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

Comparing AnyEvent-HTTP/HTTP.pm (file contents):
Revision 1.2 by root, Wed Jun 4 11:37:41 2008 UTC vs.
Revision 1.88 by root, Sun Jan 2 20:57:03 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.
13		17
		18	This module implements a simple, stateless and non-blocking HTTP
		19	client. It supports GET, POST and other request methods, cookies and more,
		20	all on a very low level. It can follow redirects supports proxies and
		21	automatically limits the number of connections to the values specified in
		22	the RFC.
		23
		24	It should generally be a "good client" that is enough for most HTTP
		25	tasks. Simple tasks should be simple, but complex tasks should still be
		26	possible as the user retains control over request and response headers.
		27
		28	The caller is responsible for authentication management, cookies (if
		29	the simplistic implementation in this module doesn't suffice), referer
		30	and other high-level protocol details for which this module offers only
		31	limited support.
		32
14	=head2 METHODS	33	=head2 METHODS
15		34
16	=over 4	35	=over 4
17		36
18	=cut	37	=cut
19		38
20	package AnyEvent::HTTP;	39	package AnyEvent::HTTP;
21		40
22	use strict;	41	use common::sense;
23	no warnings;
24		42
25	use Carp;	43	use Errno ();
26		44
27	use AnyEvent ();	45	use AnyEvent 5.0 ();
28	use AnyEvent::Util ();	46	use AnyEvent::Util ();
29	use AnyEvent::Socket ();
30	use AnyEvent::Handle ();	47	use AnyEvent::Handle ();
31		48
32	use base Exporter::;	49	use base Exporter::;
33		50
34	our $VERSION = '1.0';	51	our $VERSION = '1.5';
35		52
36	our @EXPORT = qw(http_get http_request);	53	our @EXPORT = qw(http_get http_post http_head http_request);
37		54
38	our $USERAGENT = "Mozilla/5.0 (compatible; AnyEvent::HTTP/$VERSION; +http://software.schmorp.de/pkg/AnyEvent)";	55	our $USERAGENT = "Mozilla/5.0 (compatible; U; AnyEvent-HTTP/$VERSION; +http://software.schmorp.de/pkg/AnyEvent)";
39	our $MAX_REDIRECTS = 10;	56	our $MAX_RECURSE = 10;
40	our $MAX_PERSISTENT = 8;	57	our $MAX_PERSISTENT = 8;
41	our $PERSISTENT_TIMEOUT = 2;	58	our $PERSISTENT_TIMEOUT = 2;
42	our $TIMEOUT = 300;	59	our $TIMEOUT = 300;
43		60
44	# changing these is evil	61	# changing these is evil
45	our $MAX_PERSISTENT_PER_HOST = 2;	62	our $MAX_PERSISTENT_PER_HOST = 2;
46	our $MAX_PER_HOST = 4; # not respected yet :(	63	our $MAX_PER_HOST = 4;
47		64
48	our $PROXY;	65	our $PROXY;
		66	our $ACTIVE = 0;
49		67
50	my %KA_COUNT; # number of open keep-alive connections per host	68	my %KA_COUNT; # number of open keep-alive connections per host
		69	my %CO_SLOT; # number of open connections, and wait queue, per host
51		70
52	=item http_get $url, key => value..., $cb->($data, $headers)	71	=item http_get $url, key => value..., $cb->($data, $headers)
53		72
54	Executes an HTTP-GET request. See the http_request function for details on	73	Executes an HTTP-GET request. See the http_request function for details on
55	additional parameters.	74	additional parameters and the return value.
		75
		76	=item http_head $url, key => value..., $cb->($data, $headers)
		77
		78	Executes an HTTP-HEAD request. See the http_request function for details
		79	on additional parameters and the return value.
		80
		81	=item http_post $url, $body, key => value..., $cb->($data, $headers)
		82
		83	Executes an HTTP-POST request with a request body of C<$body>. See the
		84	http_request function for details on additional parameters and the return
		85	value.
56		86
57	=item http_request $method => $url, key => value..., $cb->($data, $headers)	87	=item http_request $method => $url, key => value..., $cb->($data, $headers)
58		88
59	Executes a HTTP request of type C<$method> (e.g. C<GET>, C<POST>). The URL	89	Executes a HTTP request of type C<$method> (e.g. C<GET>, C<POST>). The URL
60	must be an absolute http or https URL.	90	must be an absolute http or https URL.
61		91
		92	When called in void context, nothing is returned. In other contexts,
		93	C<http_request> returns a "cancellation guard" - you have to keep the
		94	object at least alive until the callback get called. If the object gets
		95	destroyed before the callback is called, the request will be cancelled.
		96
62	The callback will be called with the response data as first argument	97	The callback will be called with the response body data as first argument
63	(or C<undef> if it wasn't available due to errors), and a hash-ref with	98	(or C<undef> if an error occured), and a hash-ref with response headers
64	response headers as second argument.	99	(and trailers) as second argument.
65		100
66	All the headers in that has are lowercased. In addition to the response	101	All the headers in that hash are lowercased. In addition to the response
67	headers, the three "pseudo-headers" C<HTTPVersion>, C<Status> and	102	headers, the "pseudo-headers" (uppercase to avoid clashing with possible
68	C<Reason> contain the three parts of the HTTP Status-Line of the same	103	response headers) C<HTTPVersion>, C<Status> and C<Reason> contain the
69	name.	104	three parts of the HTTP Status-Line of the same name. If an error occurs
		105	during the body phase of a request, then the original C<Status> and
		106	C<Reason> values from the header are available as C<OrigStatus> and
		107	C<OrigReason>.
		108
		109	The pseudo-header C<URL> contains the actual URL (which can differ from
		110	the requested URL when following redirects - for example, you might get
		111	an error that your URL scheme is not supported even though your URL is a
		112	valid http URL because it redirected to an ftp URL, in which case you can
		113	look at the URL pseudo header).
		114
		115	The pseudo-header C<Redirect> only exists when the request was a result
		116	of an internal redirect. In that case it is an array reference with
		117	the C<($data, $headers)> from the redirect response. Note that this
		118	response could in turn be the result of a redirect itself, and C<<
		119	$headers->{Redirect}[1]{Redirect} >> will then contain the original
		120	response, and so on.
		121
		122	If the server sends a header multiple times, then their contents will be
		123	joined together with a comma (C<,>), as per the HTTP spec.
70		124
71	If an internal error occurs, such as not being able to resolve a hostname,	125	If an internal error occurs, such as not being able to resolve a hostname,
72	then C<$data> will be C<undef>, C<< $headers->{Status} >> will be C<599>	126	then C<$data> will be C<undef>, C<< $headers->{Status} >> will be
73	and the C<Reason> pseudo-header will contain an error message.	127	C<590>-C<599> and the C<Reason> pseudo-header will contain an error
		128	message. Currently the following status codes are used:
		129
		130	=over 4
		131
		132	=item 595 - errors during connection etsbalishment, proxy handshake.
		133
		134	=item 596 - errors during TLS negotiation, request sending and header processing.
		135
		136	=item 597 - errors during body receiving or processing.
		137
		138	=item 598 - user aborted request via C<on_header> or C<on_body>.
		139
		140	=item 599 - other, usually nonretryable, errors (garbled URL etc.).
		141
		142	=back
		143
		144	A typical callback might look like this:
		145
		146	sub {
		147	my ($body, $hdr) = @_;
		148
		149	if ($hdr->{Status} =~ /^2/) {
		150	... everything should be ok
		151	} else {
		152	print "error, $hdr->{Status} $hdr->{Reason}\n";
		153	}
		154	}
74		155
75	Additional parameters are key-value pairs, and are fully optional. They	156	Additional parameters are key-value pairs, and are fully optional. They
76	include:	157	include:
77		158
78	=over 4	159	=over 4
79		160
80	=item recurse => $boolean (default: true)	161	=item recurse => $count (default: $MAX_RECURSE)
81		162
82	Whether to recurse requests or not, e.g. on redirects, authentication	163	Whether to recurse requests or not, e.g. on redirects, authentication
83	retries and so on.	164	retries and so on, and how often to do so.
84		165
85	=item headers => hashref	166	=item headers => hashref
86		167
87	The request headers to use.	168	The request headers to use. Currently, C<http_request> may provide its own
		169	C<Host:>, C<Content-Length:>, C<Connection:> and C<Cookie:> headers and
		170	will provide defaults at least for C<TE:>, C<Referer:> and C<User-Agent:>
		171	(this can be suppressed by using C<undef> for these headers in which case
		172	they won't be sent at all).
88		173
89	=item timeout => $seconds	174	=item timeout => $seconds
90		175
91	The time-out to use for various stages - each connect attempt will reset	176	The time-out to use for various stages - each connect attempt will reset
92	the timeout, as will read or write activity. Default timeout is 5 minutes.	177	the timeout, as will read or write activity, i.e. this is not an overall
		178	timeout.
		179
		180	Default timeout is 5 minutes.
93		181
94	=item proxy => [$host, $port[, $scheme]] or undef	182	=item proxy => [$host, $port[, $scheme]] or undef
95		183
96	Use the given http proxy for all requests. If not specified, then the	184	Use the given http proxy for all requests. If not specified, then the
97	default proxy (as specified by C<$ENV{http_proxy}>) is used.	185	default proxy (as specified by C<$ENV{http_proxy}>) is used.
98		186
99	C<$scheme> must be either missing or C<http> for HTTP, or C<https> for	187	C<$scheme> must be either missing, C<http> for HTTP or C<https> for
100	HTTPS.	188	HTTPS.
101		189
		190	=item body => $string
		191
		192	The request body, usually empty. Will be sent as-is (future versions of
		193	this module might offer more options).
		194
		195	=item cookie_jar => $hash_ref
		196
		197	Passing this parameter enables (simplified) cookie-processing, loosely
		198	based on the original netscape specification.
		199
		200	The C<$hash_ref> must be an (initially empty) hash reference which
		201	will get updated automatically. It is possible to save the cookie jar
		202	to persistent storage with something like JSON or Storable - see the
		203	C<AnyEvent::HTTP::cookie_jar_expire> function if you wish to remove
		204	expired or session-only cookies, and also for documentation on the format
		205	of the cookie jar.
		206
		207	Note that this cookie implementation is not meant to be complete. If
		208	you want complete cookie management you have to do that on your
		209	own. C<cookie_jar> is meant as a quick fix to get most cookie-using sites
		210	working. Cookies are a privacy disaster, do not use them unless required
		211	to.
		212
		213	When cookie processing is enabled, the C<Cookie:> and C<Set-Cookie:>
		214	headers will be set and handled by this module, otherwise they will be
		215	left untouched.
		216
		217	=item tls_ctx => $scheme \| $tls_ctx
		218
		219	Specifies the AnyEvent::TLS context to be used for https connections. This
		220	parameter follows the same rules as the C<tls_ctx> parameter to
		221	L<AnyEvent::Handle>, but additionally, the two strings C<low> or
		222	C<high> can be specified, which give you a predefined low-security (no
		223	verification, highest compatibility) and high-security (CA and common-name
		224	verification) TLS context.
		225
		226	The default for this option is C<low>, which could be interpreted as "give
		227	me the page, no matter what".
		228
		229	=item on_prepare => $callback->($fh)
		230
		231	In rare cases you need to "tune" the socket before it is used to
		232	connect (for exmaple, to bind it on a given IP address). This parameter
		233	overrides the prepare callback passed to C<AnyEvent::Socket::tcp_connect>
		234	and behaves exactly the same way (e.g. it has to provide a
		235	timeout). See the description for the C<$prepare_cb> argument of
		236	C<AnyEvent::Socket::tcp_connect> for details.
		237
		238	=item tcp_connect => $callback->($host, $service, $connect_cb, $prepare_cb)
		239
		240	In even rarer cases you want total control over how AnyEvent::HTTP
		241	establishes connections. Normally it uses L<AnyEvent::Socket::tcp_connect>
		242	to do this, but you can provide your own C<tcp_connect> function -
		243	obviously, it has to follow the same calling conventions, except that it
		244	may always return a connection guard object.
		245
		246	There are probably lots of weird uses for this function, starting from
		247	tracing the hosts C<http_request> actually tries to connect, to (inexact
		248	but fast) host => IP address caching or even socks protocol support.
		249
		250	=item on_header => $callback->($headers)
		251
		252	When specified, this callback will be called with the header hash as soon
		253	as headers have been successfully received from the remote server (not on
		254	locally-generated errors).
		255
		256	It has to return either true (in which case AnyEvent::HTTP will continue),
		257	or false, in which case AnyEvent::HTTP will cancel the download (and call
		258	the finish callback with an error code of C<598>).
		259
		260	This callback is useful, among other things, to quickly reject unwanted
		261	content, which, if it is supposed to be rare, can be faster than first
		262	doing a C<HEAD> request.
		263
		264	The downside is that cancelling the request makes it impossible to re-use
		265	the connection. Also, the C<on_header> callback will not receive any
		266	trailer (headers sent after the response body).
		267
		268	Example: cancel the request unless the content-type is "text/html".
		269
		270	on_header => sub {
		271	$_[0]{"content-type"} =~ /^text\/html\s*(?:;\|$)/
		272	},
		273
		274	=item on_body => $callback->($partial_body, $headers)
		275
		276	When specified, all body data will be passed to this callback instead of
		277	to the completion callback. The completion callback will get the empty
		278	string instead of the body data.
		279
		280	It has to return either true (in which case AnyEvent::HTTP will continue),
		281	or false, in which case AnyEvent::HTTP will cancel the download (and call
		282	the completion callback with an error code of C<598>).
		283
		284	The downside to cancelling the request is that it makes it impossible to
		285	re-use the connection.
		286
		287	This callback is useful when the data is too large to be held in memory
		288	(so the callback writes it to a file) or when only some information should
		289	be extracted, or when the body should be processed incrementally.
		290
		291	It is usually preferred over doing your own body handling via
		292	C<want_body_handle>, but in case of streaming APIs, where HTTP is
		293	only used to create a connection, C<want_body_handle> is the better
		294	alternative, as it allows you to install your own event handler, reducing
		295	resource usage.
		296
		297	=item want_body_handle => $enable
		298
		299	When enabled (default is disabled), the behaviour of AnyEvent::HTTP
		300	changes considerably: after parsing the headers, and instead of
		301	downloading the body (if any), the completion callback will be
		302	called. Instead of the C<$body> argument containing the body data, the
		303	callback will receive the L<AnyEvent::Handle> object associated with the
		304	connection. In error cases, C<undef> will be passed. When there is no body
		305	(e.g. status C<304>), the empty string will be passed.
		306
		307	The handle object might or might not be in TLS mode, might be connected to
		308	a proxy, be a persistent connection etc., and configured in unspecified
		309	ways. The user is responsible for this handle (it will not be used by this
		310	module anymore).
		311
		312	This is useful with some push-type services, where, after the initial
		313	headers, an interactive protocol is used (typical example would be the
		314	push-style twitter API which starts a JSON/XML stream).
		315
		316	If you think you need this, first have a look at C<on_body>, to see if
		317	that doesn't solve your problem in a better way.
		318
102	=back	319	=back
103		320
104	=back	321	Example: do a simple HTTP GET request for http://www.nethype.de/ and print
		322	the response body.
		323
		324	http_request GET => "http://www.nethype.de/", sub {
		325	my ($body, $hdr) = @_;
		326	print "$body\n";
		327	};
		328
		329	Example: do a HTTP HEAD request on https://www.google.com/, use a
		330	timeout of 30 seconds.
		331
		332	http_request
		333	GET => "https://www.google.com",
		334	timeout => 30,
		335	sub {
		336	my ($body, $hdr) = @_;
		337	use Data::Dumper;
		338	print Dumper $hdr;
		339	}
		340	;
		341
		342	Example: do another simple HTTP GET request, but immediately try to
		343	cancel it.
		344
		345	my $request = http_request GET => "http://www.nethype.de/", sub {
		346	my ($body, $hdr) = @_;
		347	print "$body\n";
		348	};
		349
		350	undef $request;
105		351
106	=cut	352	=cut
107		353
		354	sub _slot_schedule;
		355	sub _slot_schedule($) {
		356	my $host = shift;
		357
		358	while ($CO_SLOT{$host}[0] < $MAX_PER_HOST) {
		359	if (my $cb = shift @{ $CO_SLOT{$host}[1] }) {
		360	# somebody wants that slot
		361	++$CO_SLOT{$host}[0];
		362	++$ACTIVE;
		363
		364	$cb->(AnyEvent::Util::guard {
		365	--$ACTIVE;
		366	--$CO_SLOT{$host}[0];
		367	_slot_schedule $host;
		368	});
		369	} else {
		370	# nobody wants the slot, maybe we can forget about it
		371	delete $CO_SLOT{$host} unless $CO_SLOT{$host}[0];
		372	last;
		373	}
		374	}
		375	}
		376
		377	# wait for a free slot on host, call callback
		378	sub _get_slot($$) {
		379	push @{ $CO_SLOT{$_[0]}[1] }, $_[1];
		380
		381	_slot_schedule $_[0];
		382	}
		383
		384	#############################################################################
		385
		386	# expire cookies
		387	sub cookie_jar_expire($;$) {
		388	my ($jar, $session_end) = @_;
		389
		390	%$jar = () if $jar->{version} != 1;
		391
		392	my $anow = AE::now;
		393
		394	while (my ($chost, $paths) = each %$jar) {
		395	next unless ref $paths;
		396
		397	while (my ($cpath, $cookies) = each %$paths) {
		398	while (my ($cookie, $kv) = each %$cookies) {
		399	if (exists $kv->{_expires}) {
		400	delete $cookies->{$cookie}
		401	if $anow > $kv->{_expires};
		402	} elsif ($session_end) {
		403	delete $cookies->{$cookie};
		404	}
		405	}
		406
		407	delete $paths->{$cpath}
		408	unless %$cookies;
		409	}
		410
		411	delete $jar->{$chost}
		412	unless %$paths;
		413	}
		414	}
		415
		416	# extract cookies from jar
		417	sub cookie_jar_extract($$$$) {
		418	my ($jar, $uscheme, $uhost, $upath) = @_;
		419
		420	%$jar = () if $jar->{version} != 1;
		421
		422	my @cookies;
		423
		424	while (my ($chost, $paths) = each %$jar) {
		425	next unless ref $paths;
		426
		427	if ($chost =~ /^\./) {
		428	next unless $chost eq substr $uhost, -length $chost;
		429	} elsif ($chost =~ /\./) {
		430	next unless $chost eq $uhost;
		431	} else {
		432	next;
		433	}
		434
		435	while (my ($cpath, $cookies) = each %$paths) {
		436	next unless $cpath eq substr $upath, 0, length $cpath;
		437
		438	while (my ($cookie, $kv) = each %$cookies) {
		439	next if $uscheme ne "https" && exists $kv->{secure};
		440
		441	if (exists $kv->{_expires} and AE::now > $kv->{_expires}) {
		442	delete $cookies->{$cookie};
		443	next;
		444	}
		445
		446	my $value = $kv->{value};
		447
		448	if ($value =~ /[=;,[:space:]]/) {
		449	$value =~ s/([\\"])/\\$1/g;
		450	$value = "\"$value\"";
		451	}
		452
		453	push @cookies, "$cookie=$value";
		454	}
		455	}
		456	}
		457
		458	\@cookies
		459	}
		460
		461	# parse set_cookie header into jar
		462	sub cookie_jar_set_cookie($$$$) {
		463	my ($jar, $set_cookie, $uhost, $date) = @_;
		464
		465	my $anow = int AE::now;
		466	my $snow; # server-now
		467
		468	for ($set_cookie) {
		469	# parse NAME=VALUE
		470	my @kv;
		471
		472	# expires is not http-compliant in the original cookie-spec,
		473	# we support the official date format and some extensions
		474	while (
		475	m{
		476	\G\s*
		477	(?:
		478	expires \s=\s ([A-Z][a-z][a-z]+,\ [^,;]+)
		479	\| ([^=;,[:space:]]+) (?: \s=\s (?: "((?:[^\\"]+\|\\.))" \| ([^=;,[:space:]]) ) )?
		480	)
		481	}gcxsi
		482	) {
		483	my $name = $2;
		484	my $value = $4;
		485
		486	if (defined $1) {
		487	# expires
		488	$name = "expires";
		489	$value = $1;
		490	} elsif (defined $3) {
		491	# quoted
		492	$value = $3;
		493	$value =~ s/\\(.)/$1/gs;
		494	}
		495
		496	push @kv, lc $name, $value;
		497
		498	last unless /\G\s*;/gc;
		499	}
		500
		501	last unless @kv;
		502
		503	my $name = shift @kv;
		504	my %kv = (value => shift @kv, @kv);
		505
		506	if (exists $kv{"max-age"}) {
		507	$kv{_expires} = $anow + delete $kv{"max-age"};
		508	} elsif (exists $kv{expires}) {
		509	$snow \|\|= parse_date ($date) \|\| $anow;
		510	$kv{_expires} = $anow + (parse_date (delete $kv{expires}) - $snow);
		511	} else {
		512	delete $kv{_expires};
		513	}
		514
		515	my $cdom;
		516	my $cpath = (delete $kv{path}) \|\| "/";
		517
		518	if (exists $kv{domain}) {
		519	$cdom = delete $kv{domain};
		520
		521	$cdom =~ s/^\.?/./; # make sure it starts with a "."
		522
		523	next if $cdom =~ /\.$/;
		524
		525	# this is not rfc-like and not netscape-like. go figure.
		526	my $ndots = $cdom =~ y/.//;
		527	next if $ndots < ($cdom =~ /\.[^.][^.]\.[^.][^.]$/ ? 3 : 2);
		528	} else {
		529	$cdom = $uhost;
		530	}
		531
		532	# store it
		533	$jar->{version} = 1;
		534	$jar->{lc $cdom}{$cpath}{$name} = \%kv;
		535
		536	redo if /\G\s*,/gc;
		537	}
		538	}
		539
		540	# continue to parse $_ for headers and place them into the arg
		541	sub parse_hdr() {
		542	my %hdr;
		543
		544	# things seen, not parsed:
		545	# p3pP="NON CUR OTPi OUR NOR UNI"
		546
		547	$hdr{lc $1} .= ",$2"
		548	while /\G
		549	([^:\000-\037]*):
		550	[\011\040]*
		551	((?: [^\012]+ \| \012[\011\040] )*)
		552	\012
		553	/gxc;
		554
		555	/\G$/
		556	or return;
		557
		558	# remove the "," prefix we added to all headers above
		559	substr $_, 0, 1, ""
		560	for values %hdr;
		561
		562	\%hdr
		563	}
		564
		565	our $qr_nlnl = qr{(?<![^\012])\015?\012};
		566
		567	our $TLS_CTX_LOW = { cache => 1, sslv2 => 1 };
		568	our $TLS_CTX_HIGH = { cache => 1, verify => 1, verify_peername => "https" };
		569
108	sub http_request($$$;@) {	570	sub http_request($$@) {
109	my $cb = pop;	571	my $cb = pop;
110	my ($method, $url, %arg) = @_;	572	my ($method, $url, %arg) = @_;
111		573
112	my %hdr;	574	my %hdr;
113		575
		576	$arg{tls_ctx} = $TLS_CTX_LOW if $arg{tls_ctx} eq "low" \|\| !exists $arg{tls_ctx};
		577	$arg{tls_ctx} = $TLS_CTX_HIGH if $arg{tls_ctx} eq "high";
		578
		579	$method = uc $method;
		580
114	if (my $hdr = delete $arg{headers}) {	581	if (my $hdr = $arg{headers}) {
115	while (my ($k, $v) = each %$hdr) {	582	while (my ($k, $v) = each %$hdr) {
116	$hdr{lc $k} = $v;	583	$hdr{lc $k} = $v;
117	}	584	}
118	}	585	}
119		586
		587	# pseudo headers for all subsequent responses
		588	my @pseudo = (URL => $url);
		589	push @pseudo, Redirect => delete $arg{Redirect} if exists $arg{Redirect};
		590
		591	my $recurse = exists $arg{recurse} ? delete $arg{recurse} : $MAX_RECURSE;
		592
		593	return $cb->(undef, { @pseudo, Status => 599, Reason => "Too many redirections" })
		594	if $recurse < 0;
		595
120	my $proxy = $arg{proxy} \|\| $PROXY;	596	my $proxy = $arg{proxy} \|\| $PROXY;
121	my $timeout = $arg{timeout} \|\| $TIMEOUT;	597	my $timeout = $arg{timeout} \|\| $TIMEOUT;
122		598
123	$hdr{"user-agent"} \|\|= $USERAGENT;	599	my ($uscheme, $uauthority, $upath, $query, $fragment) =
		600	$url =~ m\|(?:([^:/?#]+):)?(?://([^/?#]))?([^?#])(?:(\?[^#]))?(?:#(.))?\|;
124		601
125	my ($host, $port, $path, $scheme);	602	$uscheme = lc $uscheme;
		603
		604	my $uport = $uscheme eq "http" ? 80
		605	: $uscheme eq "https" ? 443
		606	: return $cb->(undef, { @pseudo, Status => 599, Reason => "Only http and https URL schemes supported" });
		607
		608	$uauthority =~ /^(?: .*\@ )? ([^\@:]+) (?: : (\d+) )?$/x
		609	or return $cb->(undef, { @pseudo, Status => 599, Reason => "Unparsable URL" });
		610
		611	my $uhost = lc $1;
		612	$uport = $2 if defined $2;
		613
		614	$hdr{host} = defined $2 ? "$uhost:$2" : "$uhost"
		615	unless exists $hdr{host};
		616
		617	$uhost =~ s/^\[(.*)\]$/$1/;
		618	$upath .= $query if length $query;
		619
		620	$upath =~ s%^/?%/%;
		621
		622	# cookie processing
		623	if (my $jar = $arg{cookie_jar}) {
		624	my $cookies = cookie_jar_extract $jar, $uscheme, $uhost, $upath;
		625
		626	$hdr{cookie} = join "; ", @$cookies
		627	if @$cookies;
		628	}
		629
		630	my ($rhost, $rport, $rscheme, $rpath); # request host, port, path
126		631
127	if ($proxy) {	632	if ($proxy) {
128	($host, $port, $scheme) = @$proxy;	633	($rpath, $rhost, $rport, $rscheme) = ($url, @$proxy);
129	$path = $url;	634
		635	$rscheme = "http" unless defined $rscheme;
		636
		637	# don't support https requests over https-proxy transport,
		638	# can't be done with tls as spec'ed, unless you double-encrypt.
		639	$rscheme = "http" if $uscheme eq "https" && $rscheme eq "https";
		640
		641	$rhost = lc $rhost;
		642	$rscheme = lc $rscheme;
130	} else {	643	} else {
131	($scheme, my $authority, $path, my $query, my $fragment) =	644	($rhost, $rport, $rscheme, $rpath) = ($uhost, $uport, $uscheme, $upath);
132	$url =~ m\|(?:([^:/?#]+):)?(?://([^/?#]))?([^?#])(?:\?([^#]))?(?:#(.))?\|;
133
134	$port = $scheme eq "http" ? 80
135	: $scheme eq "https" ? 443
136	: croak "$url: only http and https URLs supported";
137
138	$authority =~ /^(?: .*\@ )? ([^\@:]+) (?: : (\d+) )?$/x
139	or croak "$authority: unparsable URL";
140
141	$host = $1;
142	$port = $2 if defined $2;
143
144	$host =~ s/^\[(.*)\]$/$1/;
145	$path .= "?$query" if length $query;
146
147	$path = "/" unless $path;
148
149	$hdr{host} = $host = lc $host;
150	}	645	}
151		646
152	$scheme = lc $scheme;	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"};
153		650
154	my %state;
155
156	my $body = "";
157	$state{body} = $body;
158
159	$hdr{"content-length"} = length $body;	651	$hdr{"content-length"} = length $arg{body}
		652	if length $arg{body} \|\| $method ne "GET";
160		653
161	$state{connect_guard} = AnyEvent::Socket::tcp_connect $host, $port, sub {	654	$hdr{connection} = "close Te"; #1.1
		655	$hdr{te} = "trailers" unless exists $hdr{te}; #1.1
		656
		657	my %state = (connect_guard => 1);
		658
		659	my $ae_error = 595; # connecting
		660
		661	# handle actual, non-tunneled, request
		662	my $handle_actual_request = sub {
		663	$ae_error = 596; # request phase
		664
		665	$state{handle}->starttls ("connect") if $uscheme eq "https" && !exists $state{handle}{tls};
		666
		667	# send request
		668	$state{handle}->push_write (
		669	"$method $rpath HTTP/1.1\015\012"
		670	. (join "", map "\u$_: $hdr{$_}\015\012", grep defined $hdr{$_}, keys %hdr)
		671	. "\015\012"
		672	. (delete $arg{body})
		673	);
		674
		675	# return if error occured during push_write()
		676	return unless %state;
		677
		678	%hdr = (); # reduce memory usage, save a kitten, also make it possible to re-use
		679
		680	# status line and headers
		681	$state{read_response} = sub {
		682	for ("$_[1]") {
		683	y/\015//d; # weed out any \015, as they show up in the weirdest of places.
		684
		685	/^HTTP\/0([0-9\.]+) \s+ ([0-9]{3}) (?: \s+ ([^\012]) )? \012/gxci
		686	or return (%state = (), $cb->(undef, { @pseudo, Status => 599, Reason => "Invalid server response" }));
		687
		688	# 100 Continue handling
		689	# should not happen as we don't send expect: 100-continue,
		690	# but we handle it just in case.
		691	# since we send the request body regardless, if we get an error
		692	# we are out of-sync, which we currently do NOT handle correctly.
		693	return $state{handle}->push_read (line => $qr_nlnl, $state{read_response})
		694	if $2 eq 100;
		695
		696	push @pseudo,
		697	HTTPVersion => $1,
		698	Status => $2,
		699	Reason => $3,
		700	;
		701
		702	my $hdr = parse_hdr
		703	or return (%state = (), $cb->(undef, { @pseudo, Status => 599, Reason => "Garbled response headers" }));
		704
		705	%hdr = (%$hdr, @pseudo);
		706	}
		707
		708	# redirect handling
		709	# microsoft and other shitheads don't give a shit for following standards,
		710	# try to support some common forms of broken Location headers.
		711	if ($hdr{location} !~ /^(?: $ \| [^:\/?\#]+ : )/x) {
		712	$hdr{location} =~ s/^\.\/+//;
		713
		714	my $url = "$rscheme://$uhost:$uport";
		715
		716	unless ($hdr{location} =~ s/^\///) {
		717	$url .= $upath;
		718	$url =~ s/\/[^\/]*$//;
		719	}
		720
		721	$hdr{location} = "$url/$hdr{location}";
		722	}
		723
		724	my $redirect;
		725
		726	if ($recurse) {
		727	my $status = $hdr{Status};
		728
		729	# industry standard is to redirect POST as GET for
		730	# 301, 302 and 303, in contrast to HTTP/1.0 and 1.1.
		731	# also, the UA should ask the user for 301 and 307 and POST,
		732	# industry standard seems to be to simply follow.
		733	# we go with the industry standard.
		734	if ($status == 301 or $status == 302 or $status == 303) {
		735	# HTTP/1.1 is unclear on how to mutate the method
		736	$method = "GET" unless $method eq "HEAD";
		737	$redirect = 1;
		738	} elsif ($status == 307) {
		739	$redirect = 1;
		740	}
		741	}
		742
		743	my $finish = sub { # ($data, $err_status, $err_reason[, $keepalive])
		744	my $may_keep_alive = $_[3];
		745
		746	$state{handle}->destroy if $state{handle};
		747	%state = ();
		748
		749	if (defined $_[1]) {
		750	$hdr{OrigStatus} = $hdr{Status}; $hdr{Status} = $_[1];
		751	$hdr{OrigReason} = $hdr{Reason}; $hdr{Reason} = $_[2];
		752	}
		753
		754	# set-cookie processing
		755	if ($arg{cookie_jar}) {
		756	cookie_jar_set_cookie $arg{cookie_jar}, $hdr{"set-cookie"}, $uhost, $hdr{date};
		757	}
		758
		759	if ($redirect && exists $hdr{location}) {
		760	# we ignore any errors, as it is very common to receive
		761	# Content-Length != 0 but no actual body
		762	# we also access %hdr, as $_[1] might be an erro
		763	http_request (
		764	$method => $hdr{location},
		765	%arg,
		766	recurse => $recurse - 1,
		767	Redirect => [$_[0], \%hdr],
		768	$cb);
		769	} else {
		770	$cb->($_[0], \%hdr);
		771	}
		772	};
		773
		774	$ae_error = 597; # body phase
		775
		776	my $len = $hdr{"content-length"};
		777
		778	# body handling, many different code paths
		779	# - no body expected
		780	# - want_body_handle
		781	# - te chunked
		782	# - 2x length known (with or without on_body)
		783	# - 2x length not known (with or without on_body)
		784	if (!$redirect && $arg{on_header} && !$arg{on_header}(\%hdr)) {
		785	$finish->(undef, 598 => "Request cancelled by on_header");
		786	} elsif (
		787	$hdr{Status} =~ /^(?:1..\|204\|205\|304)$/
		788	or $method eq "HEAD"
		789	or (defined $len && $len == 0) # == 0, not !, because "0 " is true
		790	) {
		791	# no body
		792	$finish->("", undef, undef, 1);
		793
		794	} elsif (!$redirect && $arg{want_body_handle}) {
		795	$_[0]->on_eof (undef);
		796	$_[0]->on_error (undef);
		797	$_[0]->on_read (undef);
		798
		799	$finish->(delete $state{handle});
		800
		801	} elsif ($hdr{"transfer-encoding"} =~ /\bchunked\b/i) {
		802	my $cl = 0;
		803	my $body = undef;
		804	my $on_body = $arg{on_body} \|\| sub { $body .= shift; 1 };
		805
		806	$state{read_chunk} = sub {
		807	$_[1] =~ /^([0-9a-fA-F]+)/
		808	or $finish->(undef, $ae_error => "Garbled chunked transfer encoding");
		809
		810	my $len = hex $1;
		811
		812	if ($len) {
		813	$cl += $len;
		814
		815	$_[0]->push_read (chunk => $len, sub {
		816	$on_body->($_[1], \%hdr)
		817	or return $finish->(undef, 598 => "Request cancelled by on_body");
		818
		819	$_[0]->push_read (line => sub {
		820	length $_[1]
		821	and return $finish->(undef, $ae_error => "Garbled chunked transfer encoding");
		822	$_[0]->push_read (line => $state{read_chunk});
		823	});
		824	});
		825	} else {
		826	$hdr{"content-length"} \|\|= $cl;
		827
		828	$_[0]->push_read (line => $qr_nlnl, sub {
		829	if (length $_[1]) {
		830	for ("$_[1]") {
		831	y/\015//d; # weed out any \015, as they show up in the weirdest of places.
		832
		833	my $hdr = parse_hdr
		834	or return $finish->(undef, $ae_error => "Garbled response trailers");
		835
		836	%hdr = (%hdr, %$hdr);
		837	}
		838	}
		839
		840	$finish->($body, undef, undef, 1);
		841	});
		842	}
		843	};
		844
		845	$_[0]->push_read (line => $state{read_chunk});
		846
		847	} elsif ($arg{on_body}) {
		848	if (defined $len) {
		849	$_[0]->on_read (sub {
		850	$len -= length $_[0]{rbuf};
		851
		852	$arg{on_body}(delete $_[0]{rbuf}, \%hdr)
		853	or return $finish->(undef, 598 => "Request cancelled by on_body");
		854
		855	$len > 0
		856	or $finish->("", undef, undef, 1);
		857	});
		858	} else {
		859	$_[0]->on_eof (sub {
		860	$finish->("");
		861	});
		862	$_[0]->on_read (sub {
		863	$arg{on_body}(delete $_[0]{rbuf}, \%hdr)
		864	or $finish->(undef, 598 => "Request cancelled by on_body");
		865	});
		866	}
		867	} else {
		868	$_[0]->on_eof (undef);
		869
		870	if (defined $len) {
		871	$_[0]->on_read (sub {
		872	$finish->((substr delete $_[0]{rbuf}, 0, $len, ""), undef, undef, 1)
		873	if $len <= length $_[0]{rbuf};
		874	});
		875	} else {
		876	$_[0]->on_error (sub {
		877	($! == Errno::EPIPE \|\| !$!)
		878	? $finish->(delete $_[0]{rbuf})
		879	: $finish->(undef, $ae_error => $_[2]);
		880	});
		881	$_[0]->on_read (sub { });
		882	}
		883	}
		884	};
		885
		886	$state{handle}->push_read (line => $qr_nlnl, $state{read_response});
		887	};
		888
		889	my $connect_cb = sub {
162	$state{fh} = shift	890	$state{fh} = shift
		891	or do {
		892	my $err = "$!";
		893	%state = ();
163	or return $cb->(undef, { Status => 599, Reason => "$!" });	894	return $cb->(undef, { @pseudo, Status => $ae_error, Reason => $err });
		895	};
164		896
165	delete $state{connect_guard}; # reduce memory usage, save a tree	897	return unless delete $state{connect_guard};
166		898
167	# get handle	899	# get handle
168	$state{handle} = new AnyEvent::Handle	900	$state{handle} = new AnyEvent::Handle
169	fh => $state{fh},	901	fh => $state{fh},
170	($scheme eq "https" ? (tls => "connect") : ());	902	peername => $rhost,
		903	tls_ctx => $arg{tls_ctx},
		904	# these need to be reconfigured on keepalive handles
		905	timeout => $timeout,
		906	on_error => sub {
		907	%state = ();
		908	$cb->(undef, { @pseudo, Status => $ae_error, Reason => $_[2] });
		909	},
		910	on_eof => sub {
		911	%state = ();
		912	$cb->(undef, { @pseudo, Status => $ae_error, Reason => "Unexpected end-of-file" });
		913	},
		914	;
171		915
172	# limit the number of persistent connections	916	# limit the number of persistent connections
		917	# keepalive not yet supported
173	if ($KA_COUNT{$_[1]} < $MAX_PERSISTENT_PER_HOST) {	918	# if ($KA_COUNT{$_[1]} < $MAX_PERSISTENT_PER_HOST) {
174	++$KA_COUNT{$_[1]};	919	# ++$KA_COUNT{$_[1]};
175	$state{handle}{ka_count_guard} = AnyEvent::Util::guard { --$KA_COUNT{$_[1]} };	920	# $state{handle}{ka_count_guard} = AnyEvent::Util::guard {
		921	# --$KA_COUNT{$_[1]}
		922	# };
176	$hdr{connection} = "keep-alive";	923	# $hdr{connection} = "keep-alive";
177	delete $hdr{connection}; # keep-alive not yet supported
178	} else {
179	delete $hdr{connection};
180	}	924	# }
181		925
182	# (re-)configure handle	926	$state{handle}->starttls ("connect") if $rscheme eq "https";
183	$state{handle}->timeout ($timeout);
184	$state{handle}->on_error (sub {
185	%state = ();
186	$cb->(undef, { Status => 599, Reason => "$!" });
187	});
188	$state{handle}->on_eof (sub {
189	%state = ();
190	$cb->(undef, { Status => 599, Reason => "unexpected end-of-file" });
191	});
192		927
193	# send request	928	# now handle proxy-CONNECT method
194	$state{handle}->push_write (	929	if ($proxy && $uscheme eq "https") {
195	"\U$method\E $path HTTP/1.0\015\012"	930	# oh dear, we have to wrap it into a connect request
196	. (join "", map "$_: $hdr{$_}\015\012", keys %hdr)
197	. "\015\012"
198	. (delete $state{body})
199	);
200		931
201	%hdr = (); # reduce memory usage, save a kitten	932	# maybe re-use $uauthority with patched port?
202		933	$state{handle}->push_write ("CONNECT $uhost:$uport HTTP/1.0\015\012\015\012");
203	# status line
204	$state{handle}->push_read (line => qr/\015?\012/, sub {	934	$state{handle}->push_read (line => $qr_nlnl, sub {
205	$_[1] =~ /^HTTP\/([0-9\.]+) \s+ ([0-9]{3}) \s+ ([^\015\012]+)/ix	935	$_[1] =~ /^HTTP\/([0-9\.]+) \s+ ([0-9]{3}) (?: \s+ ([^\015\012]*) )?/ix
206	or return (%state = (), $cb->(undef, { Status => 599, Reason => "invalid server response ($_[1])" }));	936	or return (%state = (), $cb->(undef, { @pseudo, Status => 599, Reason => "Invalid proxy connect response ($_[1])" }));
207		937
208	my %hdr = ( # response headers	938	if ($2 == 200) {
209	HTTPVersion => ",$1",	939	$rpath = $upath;
210	Status => ",$2",	940	$handle_actual_request->();
211	Reason => ",$3",
212	);
213
214	# headers, could be optimized a bit
215	$state{handle}->unshift_read (line => qr/\015?\012\015?\012/, sub {
216	for ("$_[1]\012") {
217	# we support spaces in field names, as lotus domino
218	# creates them.
219	$hdr{lc $1} .= ",$2"
220	while /\G
221	([^:\000-\037]+):
222	[\011\040]*
223	((?: [^\015\012]+ \| \015?\012[\011\040] )*)
224	\015?\012
225	/gxc;
226
227	/\G$/
228	or return $cb->(undef, { Status => 599, Reason => "garbled response headers" });
229	}
230
231	substr $_, 0, 1, ""
232	for values %hdr;
233
234	if (exists $hdr{"content-length"}) {
235	$_[0]->unshift_read (chunk => $hdr{"content-length"}, sub {
236	# could cache persistent connection now
237	if ($hdr{connection} =~ /\bkeep-alive\b/i) {
238	# but we don't, due to misdesigns, this is annoyingly complex
239	};
240
241	%state = ();
242	$cb->($_[1], \%hdr);
243	});
244	} else {	941	} else {
245	# too bad, need to read until we get an error or EOF,
246	# no way to detect winged data.
247	$_[0]->on_error (sub {
248	%state = ();	942	%state = ();
249	$cb->($_[0]{rbuf}, \%hdr);	943	$cb->(undef, { @pseudo, Status => $2, Reason => $3 });
250	});
251	$_[0]->on_eof (undef);
252	$_[0]->on_read (sub { });
253	}	944	}
254	});	945	});
		946	} else {
		947	$handle_actual_request->();
255	});	948	}
256	}, sub {	949	};
257	$timeout	950
		951	_get_slot $uhost, sub {
		952	$state{slot_guard} = shift;
		953
		954	return unless $state{connect_guard};
		955
		956	my $tcp_connect = $arg{tcp_connect}
		957	\|\| do { require AnyEvent::Socket; \&AnyEvent::Socket::tcp_connect };
		958
		959	$state{connect_guard} = $tcp_connect->($rhost, $rport, $connect_cb, $arg{on_prepare} \|\| sub { $timeout });
258	};	960	};
259		961
260	defined wantarray && AnyEvent::Util::guard { %state = () }	962	defined wantarray && AnyEvent::Util::guard { %state = () }
261	}	963	}
262		964
263	sub http_get($$;@) {	965	sub http_get($@) {
264	unshift @_, "GET";	966	unshift @_, "GET";
265	&http_request	967	&http_request
266	}	968	}
267		969
		970	sub http_head($@) {
		971	unshift @_, "HEAD";
		972	&http_request
		973	}
		974
		975	sub http_post($$@) {
		976	my $url = shift;
		977	unshift @_, "POST", $url, "body";
		978	&http_request
		979	}
		980
		981	=back
		982
		983	=head2 DNS CACHING
		984
		985	AnyEvent::HTTP uses the AnyEvent::Socket::tcp_connect function for
		986	the actual connection, which in turn uses AnyEvent::DNS to resolve
		987	hostnames. The latter is a simple stub resolver and does no caching
		988	on its own. If you want DNS caching, you currently have to provide
		989	your own default resolver (by storing a suitable resolver object in
		990	C<$AnyEvent::DNS::RESOLVER>).
		991
268	=head2 GLOBAL FUNCTIONS AND VARIABLES	992	=head2 GLOBAL FUNCTIONS AND VARIABLES
269		993
270	=over 4	994	=over 4
271		995
272	=item AnyEvent::HTTP::set_proxy "proxy-url"	996	=item AnyEvent::HTTP::set_proxy "proxy-url"
273		997
274	Sets the default proxy server to use. The proxy-url must begin with a	998	Sets the default proxy server to use. The proxy-url must begin with a
275	string of the form C<http://host:port> (optionally C<https:...>).	999	string of the form C<http://host:port> (optionally C<https:...>), croaks
		1000	otherwise.
276		1001
		1002	To clear an already-set proxy, use C<undef>.
		1003
		1004	=item AnyEvent::HTTP::cookie_jar_expire $jar[, $session_end]
		1005
		1006	Remove all cookies from the cookie jar that have been expired. If
		1007	C<$session_end> is given and true, then additionally remove all session
		1008	cookies.
		1009
		1010	You should call this function (with a true C<$session_end>) before you
		1011	save cookies to disk, and you should call this function after loading them
		1012	again. If you have a long-running program you can additonally call this
		1013	function from time to time.
		1014
		1015	A cookie jar is initially an empty hash-reference that is managed by this
		1016	module. It's format is subject to change, but currently it is like this:
		1017
		1018	The key C<version> has to contain C<1>, otherwise the hash gets
		1019	emptied. All other keys are hostnames or IP addresses pointing to
		1020	hash-references. The key for these inner hash references is the
		1021	server path for which this cookie is meant, and the values are again
		1022	hash-references. The keys of those hash-references is the cookie name, and
		1023	the value, you guessed it, is another hash-reference, this time with the
		1024	key-value pairs from the cookie, except for C<expires> and C<max-age>,
		1025	which have been replaced by a C<_expires> key that contains the cookie
		1026	expiry timestamp.
		1027
		1028	Here is an example of a cookie jar with a single cookie, so you have a
		1029	chance of understanding the above paragraph:
		1030
		1031	{
		1032	version => 1,
		1033	"10.0.0.1" => {
		1034	"/" => {
		1035	"mythweb_id" => {
		1036	_expires => 1293917923,
		1037	value => "ooRung9dThee3ooyXooM1Ohm",
		1038	},
		1039	},
		1040	},
		1041	}
		1042
		1043	=item $date = AnyEvent::HTTP::format_date $timestamp
		1044
		1045	Takes a POSIX timestamp (seconds since the epoch) and formats it as a HTTP
		1046	Date (RFC 2616).
		1047
		1048	=item $timestamp = AnyEvent::HTTP::parse_date $date
		1049
		1050	Takes a HTTP Date (RFC 2616) or a Cookie date (netscape cookie spec) or a
		1051	bunch of minor variations of those, and returns the corresponding POSIX
		1052	timestamp, or C<undef> if the date cannot be parsed.
		1053
277	=item $AnyEvent::HTTP::MAX_REDIRECTS	1054	=item $AnyEvent::HTTP::MAX_RECURSE
278		1055
279	The default value for the C<max_redirects> request parameter	1056	The default value for the C<recurse> request parameter (default: C<10>).
280	(default: C<10>).
281		1057
282	=item $AnyEvent::HTTP::USERAGENT	1058	=item $AnyEvent::HTTP::USERAGENT
283		1059
284	The default value for the C<User-Agent> header (the default is	1060	The default value for the C<User-Agent> header (the default is
285	C<Mozilla/5.0 (compatible; AnyEvent::HTTP/$VERSION; +http://software.schmorp.de/pkg/AnyEvent)>).	1061	C<Mozilla/5.0 (compatible; U; AnyEvent-HTTP/$VERSION; +http://software.schmorp.de/pkg/AnyEvent)>).
286		1062
287	=item $AnyEvent::HTTP::MAX_PERSISTENT	1063	=item $AnyEvent::HTTP::MAX_PER_HOST
288		1064
289	The maximum number of persistent connections to keep open (default: 8).	1065	The maximum number of concurrent connections to the same host (identified
		1066	by the hostname). If the limit is exceeded, then the additional requests
		1067	are queued until previous connections are closed.
290		1068
291	=item $AnyEvent::HTTP::PERSISTENT_TIMEOUT	1069	The default value for this is C<4>, and it is highly advisable to not
		1070	increase it.
292		1071
293	The maximum time to cache a persistent connection, in seconds (default: 2).	1072	=item $AnyEvent::HTTP::ACTIVE
		1073
		1074	The number of active connections. This is not the number of currently
		1075	running requests, but the number of currently open and non-idle TCP
		1076	connections. This number of can be useful for load-leveling.
294		1077
295	=back	1078	=back
296		1079
297	=cut	1080	=cut
298		1081
		1082	our @month = qw(Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec);
		1083	our @weekday = qw(Sun Mon Tue Wed Thu Fri Sat);
		1084
		1085	sub format_date($) {
		1086	my ($time) = @_;
		1087
		1088	# RFC 822/1123 format
		1089	my ($S, $M, $H, $mday, $mon, $year, $wday, $yday, undef) = gmtime $time;
		1090
		1091	sprintf "%s, %02d %s %04d %02d:%02d:%02d GMT",
		1092	$weekday[$wday], $mday, $month[$mon], $year + 1900,
		1093	$H, $M, $S;
		1094	}
		1095
		1096	sub parse_date($) {
		1097	my ($date) = @_;
		1098
		1099	my ($d, $m, $y, $H, $M, $S);
		1100
		1101	if ($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$/) {
		1102	# RFC 822/1123, required by RFC 2616 (with " ")
		1103	# cookie dates (with "-")
		1104
		1105	($d, $m, $y, $H, $M, $S) = ($1, $2, $3, $4, $5, $6);
		1106
		1107	} elsif ($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$/) {
		1108	# RFC 850
		1109	($d, $m, $y, $H, $M, $S) = ($1, $2, $3 < 69 ? $3 + 2000 : $3 + 1900, $4, $5, $6);
		1110
		1111	} elsif ($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])$/) {
		1112	# ISO C's asctime
		1113	($d, $m, $y, $H, $M, $S) = ($2, $1, $6, $3, $4, $5);
		1114	}
		1115	# other formats fail in the loop below
		1116
		1117	for (0..11) {
		1118	if ($m eq $month[$_]) {
		1119	require Time::Local;
		1120	return Time::Local::timegm ($S, $M, $H, $d, $_, $y);
		1121	}
		1122	}
		1123
		1124	undef
		1125	}
		1126
299	sub set_proxy($) {	1127	sub set_proxy($) {
		1128	if (length $_[0]) {
300	$PROXY = [$2, $3 \|\| 3128, $1] if $_[0] =~ m%^(https?):// ([^:/]+) (?: : (\d*) )?%ix;	1129	$_[0] =~ m%^(https?):// ([^:/]+) (?: : (\d*) )?%ix
		1130	or Carp::croak "$_[0]: invalid proxy URL";
		1131	$PROXY = [$2, $3 \|\| 3128, $1]
		1132	} else {
		1133	undef $PROXY;
		1134	}
301	}	1135	}
302		1136
303	# initialise proxy from environment	1137	# initialise proxy from environment
		1138	eval {
304	set_proxy $ENV{http_proxy};	1139	set_proxy $ENV{http_proxy};
		1140	};
		1141
		1142	=head2 SOCKS PROXIES
		1143
		1144	Socks proxies are not directly supported by AnyEvent::HTTP. You can
		1145	compile your perl to support socks, or use an external program such as
		1146	F<socksify> (dante) or F<tsocks> to make your program use a socks proxy
		1147	transparently.
		1148
		1149	Alternatively, for AnyEvent::HTTP only, you can use your own
		1150	C<tcp_connect> function that does the proxy handshake - here is an example
		1151	that works with socks4a proxies:
		1152
		1153	use Errno;
		1154	use AnyEvent::Util;
		1155	use AnyEvent::Socket;
		1156	use AnyEvent::Handle;
		1157
		1158	# host, port and username of/for your socks4a proxy
		1159	my $socks_host = "10.0.0.23";
		1160	my $socks_port = 9050;
		1161	my $socks_user = "";
		1162
		1163	sub socks4a_connect {
		1164	my ($host, $port, $connect_cb, $prepare_cb) = @_;
		1165
		1166	my $hdl = new AnyEvent::Handle
		1167	connect => [$socks_host, $socks_port],
		1168	on_prepare => sub { $prepare_cb->($_[0]{fh}) },
		1169	on_error => sub { $connect_cb->() },
		1170	;
		1171
		1172	$hdl->push_write (pack "CCnNZZ", 4, 1, $port, 1, $socks_user, $host);
		1173
		1174	$hdl->push_read (chunk => 8, sub {
		1175	my ($hdl, $chunk) = @_;
		1176	my ($status, $port, $ipn) = unpack "xCna4", $chunk;
		1177
		1178	if ($status == 0x5a) {
		1179	$connect_cb->($hdl->{fh}, (format_address $ipn) . ":$port");
		1180	} else {
		1181	$! = Errno::ENXIO; $connect_cb->();
		1182	}
		1183	});
		1184
		1185	$hdl
		1186	}
		1187
		1188	Use C<socks4a_connect> instead of C<tcp_connect> when doing C<http_request>s,
		1189	possibly after switching off other proxy types:
		1190
		1191	AnyEvent::HTTP::set_proxy undef; # usually you do not want other proxies
		1192
		1193	http_get 'http://www.google.com', tcp_connect => \&socks4a_connect, sub {
		1194	my ($data, $headers) = @_;
		1195	...
		1196	};
305		1197
306	=head1 SEE ALSO	1198	=head1 SEE ALSO
307		1199
308	L<AnyEvent>.	1200	L<AnyEvent>.
309		1201
310	=head1 AUTHOR	1202	=head1 AUTHOR
311		1203
312	Marc Lehmann <schmorp@schmorp.de>	1204	Marc Lehmann <schmorp@schmorp.de>
313	http://home.schmorp.de/	1205	http://home.schmorp.de/
		1206
		1207	With many thanks to Дмитрий Шалашов, who provided countless
		1208	testcases and bugreports.
314		1209
315	=cut	1210	=cut
316		1211
317	1	1212	1
318		1213

Diff Legend

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

Comparing AnyEvent-HTTP/HTTP.pm (file contents): Revision 1.2 by root, Wed Jun 4 11:37:41 2008 UTC vs. Revision 1.88 by root, Sun Jan 2 20:57:03 2011 UTC

Diff Legend

Comparing AnyEvent-HTTP/HTTP.pm (file contents):
Revision 1.2 by root, Wed Jun 4 11:37:41 2008 UTC vs.
Revision 1.88 by root, Sun Jan 2 20:57:03 2011 UTC