing
378	if ($arg{cookie_jar} && exists $hdr{"set-cookie"}) {	760	if ($arg{cookie_jar}) {
379	for (split /\x00/, $hdr{"set-cookie"}) {	761	cookie_jar_set_cookie $arg{cookie_jar}, $hdr{"set-cookie"}, $uhost, $hdr{date};
380	my ($cookie, @arg) = split /;\s*/;	762	}
381	my ($name, $value) = split /=/, $cookie, 2;
382	my %kv = (value => $value, map { split /=/, $_, 2 } @arg);
383
384	my $cdom = (delete $kv{domain}) \|\| $uhost;
385	my $cpath = (delete $kv{path}) \|\| "/";
386
387	$cdom =~ s/^.?/./; # make sure it starts with a "."
388		763
389	next if $cdom =~ /\.$/;	764	if ($redirect && exists $hdr{location}) {
390		765	# we ignore any errors, as it is very common to receive
391	# this is not rfc-like and not netscape-like. go figure.	766	# Content-Length != 0 but no actual body
392	my $ndots = $cdom =~ y/.//;	767	# we also access %hdr, as $_[1] might be an erro
393	next if $ndots < ($cdom =~ /\.[^.][^.]\.[^.][^.]$/ ? 3 : 2);	768	http_request (
394		769	$method => $hdr{location},
		770	%arg,
		771	recurse => $recurse - 1,
		772	Redirect => [$_[0], \%hdr],
		773	$cb);
		774	} else {
		775	$cb->($_[0], \%hdr);
		776	}
		777	};
		778
		779	$ae_error = 597; # body phase
		780
		781	my $len = $hdr{"content-length"};
		782
		783	if (!$redirect && $arg{on_header} && !$arg{on_header}(\%hdr)) {
		784	$finish->(undef, 598 => "Request cancelled by on_header");
		785	} elsif (
		786	$hdr{Status} =~ /^(?:1..\|204\|205\|304)$/
		787	or $method eq "HEAD"
		788	or (defined $len && !$len)
		789	) {
		790	# no body
		791	$finish->("", undef, undef, 1);
		792	} else {
		793	# body handling, many different code paths
		794	# - no body expected
		795	# - want_body_handle
		796	# - te chunked
		797	# - 2x length known (with or without on_body)
		798	# - 2x length not known (with or without on_body)
		799	if (!$redirect && $arg{want_body_handle}) {
		800	$_[0]->on_eof (undef);
		801	$_[0]->on_error (undef);
		802	$_[0]->on_read (undef);
		803
		804	$finish->(delete $state{handle});
		805
		806	} elsif ($hdr{"transfer-encoding"} =~ /\bchunked\b/i) {
		807	my $cl = 0;
		808	my $body = undef;
		809	my $on_body = $arg{on_body} \|\| sub { $body .= shift; 1 };
		810
		811	$state{read_chunk} = sub {
		812	$_[1] =~ /^([0-9a-fA-F]+)/
		813	or $finish->(undef, $ae_error => "Garbled chunked transfer encoding");
		814
		815	my $len = hex $1;
		816
		817	if ($len) {
		818	$cl += $len;
		819
		820	$_[0]->push_read (chunk => $len, sub {
		821	$on_body->($_[1], \%hdr)
		822	or return $finish->(undef, 598 => "Request cancelled by on_body");
		823
		824	$_[0]->push_read (line => sub {
		825	length $_[1]
		826	and return $finish->(undef, $ae_error => "Garbled chunked transfer encoding");
		827	$_[0]->push_read (line => $state{read_chunk});
		828	});
395	# store it	829	});
396	$arg{cookie_jar}{version} = 1;	830	} else {
397	$arg{cookie_jar}{$cdom}{$cpath}{$name} = \%kv;	831	$hdr{"content-length"} \|\|= $cl;
		832
		833	$_[0]->push_read (line => $qr_nlnl, sub {
		834	if (length $_[1]) {
		835	for ("$_[1]") {
		836	y/\015//d; # weed out any \015, as they show up in the weirdest of places.
		837
		838	my $hdr = parse_hdr
		839	or return $finish->(undef, $ae_error => "Garbled response trailers");
		840
		841	%hdr = (%hdr, %$hdr);
		842	}
		843	}
		844
		845	$finish->($body, undef, undef, 1);
		846	});
398	}	847	}
399	}	848	};
400		849
401	if ($_[1]{Status} =~ /^x30[12]$/ && $recurse) {	850	$_[0]->push_read (line => $state{read_chunk});
402	# microsoft and other assholes don't give a shit for following standards,
403	# try to support a common form of broken Location header.
404	$_[1]{location} =~ s%^/%$scheme://$uhost:$uport/%;
405		851
406	http_request ($method, $_[1]{location}, %arg, recurse => $recurse - 1, $cb);	852	} elsif ($arg{on_body}) {
407	} else {	853	if ($len) {
408	$cb->($_[0], $_[1]);	854	$_[0]->on_read (sub {
409	}	855	$len -= length $_[0]{rbuf};
410	};
411		856
412	if ($hdr{Status} =~ /^(?:1..\|204\|304)$/ or $method eq "HEAD") {	857	$arg{on_body}(delete $_[0]{rbuf}, \%hdr)
413	$finish->(undef, \%hdr);	858	or return $finish->(undef, 598 => "Request cancelled by on_body");
414	} else {	859
415	if (exists $hdr{"content-length"}) {
416	$_[0]->unshift_read (chunk => $hdr{"content-length"}, sub {
417	# could cache persistent connection now
418	if ($hdr{connection} =~ /\bkeep-alive\b/i) {
419	# but we don't, due to misdesigns, this is annoyingly complex
420	};	860	$len > 0
421		861	or $finish->("", undef, undef, 1);
422	$finish->($_[1], \%hdr);
423	});	862	});
424	} else {	863	} else {
425	# too bad, need to read until we get an error or EOF,	864	$_[0]->on_eof (sub {
426	# no way to detect winged data.	865	$finish->("");
		866	});
		867	$_[0]->on_read (sub {
		868	$arg{on_body}(delete $_[0]{rbuf}, \%hdr)
		869	or $finish->(undef, 598 => "Request cancelled by on_body");
		870	});
		871	}
		872	} else {
		873	$_[0]->on_eof (undef);
		874
		875	if ($len) {
		876	$_[0]->on_read (sub {
		877	$finish->((substr delete $_[0]{rbuf}, 0, $len, ""), undef, undef, 1)
		878	if $len <= length $_[0]{rbuf};
		879	});
		880	} else {
427	$_[0]->on_error (sub {	881	$_[0]->on_error (sub {
		882	($! == Errno::EPIPE \|\| !$!)
428	$finish->($_[0]{rbuf}, \%hdr);	883	? $finish->(delete $_[0]{rbuf})
		884	: $finish->(undef, $ae_error => $_[2]);
429	});	885	});
430	$_[0]->on_eof (undef);
431	$_[0]->on_read (sub { });	886	$_[0]->on_read (sub { });
432	}	887	}
433	}	888	}
		889	}
		890	};
		891
		892	$state{handle}->push_read (line => $qr_nlnl, $state{read_response});
		893	};
		894
		895	my $connect_cb = sub {
		896	$state{fh} = shift
		897	or do {
		898	my $err = "$!";
		899	%state = ();
		900	return $cb->(undef, { @pseudo, Status => $ae_error, Reason => $err });
		901	};
		902
		903	return unless delete $state{connect_guard};
		904
		905	# get handle
		906	$state{handle} = new AnyEvent::Handle
		907	fh => $state{fh},
		908	peername => $rhost,
		909	tls_ctx => $arg{tls_ctx},
		910	# these need to be reconfigured on keepalive handles
		911	timeout => $timeout,
		912	on_error => sub {
		913	%state = ();
		914	$cb->(undef, { @pseudo, Status => $ae_error, Reason => $_[2] });
		915	},
		916	on_eof => sub {
		917	%state = ();
		918	$cb->(undef, { @pseudo, Status => $ae_error, Reason => "Unexpected end-of-file" });
		919	},
		920	;
		921
		922	# limit the number of persistent connections
		923	# keepalive not yet supported
		924	# if ($KA_COUNT{$_[1]} < $MAX_PERSISTENT_PER_HOST) {
		925	# ++$KA_COUNT{$_[1]};
		926	# $state{handle}{ka_count_guard} = AnyEvent::Util::guard {
		927	# --$KA_COUNT{$_[1]}
		928	# };
		929	# $hdr{connection} = "keep-alive";
		930	# }
		931
		932	$state{handle}->starttls ("connect") if $rscheme eq "https";
		933
		934	# now handle proxy-CONNECT method
		935	if ($proxy && $uscheme eq "https") {
		936	# oh dear, we have to wrap it into a connect request
		937
		938	# maybe re-use $uauthority with patched port?
		939	$state{handle}->push_write ("CONNECT $uhost:$uport HTTP/1.0\015\012Host: $uhost\015\012\015\012");
		940	$state{handle}->push_read (line => $qr_nlnl, sub {
		941	$_[1] =~ /^HTTP\/([0-9\.]+) \s+ ([0-9]{3}) (?: \s+ ([^\015\012]*) )?/ix
		942	or return (%state = (), $cb->(undef, { @pseudo, Status => 599, Reason => "Invalid proxy connect response ($_[1])" }));
		943
		944	if ($2 == 200) {
		945	$rpath = $upath;
		946	$handle_actual_request->();
		947	} else {
		948	%state = ();
		949	$cb->(undef, { @pseudo, Status => $2, Reason => $3 });
		950	}
434	});	951	});
		952	} else {
		953	$handle_actual_request->();
435	});	954	}
436	}, sub {
437	$timeout
438	};	955	};
		956
		957	my $tcp_connect = $arg{tcp_connect}
		958	\|\| do { require AnyEvent::Socket; \&AnyEvent::Socket::tcp_connect };
		959
		960	$state{connect_guard} = $tcp_connect->($rhost, $rport, $connect_cb, $arg{on_prepare} \|\| sub { $timeout });
439	};	961	};
440		962
441	defined wantarray && AnyEvent::Util::guard { %state = () }	963	defined wantarray && AnyEvent::Util::guard { %state = () }
442	}	964	}
443		965
444	sub http_get($$;@) {	966	sub http_get($@) {
445	unshift @_, "GET";	967	unshift @_, "GET";
446	&http_request	968	&http_request
447	}	969	}
448		970
449	sub http_head($$;@) {	971	sub http_head($@) {
450	unshift @_, "HEAD";	972	unshift @_, "HEAD";
451	&http_request	973	&http_request
452	}	974	}
453		975
454	sub http_post($$$;@) {	976	sub http_post($$@) {
		977	my $url = shift;
455	unshift @_, "POST", "body";	978	unshift @_, "POST", $url, "body";
456	&http_request	979	&http_request
457	}	980	}
458		981
459	=back	982	=back
460		983
		984	=head2 DNS CACHING
		985
		986	AnyEvent::HTTP uses the AnyEvent::Socket::tcp_connect function for
		987	the actual connection, which in turn uses AnyEvent::DNS to resolve
		988	hostnames. The latter is a simple stub resolver and does no caching
		989	on its own. If you want DNS caching, you currently have to provide
		990	your own default resolver (by storing a suitable resolver object in
		991	C<$AnyEvent::DNS::RESOLVER>).
		992
461	=head2 GLOBAL FUNCTIONS AND VARIABLES	993	=head2 GLOBAL FUNCTIONS AND VARIABLES
462		994
463	=over 4	995	=over 4
464		996
465	=item AnyEvent::HTTP::set_proxy "proxy-url"	997	=item AnyEvent::HTTP::set_proxy "proxy-url"
466		998
467	Sets the default proxy server to use. The proxy-url must begin with a	999	Sets the default proxy server to use. The proxy-url must begin with a
468	string of the form C<http://host:port> (optionally C<https:...>).	1000	string of the form C<http://host:port> (optionally C<https:...>), croaks
		1001	otherwise.
		1002
		1003	To clear an already-set proxy, use C<undef>.
		1004
		1005	=item AnyEvent::HTTP::cookie_jar_expire $jar[, $session_end]
		1006
		1007	Remove all cookies from the cookie jar that have been expired. If
		1008	C<$session_end> is given and true, then additionally remove all session
		1009	cookies.
		1010
		1011	You should call this function (with a true C<$session_end>) before you
		1012	save cookies to disk, and you should call this function after loading them
		1013	again. If you have a long-running program you can additonally call this
		1014	function from time to time.
		1015
		1016	A cookie jar is initially an empty hash-reference that is managed by this
		1017	module. It's format is subject to change, but currently it is like this:
		1018
		1019	The key C<version> has to contain C<1>, otherwise the hash gets
		1020	emptied. All other keys are hostnames or IP addresses pointing to
		1021	hash-references. The key for these inner hash references is the
		1022	server path for which this cookie is meant, and the values are again
		1023	hash-references. The keys of those hash-references is the cookie name, and
		1024	the value, you guessed it, is another hash-reference, this time with the
		1025	key-value pairs from the cookie, except for C<expires> and C<max-age>,
		1026	which have been replaced by a C<_expires> key that contains the cookie
		1027	expiry timestamp.
		1028
		1029	Here is an example of a cookie jar with a single cookie, so you have a
		1030	chance of understanding the above paragraph:
		1031
		1032	{
		1033	version => 1,
		1034	"10.0.0.1" => {
		1035	"/" => {
		1036	"mythweb_id" => {
		1037	_expires => 1293917923,
		1038	value => "ooRung9dThee3ooyXooM1Ohm",
		1039	},
		1040	},
		1041	},
		1042	}
		1043
		1044	=item $date = AnyEvent::HTTP::format_date $timestamp
		1045
		1046	Takes a POSIX timestamp (seconds since the epoch) and formats it as a HTTP
		1047	Date (RFC 2616).
		1048
		1049	=item $timestamp = AnyEvent::HTTP::parse_date $date
		1050
		1051	Takes a HTTP Date (RFC 2616) or a Cookie date (netscape cookie spec) or a
		1052	bunch of minor variations of those, and returns the corresponding POSIX
		1053	timestamp, or C<undef> if the date cannot be parsed.
469		1054
470	=item $AnyEvent::HTTP::MAX_RECURSE	1055	=item $AnyEvent::HTTP::MAX_RECURSE
471		1056
472	The default value for the C<recurse> request parameter (default: C<10>).	1057	The default value for the C<recurse> request parameter (default: C<10>).
473		1058
474	=item $AnyEvent::HTTP::USERAGENT	1059	=item $AnyEvent::HTTP::USERAGENT
475		1060
476	The default value for the C<User-Agent> header (the default is	1061	The default value for the C<User-Agent> header (the default is
477	C<Mozilla/5.0 (compatible; AnyEvent::HTTP/$VERSION; +http://software.schmorp.de/pkg/AnyEvent)>).	1062	C<Mozilla/5.0 (compatible; U; AnyEvent-HTTP/$VERSION; +http://software.schmorp.de/pkg/AnyEvent)>).
478		1063
479	=item $AnyEvent::HTTP::MAX_PERSISTENT	1064	=item $AnyEvent::HTTP::MAX_PER_HOST
480		1065
481	The maximum number of persistent connections to keep open (default: 8).	1066	The maximum number of concurrent connections to the same host (identified
		1067	by the hostname). If the limit is exceeded, then the additional requests
		1068	are queued until previous connections are closed.
482		1069
483	Not implemented currently.	1070	The default value for this is C<4>, and it is highly advisable to not
		1071	increase it.
484		1072
485	=item $AnyEvent::HTTP::PERSISTENT_TIMEOUT	1073	=item $AnyEvent::HTTP::ACTIVE
486		1074
487	The maximum time to cache a persistent connection, in seconds (default: 2).	1075	The number of active connections. This is not the number of currently
488		1076	running requests, but the number of currently open and non-idle TCP
489	Not implemented currently.	1077	connections. This number of can be useful for load-leveling.
490		1078
491	=back	1079	=back
492		1080
493	=cut	1081	=cut
494		1082
		1083	our @month = qw(Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec);
		1084	our @weekday = qw(Sun Mon Tue Wed Thu Fri Sat);
		1085
		1086	sub format_date($) {
		1087	my ($time) = @_;
		1088
		1089	# RFC 822/1123 format
		1090	my ($S, $M, $H, $mday, $mon, $year, $wday, $yday, undef) = gmtime $time;
		1091
		1092	sprintf "%s, %02d %s %04d %02d:%02d:%02d GMT",
		1093	$weekday[$wday], $mday, $month[$mon], $year + 1900,
		1094	$H, $M, $S;
		1095	}
		1096
		1097	sub parse_date($) {
		1098	my ($date) = @_;
		1099
		1100	my ($d, $m, $y, $H, $M, $S);
		1101
		1102	if ($date =~ /^[A-Z][a-z][a-z]+, ([0-9][0-9]?)[\- ]([A-Z][a-z][a-z])[\- ]([0-9][0-9][0-9][0-9]) ([0-9][0-9]?):([0-9][0-9]?):([0-9][0-9]?) GMT$/) {
		1103	# RFC 822/1123, required by RFC 2616 (with " ")
		1104	# cookie dates (with "-")
		1105
		1106	($d, $m, $y, $H, $M, $S) = ($1, $2, $3, $4, $5, $6);
		1107
		1108	} elsif ($date =~ /^[A-Z][a-z][a-z]+, ([0-9][0-9]?)-([A-Z][a-z][a-z])-([0-9][0-9]) ([0-9][0-9]?):([0-9][0-9]?):([0-9][0-9]?) GMT$/) {
		1109	# RFC 850
		1110	($d, $m, $y, $H, $M, $S) = ($1, $2, $3 < 69 ? $3 + 2000 : $3 + 1900, $4, $5, $6);
		1111
		1112	} elsif ($date =~ /^[A-Z][a-z][a-z]+ ([A-Z][a-z][a-z]) ([0-9 ]?[0-9]) ([0-9][0-9]?):([0-9][0-9]?):([0-9][0-9]?) ([0-9][0-9][0-9][0-9])$/) {
		1113	# ISO C's asctime
		1114	($d, $m, $y, $H, $M, $S) = ($2, $1, $6, $3, $4, $5);
		1115	}
		1116	# other formats fail in the loop below
		1117
		1118	for (0..11) {
		1119	if ($m eq $month[$_]) {
		1120	require Time::Local;
		1121	return Time::Local::timegm ($S, $M, $H, $d, $_, $y);
		1122	}
		1123	}
		1124
		1125	undef
		1126	}
		1127
495	sub set_proxy($) {	1128	sub set_proxy($) {
		1129	if (length $_[0]) {
496	$PROXY = [$2, $3 \|\| 3128, $1] if $_[0] =~ m%^(https?):// ([^:/]+) (?: : (\d*) )?%ix;	1130	$_[0] =~ m%^(https?):// ([^:/]+) (?: : (\d*) )?%ix
		1131	or Carp::croak "$_[0]: invalid proxy URL";
		1132	$PROXY = [$2, $3 \|\| 3128, $1]
		1133	} else {
		1134	undef $PROXY;
		1135	}
497	}	1136	}
498		1137
499	# initialise proxy from environment	1138	# initialise proxy from environment
		1139	eval {
500	set_proxy $ENV{http_proxy};	1140	set_proxy $ENV{http_proxy};
		1141	};
		1142
		1143	=head2 SOCKS PROXIES
		1144
		1145	Socks proxies are not directly supported by AnyEvent::HTTP. You can
		1146	compile your perl to support socks, or use an external program such as
		1147	F<socksify> (dante) or F<tsocks> to make your program use a socks proxy
		1148	transparently.
		1149
		1150	Alternatively, for AnyEvent::HTTP only, you can use your own
		1151	C<tcp_connect> function that does the proxy handshake - here is an example
		1152	that works with socks4a proxies:
		1153
		1154	use Errno;
		1155	use AnyEvent::Util;
		1156	use AnyEvent::Socket;
		1157	use AnyEvent::Handle;
		1158
		1159	# host, port and username of/for your socks4a proxy
		1160	my $socks_host = "10.0.0.23";
		1161	my $socks_port = 9050;
		1162	my $socks_user = "";
		1163
		1164	sub socks4a_connect {
		1165	my ($host, $port, $connect_cb, $prepare_cb) = @_;
		1166
		1167	my $hdl = new AnyEvent::Handle
		1168	connect => [$socks_host, $socks_port],
		1169	on_prepare => sub { $prepare_cb->($_[0]{fh}) },
		1170	on_error => sub { $connect_cb->() },
		1171	;
		1172
		1173	$hdl->push_write (pack "CCnNZZ", 4, 1, $port, 1, $socks_user, $host);
		1174
		1175	$hdl->push_read (chunk => 8, sub {
		1176	my ($hdl, $chunk) = @_;
		1177	my ($status, $port, $ipn) = unpack "xCna4", $chunk;
		1178
		1179	if ($status == 0x5a) {
		1180	$connect_cb->($hdl->{fh}, (format_address $ipn) . ":$port");
		1181	} else {
		1182	$! = Errno::ENXIO; $connect_cb->();
		1183	}
		1184	});
		1185
		1186	$hdl
		1187	}
		1188
		1189	Use C<socks4a_connect> instead of C<tcp_connect> when doing C<http_request>s,
		1190	possibly after switching off other proxy types:
		1191
		1192	AnyEvent::HTTP::set_proxy undef; # usually you do not want other proxies
		1193
		1194	http_get 'http://www.google.com', tcp_connect => \&socks4a_connect, sub {
		1195	my ($data, $headers) = @_;
		1196	...
		1197	};
501		1198
502	=head1 SEE ALSO	1199	=head1 SEE ALSO
503		1200
504	L<AnyEvent>.	1201	L<AnyEvent>.
505		1202
506	=head1 AUTHOR	1203	=head1 AUTHOR
507		1204
508	Marc Lehmann <schmorp@schmorp.de>	1205	Marc Lehmann <schmorp@schmorp.de>
509	http://home.schmorp.de/	1206	http://home.schmorp.de/
		1207
		1208	With many thanks to Дмитрий Шалашов, who provided countless
		1209	testcases and bugreports.
510		1210
511	=cut	1211	=cut
512		1212
513	1	1213	1
514		1214

Diff Legend

-–
+Removed lines
-+
+Added lines
-<
+Changed lines
->
+Changed lines

Comparing AnyEvent-HTTP/HTTP.pm (file contents): Revision 1.11 by root, Thu Jun 5 15:34:00 2008 UTC vs. Revision 1.83 by root, Sun Jan 2 05:02:28 2011 UTC

Diff Legend

Comparing AnyEvent-HTTP/HTTP.pm (file contents):
Revision 1.11 by root, Thu Jun 5 15:34:00 2008 UTC vs.
Revision 1.83 by root, Sun Jan 2 05:02:28 2011 UTC