[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 1.70 by root, Fri Dec 31 20:31:47 2010 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<< $headers->{Status} >> will be C<59x>
100	and the C<Reason> pseudo-header will contain an error message.	128	(usually C<599>) and the C<Reason> pseudo-header will contain an error
		129	message.
101		130
102	A typical callback might look like this:	131	A typical callback might look like this:
103		132
104	sub {	133	sub {
105	my ($body, $hdr) = @_;	134	my ($body, $hdr) = @_;
…		…
121	Whether to recurse requests or not, e.g. on redirects, authentication	150	Whether to recurse requests or not, e.g. on redirects, authentication
122	retries and so on, and how often to do so.	151	retries and so on, and how often to do so.
123		152
124	=item headers => hashref	153	=item headers => hashref
125		154
126	The request headers to use.	155	The request headers to use. Currently, C<http_request> may provide its own
		156	C<Host:>, C<Content-Length:>, C<Connection:> and C<Cookie:> headers and
		157	will provide defaults at least for C<TE:>, C<Referer:> and C<User-Agent:>
		158	(this can be suppressed by using C<undef> for these headers in which case
		159	they won't be sent at all).
127		160
128	=item timeout => $seconds	161	=item timeout => $seconds
129		162
130	The time-out to use for various stages - each connect attempt will reset	163	The time-out to use for various stages - each connect attempt will reset
131	the timeout, as will read or write activity. Default timeout is 5 minutes.	164	the timeout, as will read or write activity, i.e. this is not an overall
		165	timeout.
		166
		167	Default timeout is 5 minutes.
132		168
133	=item proxy => [$host, $port[, $scheme]] or undef	169	=item proxy => [$host, $port[, $scheme]] or undef
134		170
135	Use the given http proxy for all requests. If not specified, then the	171	Use the given http proxy for all requests. If not specified, then the
136	default proxy (as specified by C<$ENV{http_proxy}>) is used.	172	default proxy (as specified by C<$ENV{http_proxy}>) is used.
137		173
138	C<$scheme> must be either missing or C<http> for HTTP, or C<https> for	174	C<$scheme> must be either missing, C<http> for HTTP or C<https> for
139	HTTPS.	175	HTTPS.
140		176
141	=item body => $string	177	=item body => $string
142		178
143	The request body, usually empty. Will be-sent as-is (future versions of	179	The request body, usually empty. Will be sent as-is (future versions of
144	this module might offer more options).	180	this module might offer more options).
145		181
146	=item cookie_jar => $hash_ref	182	=item cookie_jar => $hash_ref
147		183
148	Passing this parameter enables (simplified) cookie-processing, loosely	184	Passing this parameter enables (simplified) cookie-processing, loosely
149	based on the original netscape specification.	185	based on the original netscape specification.
150		186
151	The C<$hash_ref> must be an (initially empty) hash reference which will	187	The C<$hash_ref> must be an (initially empty) hash reference which will
152	get updated automatically. It is possible to save the cookie_jar to	188	get updated automatically. It is possible to save the cookie jar to
153	persistent storage with something like JSON or Storable, but this is not	189	persistent storage with something like JSON or Storable, but this is not
154	recommended, as expire times are currently being ignored.	190	recommended, as session-only cookies might survive longer than expected.
155		191
156	Note that this cookie implementation is not of very high quality, nor	192	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	193	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	194	own. C<cookie_jar> is meant as a quick fix to get some cookie-using sites
159	cookie-using sites working. Cookies are a privacy disaster, do not use	195	working. Cookies are a privacy disaster, do not use them unless required
160	them unless required to.	196	to.
		197
		198	When cookie processing is enabled, the C<Cookie:> and C<Set-Cookie:>
		199	headers will be set and handled by this module, otherwise they will be
		200	left untouched.
		201
		202	=item tls_ctx => $scheme \| $tls_ctx
		203
		204	Specifies the AnyEvent::TLS context to be used for https connections. This
		205	parameter follows the same rules as the C<tls_ctx> parameter to
		206	L<AnyEvent::Handle>, but additionally, the two strings C<low> or
		207	C<high> can be specified, which give you a predefined low-security (no
		208	verification, highest compatibility) and high-security (CA and common-name
		209	verification) TLS context.
		210
		211	The default for this option is C<low>, which could be interpreted as "give
		212	me the page, no matter what".
		213
		214	=item on_prepare => $callback->($fh)
		215
		216	In rare cases you need to "tune" the socket before it is used to
		217	connect (for exmaple, to bind it on a given IP address). This parameter
		218	overrides the prepare callback passed to C<AnyEvent::Socket::tcp_connect>
		219	and behaves exactly the same way (e.g. it has to provide a
		220	timeout). See the description for the C<$prepare_cb> argument of
		221	C<AnyEvent::Socket::tcp_connect> for details.
		222
		223	=item tcp_connect => $callback->($host, $service, $connect_cb, $prepare_cb)
		224
		225	In even rarer cases you want total control over how AnyEvent::HTTP
		226	establishes connections. Normally it uses L<AnyEvent::Socket::tcp_connect>
		227	to do this, but you can provide your own C<tcp_connect> function -
		228	obviously, it has to follow the same calling conventions, except that it
		229	may always return a connection guard object.
		230
		231	There are probably lots of weird uses for this function, starting from
		232	tracing the hosts C<http_request> actually tries to connect, to (inexact
		233	but fast) host => IP address caching or even socks protocol support.
		234
		235	=item on_header => $callback->($headers)
		236
		237	When specified, this callback will be called with the header hash as soon
		238	as headers have been successfully received from the remote server (not on
		239	locally-generated errors).
		240
		241	It has to return either true (in which case AnyEvent::HTTP will continue),
		242	or false, in which case AnyEvent::HTTP will cancel the download (and call
		243	the finish callback with an error code of C<598>).
		244
		245	This callback is useful, among other things, to quickly reject unwanted
		246	content, which, if it is supposed to be rare, can be faster than first
		247	doing a C<HEAD> request.
		248
		249	The downside is that cancelling the request makes it impossible to re-use
		250	the connection. Also, the C<on_header> callback will not receive any
		251	trailer (headers sent after the response body).
		252
		253	Example: cancel the request unless the content-type is "text/html".
		254
		255	on_header => sub {
		256	$_[0]{"content-type"} =~ /^text\/html\s*(?:;\|$)/
		257	},
		258
		259	=item on_body => $callback->($partial_body, $headers)
		260
		261	When specified, all body data will be passed to this callback instead of
		262	to the completion callback. The completion callback will get the empty
		263	string instead of the body data.
		264
		265	It has to return either true (in which case AnyEvent::HTTP will continue),
		266	or false, in which case AnyEvent::HTTP will cancel the download (and call
		267	the completion callback with an error code of C<598>).
		268
		269	The downside to cancelling the request is that it makes it impossible to
		270	re-use the connection.
		271
		272	This callback is useful when the data is too large to be held in memory
		273	(so the callback writes it to a file) or when only some information should
		274	be extracted, or when the body should be processed incrementally.
		275
		276	It is usually preferred over doing your own body handling via
		277	C<want_body_handle>, but in case of streaming APIs, where HTTP is
		278	only used to create a connection, C<want_body_handle> is the better
		279	alternative, as it allows you to install your own event handler, reducing
		280	resource usage.
		281
		282	=item want_body_handle => $enable
		283
		284	When enabled (default is disabled), the behaviour of AnyEvent::HTTP
		285	changes considerably: after parsing the headers, and instead of
		286	downloading the body (if any), the completion callback will be
		287	called. Instead of the C<$body> argument containing the body data, the
		288	callback will receive the L<AnyEvent::Handle> object associated with the
		289	connection. In error cases, C<undef> will be passed. When there is no body
		290	(e.g. status C<304>), the empty string will be passed.
		291
		292	The handle object might or might not be in TLS mode, might be connected to
		293	a proxy, be a persistent connection etc., and configured in unspecified
		294	ways. The user is responsible for this handle (it will not be used by this
		295	module anymore).
		296
		297	This is useful with some push-type services, where, after the initial
		298	headers, an interactive protocol is used (typical example would be the
		299	push-style twitter API which starts a JSON/XML stream).
		300
		301	If you think you need this, first have a look at C<on_body>, to see if
		302	that doesn't solve your problem in a better way.
161		303
162	=back	304	=back
163		305
164	Example: make a simple HTTP GET request for http://www.nethype.de/	306	Example: do a simple HTTP GET request for http://www.nethype.de/ and print
		307	the response body.
165		308
166	http_request GET => "http://www.nethype.de/", sub {	309	http_request GET => "http://www.nethype.de/", sub {
167	my ($body, $hdr) = @_;	310	my ($body, $hdr) = @_;
168	print "$body\n";	311	print "$body\n";
169	};	312	};
170		313
171	Example: make a HTTP HEAD request on https://www.google.com/, use a	314	Example: do a HTTP HEAD request on https://www.google.com/, use a
172	timeout of 30 seconds.	315	timeout of 30 seconds.
173		316
174	http_request	317	http_request
175	GET => "https://www.google.com",	318	GET => "https://www.google.com",
176	timeout => 30,	319	timeout => 30,
…		…
179	use Data::Dumper;	322	use Data::Dumper;
180	print Dumper $hdr;	323	print Dumper $hdr;
181	}	324	}
182	;	325	;
183		326
		327	Example: do another simple HTTP GET request, but immediately try to
		328	cancel it.
		329
		330	my $request = http_request GET => "http://www.nethype.de/", sub {
		331	my ($body, $hdr) = @_;
		332	print "$body\n";
		333	};
		334
		335	undef $request;
		336
184	=cut	337	=cut
185		338
		339	sub _slot_schedule;
186	sub _slot_schedule($) {	340	sub _slot_schedule($) {
187	my $host = shift;	341	my $host = shift;
188		342
189	while ($CO_SLOT{$host}[0] < $MAX_PER_HOST) {	343	while ($CO_SLOT{$host}[0] < $MAX_PER_HOST) {
190	if (my $cb = shift @{ $CO_SLOT{$host}[1] }) {	344	if (my $cb = shift @{ $CO_SLOT{$host}[1] }) {
191	# somebody wnats that slot	345	# somebody wants that slot
192	++$CO_SLOT{$host}[0];	346	++$CO_SLOT{$host}[0];
		347	++$ACTIVE;
193		348
194	$cb->(AnyEvent::Util::guard {	349	$cb->(AnyEvent::Util::guard {
		350	--$ACTIVE;
195	--$CO_SLOT{$host}[0];	351	--$CO_SLOT{$host}[0];
196	_slot_schedule $host;	352	_slot_schedule $host;
197	});	353	});
198	} else {	354	} else {
199	# nobody wants the slot, maybe we can forget about it	355	# nobody wants the slot, maybe we can forget about it
200	delete $CO_SLOT{$host} unless $CO_SLOT{$host}[0];	356	delete $CO_SLOT{$host} unless $CO_SLOT{$host}[0];
201	warn "$host deleted" unless $CO_SLOT{$host}[0];#d#
202	last;	357	last;
203	}	358	}
204	}	359	}
205	}	360	}
206		361
…		…
209	push @{ $CO_SLOT{$_[0]}[1] }, $_[1];	364	push @{ $CO_SLOT{$_[0]}[1] }, $_[1];
210		365
211	_slot_schedule $_[0];	366	_slot_schedule $_[0];
212	}	367	}
213		368
		369	# continue to parse $_ for headers and place them into the arg
		370	sub parse_hdr() {
		371	my %hdr;
		372
		373	# things seen, not parsed:
		374	# p3pP="NON CUR OTPi OUR NOR UNI"
		375
		376	$hdr{lc $1} .= ",$2"
		377	while /\G
		378	([^:\000-\037]*):
		379	[\011\040]*
		380	((?: [^\012]+ \| \012[\011\040] )*)
		381	\012
		382	/gxc;
		383
		384	/\G$/
		385	or return;
		386
		387	# remove the "," prefix we added to all headers above
		388	substr $_, 0, 1, ""
		389	for values %hdr;
		390
		391	\%hdr
		392	}
		393
		394	our $qr_nlnl = qr{(?<![^\012])\015?\012};
		395
		396	our $TLS_CTX_LOW = { cache => 1, sslv2 => 1 };
		397	our $TLS_CTX_HIGH = { cache => 1, verify => 1, verify_peername => "https" };
		398
214	sub http_request($$$;@) {	399	sub http_request($$@) {
215	my $cb = pop;	400	my $cb = pop;
216	my ($method, $url, %arg) = @_;	401	my ($method, $url, %arg) = @_;
217		402
218	my %hdr;	403	my %hdr;
		404
		405	$arg{tls_ctx} = $TLS_CTX_LOW if $arg{tls_ctx} eq "low" \|\| !exists $arg{tls_ctx};
		406	$arg{tls_ctx} = $TLS_CTX_HIGH if $arg{tls_ctx} eq "high";
219		407
220	$method = uc $method;	408	$method = uc $method;
221		409
222	if (my $hdr = $arg{headers}) {	410	if (my $hdr = $arg{headers}) {
223	while (my ($k, $v) = each %$hdr) {	411	while (my ($k, $v) = each %$hdr) {
224	$hdr{lc $k} = $v;	412	$hdr{lc $k} = $v;
225	}	413	}
226	}	414	}
227		415
		416	# pseudo headers for all subsequent responses
		417	my @pseudo = (URL => $url);
		418	push @pseudo, Redirect => delete $arg{Redirect} if exists $arg{Redirect};
		419
228	my $recurse = exists $arg{recurse} ? $arg{recurse} : $MAX_RECURSE;	420	my $recurse = exists $arg{recurse} ? delete $arg{recurse} : $MAX_RECURSE;
229		421
230	return $cb->(undef, { Status => 599, Reason => "recursion limit reached" })	422	return $cb->(undef, { @pseudo, Status => 599, Reason => "Too many redirections" })
231	if $recurse < 0;	423	if $recurse < 0;
232		424
233	my $proxy = $arg{proxy} \|\| $PROXY;	425	my $proxy = $arg{proxy} \|\| $PROXY;
234	my $timeout = $arg{timeout} \|\| $TIMEOUT;	426	my $timeout = $arg{timeout} \|\| $TIMEOUT;
235		427
236	$hdr{"user-agent"} \|\|= $USERAGENT;
237
238	my ($scheme, $authority, $upath, $query, $fragment) =	428	my ($uscheme, $uauthority, $upath, $query, $fragment) =
239	$url =~ m\|(?:([^:/?#]+):)?(?://([^/?#]))?([^?#])(?:\?([^#]))?(?:#(.))?\|;	429	$url =~ m\|(?:([^:/?#]+):)?(?://([^/?#]))?([^?#])(?:(\?[^#]))?(?:#(.))?\|;
240		430
241	$scheme = lc $scheme;	431	$uscheme = lc $uscheme;
242		432
243	my $uport = $scheme eq "http" ? 80	433	my $uport = $uscheme eq "http" ? 80
244	: $scheme eq "https" ? 443	434	: $uscheme eq "https" ? 443
245	: return $cb->(undef, { Status => 599, Reason => "only http and https URL schemes supported" });	435	: return $cb->(undef, { @pseudo, Status => 599, Reason => "Only http and https URL schemes supported" });
246		436
247	$authority =~ /^(?: .*\@ )? ([^\@:]+) (?: : (\d+) )?$/x	437	$uauthority =~ /^(?: .*\@ )? ([^\@:]+) (?: : (\d+) )?$/x
248	or return $cb->(undef, { Status => 599, Reason => "unparsable URL" });	438	or return $cb->(undef, { @pseudo, Status => 599, Reason => "Unparsable URL" });
249		439
250	my $uhost = $1;	440	my $uhost = $1;
251	$uport = $2 if defined $2;	441	$uport = $2 if defined $2;
252		442
		443	$hdr{host} = defined $2 ? "$uhost:$2" : "$uhost"
		444	unless exists $hdr{host};
		445
253	$uhost =~ s/^\[(.*)\]$/$1/;	446	$uhost =~ s/^\[(.*)\]$/$1/;
254	$upath .= "?$query" if length $query;	447	$upath .= $query if length $query;
255		448
256	$upath =~ s%^/?%/%;	449	$upath =~ s%^/?%/%;
257		450
258	# cookie processing	451	# cookie processing
259	if (my $jar = $arg{cookie_jar}) {	452	if (my $jar = $arg{cookie_jar}) {
260	%$jar = () if $jar->{version} < 1;	453	%$jar = () if $jar->{version} != 1;
261		454
262	my @cookie;	455	my @cookie;
263		456
264	while (my ($chost, $v) = each %$jar) {	457	while (my ($chost, $paths) = each %$jar) {
		458	if ($chost =~ /^\./) {
265	next unless $chost eq substr $uhost, -length $chost;	459	next unless $chost eq substr $uhost, -length $chost;
266	next unless $chost =~ /^\./;	460	} elsif ($chost =~ /\./) {
267		461	next unless $chost eq $uhost;
		462	} else {
		463	next;
		464	}
		465
268	while (my ($cpath, $v) = each %$v) {	466	while (my ($cpath, $cookies) = each %$paths) {
269	next unless $cpath eq substr $upath, 0, length $cpath;	467	next unless $cpath eq substr $upath, 0, length $cpath;
270		468
271	while (my ($k, $v) = each %$v) {	469	while (my ($cookie, $kv) = each %$cookies) {
272	next if $scheme ne "https" && exists $v->{secure};	470	next if $uscheme ne "https" && exists $kv->{secure};
		471
		472	if (exists $kv->{expires}) {
		473	if (AE::now > parse_date ($kv->{expires})) {
		474	delete $cookies->{$cookie};
		475	next;
		476	}
		477	}
		478
		479	my $value = $kv->{value};
		480	$value =~ s/([\\"])/\\$1/g;
273	push @cookie, "$k=$v->{value}";	481	push @cookie, "$cookie=\"$value\"";
274	}	482	}
275	}	483	}
276	}	484	}
277		485
278	$hdr{cookie} = join "; ", @cookie	486	$hdr{cookie} = join "; ", @cookie
279	if @cookie;	487	if @cookie;
280	}	488	}
281		489
282	my ($rhost, $rport, $rpath); # request host, port, path	490	my ($rhost, $rport, $rscheme, $rpath); # request host, port, path
283		491
284	if ($proxy) {	492	if ($proxy) {
285	($rhost, $rport, $scheme) = @$proxy;	493	($rpath, $rhost, $rport, $rscheme) = ($url, @$proxy);
286	$rpath = $url;	494
		495	$rscheme = "http" unless defined $rscheme;
		496
		497	# don't support https requests over https-proxy transport,
		498	# can't be done with tls as spec'ed, unless you double-encrypt.
		499	$rscheme = "http" if $uscheme eq "https" && $rscheme eq "https";
287	} else {	500	} else {
288	($rhost, $rport, $rpath) = ($uhost, $uport, $upath);	501	($rhost, $rport, $rscheme, $rpath) = ($uhost, $uport, $uscheme, $upath);
289	$hdr{host} = $uhost;
290	}	502	}
291		503
		504	# leave out fragment and query string, just a heuristic
		505	$hdr{referer} = "$uscheme://$uauthority$upath" unless exists $hdr{referer};
		506	$hdr{"user-agent"} = $USERAGENT unless exists $hdr{"user-agent"};
		507
292	$hdr{"content-length"} = length $arg{body};	508	$hdr{"content-length"} = length $arg{body}
		509	if length $arg{body} \|\| $method ne "GET";
		510
		511	$hdr{connection} = "close TE"; #1.1
		512	$hdr{te} = "trailers" unless exists $hdr{te}; #1.1
293		513
294	my %state = (connect_guard => 1);	514	my %state = (connect_guard => 1);
295		515
296	_get_slot $uhost, sub {	516	_get_slot $uhost, sub {
297	$state{slot_guard} = shift;	517	$state{slot_guard} = shift;
298		518
299	return unless $state{connect_guard};	519	return unless $state{connect_guard};
300		520
301	$state{connect_guard} = AnyEvent::Socket::tcp_connect $rhost, $rport, sub {	521	my $connect_cb = sub {
302	$state{fh} = shift	522	$state{fh} = shift
		523	or do {
		524	my $err = "$!";
		525	%state = ();
303	or return $cb->(undef, { Status => 599, Reason => "$!" });	526	return $cb->(undef, { @pseudo, Status => 599, Reason => $err });
		527	};
304		528
305	delete $state{connect_guard}; # reduce memory usage, save a tree	529	pop; # free memory, save a tree
		530
		531	return unless delete $state{connect_guard};
306		532
307	# get handle	533	# get handle
308	$state{handle} = new AnyEvent::Handle	534	$state{handle} = new AnyEvent::Handle
309	fh => $state{fh},	535	fh => $state{fh},
310	($scheme eq "https" ? (tls => "connect") : ());	536	peername => $rhost,
		537	tls_ctx => $arg{tls_ctx},
		538	# these need to be reconfigured on keepalive handles
		539	timeout => $timeout,
		540	on_error => sub {
		541	%state = ();
		542	$cb->(undef, { @pseudo, Status => 599, Reason => $_[2] });
		543	},
		544	on_eof => sub {
		545	%state = ();
		546	$cb->(undef, { @pseudo, Status => 599, Reason => "Unexpected end-of-file" });
		547	},
		548	;
311		549
312	# limit the number of persistent connections	550	# limit the number of persistent connections
		551	# keepalive not yet supported
313	if ($KA_COUNT{$_[1]} < $MAX_PERSISTENT_PER_HOST) {	552	# if ($KA_COUNT{$_[1]} < $MAX_PERSISTENT_PER_HOST) {
314	++$KA_COUNT{$_[1]};	553	# ++$KA_COUNT{$_[1]};
315	$state{handle}{ka_count_guard} = AnyEvent::Util::guard { --$KA_COUNT{$_[1]} };	554	# $state{handle}{ka_count_guard} = AnyEvent::Util::guard {
		555	# --$KA_COUNT{$_[1]}
		556	# };
316	$hdr{connection} = "keep-alive";	557	# $hdr{connection} = "keep-alive";
317	delete $hdr{connection}; # keep-alive not yet supported
318	} else {
319	delete $hdr{connection};
320	}	558	# }
321		559
322	# (re-)configure handle	560	$state{handle}->starttls ("connect") if $rscheme eq "https";
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		561
		562	# handle actual, non-tunneled, request
		563	my $handle_actual_request = sub {
		564	$state{handle}->starttls ("connect") if $uscheme eq "https" && !exists $state{handle}{tls};
		565
333	# send request	566	# send request
334	$state{handle}->push_write (	567	$state{handle}->push_write (
335	"$method $rpath HTTP/1.0\015\012"	568	"$method $rpath HTTP/1.1\015\012"
336	. (join "", map "$_: $hdr{$_}\015\012", keys %hdr)	569	. (join "", map "\u$_: $hdr{$_}\015\012", grep defined $hdr{$_}, keys %hdr)
337	. "\015\012"	570	. "\015\012"
338	. (delete $arg{body})	571	. (delete $arg{body})
339	);
340
341	%hdr = (); # reduce memory usage, save a kitten
342
343	# status line
344	$state{handle}->push_read (line => qr/\015?\012/, sub {
345	$_[1] =~ /^HTTP\/([0-9\.]+) \s+ ([0-9]{3}) \s+ ([^\015\012]+)/ix
346	or return (%state = (), $cb->(undef, { Status => 599, Reason => "invalid server response ($_[1])" }));
347
348	my %hdr = ( # response headers
349	HTTPVersion => "\x00$1",
350	Status => "\x00$2",
351	Reason => "\x00$3",
352	);	572	);
353		573
354	# headers, could be optimized a bit	574	# return if error occured during push_write()
355	$state{handle}->unshift_read (line => qr/\015?\012\015?\012/, sub {	575	return unless %state;
		576
		577	%hdr = (); # reduce memory usage, save a kitten, also make it possible to re-use
		578
		579	# status line and headers
		580	$state{read_response} = sub {
356	for ("$_[1]\012") {	581	for ("$_[1]") {
357	# we support spaces in field names, as lotus domino	582	y/\015//d; # weed out any \015, as they show up in the weirdest of places.
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		583
		584	/^HTTP\/([0-9\.]+) \s+ ([0-9]{3}) (?: \s+ ([^\012]*) )? \012/igxc
		585	or return (%state = (), $cb->(undef, { @pseudo, Status => 599, Reason => "Invalid server response" }));
		586
		587	# 100 Continue handling
		588	# should not happen as we don't send expect: 100-continue,
		589	# but we handle it just in case.
		590	# since we send the request body regardless, if we get an error
		591	# we are out of-sync, which we currently do NOT handle correctly.
		592	return $state{handle}->push_read (line => $qr_nlnl, $state{read_response})
		593	if $2 eq 100;
		594
		595	push @pseudo,
		596	HTTPVersion => $1,
		597	Status => $2,
		598	Reason => $3,
367	/\G$/	599	;
		600
		601	my $hdr = parse_hdr
368	or return (%state = (), $cb->(undef, { Status => 599, Reason => "garbled response headers" }));	602	or return (%state = (), $cb->(undef, { @pseudo, Status => 599, Reason => "Garbled response headers" }));
		603
		604	%hdr = (%$hdr, @pseudo);
369	}	605	}
370		606
371	substr $_, 0, 1, ""	607	# redirect handling
372	for values %hdr;	608	# microsoft and other shitheads don't give a shit for following standards,
		609	# try to support some common forms of broken Location headers.
		610	if ($hdr{location} !~ /^(?: $ \| [^:\/?\#]+ : )/x) {
		611	$hdr{location} =~ s/^\.\/+//;
373		612
374	my $finish = sub {	613	my $url = "$rscheme://$uhost:$uport";
		614
		615	unless ($hdr{location} =~ s/^\///) {
		616	$url .= $upath;
		617	$url =~ s/\/[^\/]*$//;
		618	}
		619
		620	$hdr{location} = "$url/$hdr{location}";
		621	}
		622
		623	my $redirect;
		624
		625	if ($recurse) {
		626	my $status = $hdr{Status};
		627
		628	# industry standard is to redirect POST as GET for
		629	# 301, 302 and 303, in contrast to http/1.0 and 1.1.
		630	# also, the UA should ask the user for 301 and 307 and POST,
		631	# industry standard seems to be to simply follow.
		632	# we go with the industry standard.
		633	if ($status == 301 or $status == 302 or $status == 303) {
		634	# HTTP/1.1 is unclear on how to mutate the method
		635	$method = "GET" unless $method eq "HEAD";
		636	$redirect = 1;
		637	} elsif ($status == 307) {
		638	$redirect = 1;
		639	}
		640	}
		641
		642	my $finish = sub { # ($data, $err_status, $err_reason[, $keepalive])
		643	my $keepalive = pop;
		644
		645	$state{handle}->destroy if $state{handle};
375	%state = ();	646	%state = ();
376		647
		648	if (defined $_[1]) {
		649	$hdr{OrigStatus} = $hdr{Status}; $hdr{Status} = $_[1];
		650	$hdr{OrigReason} = $hdr{Reason}; $hdr{Reason} = $_[2];
		651	}
		652
377	# set-cookie processing	653	# set-cookie processing
378	if ($arg{cookie_jar} && exists $hdr{"set-cookie"}) {	654	if ($arg{cookie_jar}) {
379	for (split /\x00/, $hdr{"set-cookie"}) {	655	for ($hdr{"set-cookie"}) {
380	my ($cookie, @arg) = split /;\s*/;	656	# parse NAME=VALUE
381	my ($name, $value) = split /=/, $cookie, 2;	657	my @kv;
382	my %kv = (value => $value, map { split /=/, $_, 2 } @arg);	658
		659	while (
		660	m{
		661	\G\s*
		662	(?:
		663	expires \s=\s ([A-Z][a-z][a-z],\ [^,;]+)
		664	\| ([^=;,[:space:]]+) \s=\s (?: "((?:[^\\"]+\|\\.))" \| ([^=;,[:space:]]) )
		665	)
		666	}gcxsi
		667	) {
		668	my $name = $2;
		669	my $value = $4;
		670
		671	unless (defined $name) {
		672	# expires
		673	$name = "expires";
		674	$value = $1;
		675	} elsif (!defined $value) {
		676	# quoted
		677	$value = $3;
		678	$value =~ s/\\(.)/$1/gs;
		679	}
		680
		681	push @kv, lc $name, $value;
		682
		683	last unless /\G\s*;/gc;
		684	}
		685
		686	last unless @kv;
		687
		688	my $name = shift @kv;
		689	my %kv = (value => shift @kv, @kv);
		690
		691	$kv{expires} \|\|= format_date (AE::now + $kv{"max-age"})
		692	if exists $kv{"max-age"};
		693
		694	my $cdom;
		695	my $cpath = (delete $kv{path}) \|\| "/";
		696
		697	if (exists $kv{domain}) {
		698	$cdom = delete $kv{domain};
383		699
384	my $cdom = (delete $kv{domain}) \|\| $uhost;
385	my $cpath = (delete $kv{path}) \|\| "/";
386
387	$cdom =~ s/^.?/./; # make sure it starts with a "."	700	$cdom =~ s/^\.?/./; # make sure it starts with a "."
388		701
389	next if $cdom =~ /\.$/;	702	next if $cdom =~ /\.$/;
390		703
391	# this is not rfc-like and not netscape-like. go figure.	704	# this is not rfc-like and not netscape-like. go figure.
392	my $ndots = $cdom =~ y/.//;	705	my $ndots = $cdom =~ y/.//;
393	next if $ndots < ($cdom =~ /\.[^.][^.]\.[^.][^.]$/ ? 3 : 2);	706	next if $ndots < ($cdom =~ /\.[^.][^.]\.[^.][^.]$/ ? 3 : 2);
		707	} else {
		708	$cdom = $uhost;
		709	}
394		710
395	# store it	711	# store it
396	$arg{cookie_jar}{version} = 1;	712	$arg{cookie_jar}{version} = 1;
397	$arg{cookie_jar}{$cdom}{$cpath}{$name} = \%kv;	713	$arg{cookie_jar}{$cdom}{$cpath}{$name} = \%kv;
		714
		715	redo if /\G\s*,/gc;
398	}	716	}
399	}	717	}
400		718
401	if ($_[1]{Status} =~ /^x30[12]$/ && $recurse) {	719	if ($redirect && exists $hdr{location}) {
402	# microsoft and other assholes don't give a shit for following standards,	720	# we ignore any errors, as it is very common to receive
403	# try to support a common form of broken Location header.	721	# Content-Length != 0 but no actual body
404	$_[1]{location} =~ s%^/%$scheme://$uhost:$uport/%;	722	# we also access %hdr, as $_[1] might be an erro
405		723	http_request (
406	http_request ($method, $_[1]{location}, %arg, recurse => $recurse - 1, $cb);	724	$method => $hdr{location},
		725	%arg,
		726	recurse => $recurse - 1,
		727	Redirect => [$_[0], \%hdr],
		728	$cb);
407	} else {	729	} else {
408	$cb->($_[0], $_[1]);	730	$cb->($_[0], \%hdr);
409	}	731	}
410	};	732	};
411		733
412	if ($hdr{Status} =~ /^(?:1..\|204\|304)$/ or $method eq "HEAD") {	734	my $len = $hdr{"content-length"};
		735
		736	if (!$redirect && $arg{on_header} && !$arg{on_header}(\%hdr)) {
		737	$finish->(undef, 598 => "Request cancelled by on_header");
		738	} elsif (
		739	$hdr{Status} =~ /^(?:1..\|204\|205\|304)$/
		740	or $method eq "HEAD"
		741	or (defined $len && !$len)
		742	) {
		743	# no body
413	$finish->(undef, \%hdr);	744	$finish->("", undef, undef, 1);
414	} else {	745	} else {
415	if (exists $hdr{"content-length"}) {	746	# body handling, many different code paths
		747	# - no body expected
		748	# - want_body_handle
		749	# - te chunked
		750	# - 2x length known (with or without on_body)
		751	# - 2x length not known (with or without on_body)
		752	if (!$redirect && $arg{want_body_handle}) {
		753	$_[0]->on_eof (undef);
		754	$_[0]->on_error (undef);
		755	$_[0]->on_read (undef);
		756
		757	$finish->(delete $state{handle});
		758
		759	} elsif ($hdr{"transfer-encoding"} =~ /\bchunked\b/i) {
		760	my $cl = 0;
		761	my $body = undef;
		762	my $on_body = $arg{on_body} \|\| sub { $body .= shift; 1 };
		763
		764	$_[0]->on_error (sub { $finish->(undef, 599 => $_[2]) });
		765
		766	my $read_chunk; $read_chunk = sub {
		767	$_[1] =~ /^([0-9a-fA-F]+)/
		768	or $finish->(undef, 599 => "Garbled chunked transfer encoding");
		769
		770	my $len = hex $1;
		771
		772	if ($len) {
		773	$cl += $len;
		774
416	$_[0]->unshift_read (chunk => $hdr{"content-length"}, sub {	775	$_[0]->push_read (chunk => $len, sub {
417	# could cache persistent connection now	776	$on_body->($_[1], \%hdr)
418	if ($hdr{connection} =~ /\bkeep-alive\b/i) {	777	or return $finish->(undef, 598 => "Request cancelled by on_body");
419	# but we don't, due to misdesigns, this is annoyingly complex	778
		779	$_[0]->push_read (line => sub {
		780	length $_[1]
		781	and return $finish->(undef, 599 => "Garbled chunked transfer encoding");
		782	$_[0]->push_read (line => $read_chunk);
		783	});
		784	});
		785	} else {
		786	$hdr{"content-length"} \|\|= $cl;
		787
		788	$_[0]->push_read (line => $qr_nlnl, sub {
		789	if (length $_[1]) {
		790	for ("$_[1]") {
		791	y/\015//d; # weed out any \015, as they show up in the weirdest of places.
		792
		793	my $hdr = parse_hdr
		794	or return $finish->(undef, 599 => "Garbled response trailers");
		795
		796	%hdr = (%hdr, %$hdr);
		797	}
		798	}
		799
		800	$finish->($body, undef, undef, 1);
		801	});
		802	}
		803	};
		804
		805	$_[0]->push_read (line => $read_chunk);
		806
		807	} elsif ($arg{on_body}) {
		808	$_[0]->on_error (sub { $finish->(undef, 599 => $_[2]) });
		809
		810	if ($len) {
		811	$_[0]->on_read (sub {
		812	$len -= length $_[0]{rbuf};
		813
		814	$arg{on_body}(delete $_[0]{rbuf}, \%hdr)
		815	or return $finish->(undef, 598 => "Request cancelled by on_body");
		816
		817	$len > 0
		818	or $finish->("", undef, undef, 1);
420	};	819	});
421		820	} else {
		821	$_[0]->on_eof (sub {
422	$finish->($_[1], \%hdr);	822	$finish->("");
		823	});
		824	$_[0]->on_read (sub {
		825	$arg{on_body}(delete $_[0]{rbuf}, \%hdr)
		826	or $finish->(undef, 598 => "Request cancelled by on_body");
		827	});
423	});	828	}
424	} else {	829	} else {
425	# too bad, need to read until we get an error or EOF,
426	# no way to detect winged data.
427	$_[0]->on_error (sub {
428	$finish->($_[0]{rbuf}, \%hdr);
429	});
430	$_[0]->on_eof (undef);	830	$_[0]->on_eof (undef);
		831
		832	if ($len) {
		833	$_[0]->on_error (sub { $finish->(undef, 599 => $_[2]) });
		834	$_[0]->on_read (sub {
		835	$finish->((substr delete $_[0]{rbuf}, 0, $len, ""), undef, undef, 1)
		836	if $len <= length $_[0]{rbuf};
		837	});
		838	} else {
		839	$_[0]->on_error (sub {
		840	($! == Errno::EPIPE \|\| !$!)
		841	? $finish->(delete $_[0]{rbuf})
		842	: $finish->(undef, 599 => $_[2]);
		843	});
431	$_[0]->on_read (sub { });	844	$_[0]->on_read (sub { });
		845	}
432	}	846	}
433	}	847	}
		848	};
		849
		850	$state{handle}->push_read (line => $qr_nlnl, $state{read_response});
		851	};
		852
		853	# now handle proxy-CONNECT method
		854	if ($proxy && $uscheme eq "https") {
		855	# oh dear, we have to wrap it into a connect request
		856
		857	# maybe re-use $uauthority with patched port?
		858	$state{handle}->push_write ("CONNECT $uhost:$uport HTTP/1.0\015\012Host: $uhost\015\012\015\012");
		859	$state{handle}->push_read (line => $qr_nlnl, sub {
		860	$_[1] =~ /^HTTP\/([0-9\.]+) \s+ ([0-9]{3}) (?: \s+ ([^\015\012]*) )?/ix
		861	or return (%state = (), $cb->(undef, { @pseudo, Status => 599, Reason => "Invalid proxy connect response ($_[1])" }));
		862
		863	if ($2 == 200) {
		864	$rpath = $upath;
		865	&$handle_actual_request;
		866	} else {
		867	%state = ();
		868	$cb->(undef, { @pseudo, Status => $2, Reason => $3 });
		869	}
434	});	870	});
		871	} else {
		872	&$handle_actual_request;
435	});	873	}
436	}, sub {
437	$timeout
438	};	874	};
		875
		876	my $tcp_connect = $arg{tcp_connect}
		877	\|\| do { require AnyEvent::Socket; \&AnyEvent::Socket::tcp_connect };
		878
		879	$state{connect_guard} = $tcp_connect->($rhost, $rport, $connect_cb, $arg{on_prepare} \|\| sub { $timeout });
		880
439	};	881	};
440		882
441	defined wantarray && AnyEvent::Util::guard { %state = () }	883	defined wantarray && AnyEvent::Util::guard { %state = () }
442	}	884	}
443		885
444	sub http_get($$;@) {	886	sub http_get($@) {
445	unshift @_, "GET";	887	unshift @_, "GET";
446	&http_request	888	&http_request
447	}	889	}
448		890
449	sub http_head($$;@) {	891	sub http_head($@) {
450	unshift @_, "HEAD";	892	unshift @_, "HEAD";
451	&http_request	893	&http_request
452	}	894	}
453		895
454	sub http_post($$$;@) {	896	sub http_post($$@) {
		897	my $url = shift;
455	unshift @_, "POST", "body";	898	unshift @_, "POST", $url, "body";
456	&http_request	899	&http_request
457	}	900	}
458		901
459	=back	902	=back
460		903
		904	=head2 DNS CACHING
		905
		906	AnyEvent::HTTP uses the AnyEvent::Socket::tcp_connect function for
		907	the actual connection, which in turn uses AnyEvent::DNS to resolve
		908	hostnames. The latter is a simple stub resolver and does no caching
		909	on its own. If you want DNS caching, you currently have to provide
		910	your own default resolver (by storing a suitable resolver object in
		911	C<$AnyEvent::DNS::RESOLVER>).
		912
461	=head2 GLOBAL FUNCTIONS AND VARIABLES	913	=head2 GLOBAL FUNCTIONS AND VARIABLES
462		914
463	=over 4	915	=over 4
464		916
465	=item AnyEvent::HTTP::set_proxy "proxy-url"	917	=item AnyEvent::HTTP::set_proxy "proxy-url"
466		918
467	Sets the default proxy server to use. The proxy-url must begin with a	919	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:...>).	920	string of the form C<http://host:port> (optionally C<https:...>), croaks
		921	otherwise.
		922
		923	To clear an already-set proxy, use C<undef>.
		924
		925	=item $date = AnyEvent::HTTP::format_date $timestamp
		926
		927	Takes a POSIX timestamp (seconds since the epoch) and formats it as a HTTP
		928	Date (RFC 2616).
		929
		930	=item $timestamp = AnyEvent::HTTP::parse_date $date
		931
		932	Takes a HTTP Date (RFC 2616) or a Cookie date (netscape cookie spec) and
		933	returns the corresponding POSIX timestamp, or C<undef> if the date cannot
		934	be parsed.
469		935
470	=item $AnyEvent::HTTP::MAX_RECURSE	936	=item $AnyEvent::HTTP::MAX_RECURSE
471		937
472	The default value for the C<recurse> request parameter (default: C<10>).	938	The default value for the C<recurse> request parameter (default: C<10>).
473		939
474	=item $AnyEvent::HTTP::USERAGENT	940	=item $AnyEvent::HTTP::USERAGENT
475		941
476	The default value for the C<User-Agent> header (the default is	942	The default value for the C<User-Agent> header (the default is
477	C<Mozilla/5.0 (compatible; AnyEvent::HTTP/$VERSION; +http://software.schmorp.de/pkg/AnyEvent)>).	943	C<Mozilla/5.0 (compatible; U; AnyEvent-HTTP/$VERSION; +http://software.schmorp.de/pkg/AnyEvent)>).
478		944
479	=item $AnyEvent::HTTP::MAX_PERSISTENT	945	=item $AnyEvent::HTTP::MAX_PER_HOST
480		946
481	The maximum number of persistent connections to keep open (default: 8).	947	The maximum number of concurrent connections to the same host (identified
		948	by the hostname). If the limit is exceeded, then the additional requests
		949	are queued until previous connections are closed.
482		950
483	Not implemented currently.	951	The default value for this is C<4>, and it is highly advisable to not
		952	increase it.
484		953
485	=item $AnyEvent::HTTP::PERSISTENT_TIMEOUT	954	=item $AnyEvent::HTTP::ACTIVE
486		955
487	The maximum time to cache a persistent connection, in seconds (default: 2).	956	The number of active connections. This is not the number of currently
488		957	running requests, but the number of currently open and non-idle TCP
489	Not implemented currently.	958	connections. This number of can be useful for load-leveling.
490		959
491	=back	960	=back
492		961
493	=cut	962	=cut
494		963
		964	our @month = qw(Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec);
		965	our @weekday = qw(Sun Mon Tue Wed Thu Fri Sat);
		966
		967	sub format_date($) {
		968	my ($time) = @_;
		969
		970	# RFC 822/1123 format
		971	my ($S, $M, $H, $mday, $mon, $year, $wday, $yday, undef) = gmtime $time;
		972
		973	sprintf "%s, %02d %s %04d %02d:%02d:%02d GMT",
		974	$weekday[$wday], $mday, $month[$mon], $year + 1900,
		975	$H, $M, $S;
		976	}
		977
		978	sub parse_date($) {
		979	my ($date) = @_;
		980
		981	my ($d, $m, $y, $H, $M, $S);
		982
		983	if ($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$/) {
		984	# RFC 822/1123, required by RFC 2616 (with " ")
		985	# cookie dates (with "-")
		986
		987	($d, $m, $y, $H, $M, $S) = ($1, $2, $3, $4, $5, $6);
		988
		989	} elsif ($date =~ /^[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$/) {
		990	# RFC 850
		991	($d, $m, $y, $H, $M, $S) = ($1, $2, $3 < 69 ? $3 + 2000 : $3 + 1900, $4, $5, $6);
		992
		993	} elsif ($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])$/) {
		994	# ISO C's asctime
		995	($d, $m, $y, $H, $M, $S) = ($2, $1, $6, $3, $4, $5);
		996	}
		997	# other formats fail in the loop below
		998
		999	for (0..11) {
		1000	if ($m eq $month[$_]) {
		1001	require Time::Local;
		1002	return Time::Local::timegm ($S, $M, $H, $d, $_, $y);
		1003	}
		1004	}
		1005
		1006	undef
		1007	}
		1008
495	sub set_proxy($) {	1009	sub set_proxy($) {
		1010	if (length $_[0]) {
496	$PROXY = [$2, $3 \|\| 3128, $1] if $_[0] =~ m%^(https?):// ([^:/]+) (?: : (\d*) )?%ix;	1011	$_[0] =~ m%^(https?):// ([^:/]+) (?: : (\d*) )?%ix
		1012	or Carp::croak "$_[0]: invalid proxy URL";
		1013	$PROXY = [$2, $3 \|\| 3128, $1]
		1014	} else {
		1015	undef $PROXY;
		1016	}
497	}	1017	}
498		1018
499	# initialise proxy from environment	1019	# initialise proxy from environment
		1020	eval {
500	set_proxy $ENV{http_proxy};	1021	set_proxy $ENV{http_proxy};
		1022	};
		1023
		1024	=head2 SOCKS PROXIES
		1025
		1026	Socks proxies are not directly supported by AnyEvent::HTTP. You can
		1027	compile your perl to support socks, or use an external program such as
		1028	F<socksify> (dante) or F<tsocks> to make your program use a socks proxy
		1029	transparently.
		1030
		1031	Alternatively, for AnyEvent::HTTP only, you can use your own
		1032	C<tcp_connect> function that does the proxy handshake - here is an example
		1033	that works with socks4a proxies:
		1034
		1035	use Errno;
		1036	use AnyEvent::Util;
		1037	use AnyEvent::Socket;
		1038	use AnyEvent::Handle;
		1039
		1040	# host, port and username of/for your socks4a proxy
		1041	my $socks_host = "10.0.0.23";
		1042	my $socks_port = 9050;
		1043	my $socks_user = "";
		1044
		1045	sub socks4a_connect {
		1046	my ($host, $port, $connect_cb, $prepare_cb) = @_;
		1047
		1048	my $hdl = new AnyEvent::Handle
		1049	connect => [$socks_host, $socks_port],
		1050	on_prepare => sub { $prepare_cb->($_[0]{fh}) },
		1051	on_error => sub { $connect_cb->() },
		1052	;
		1053
		1054	$hdl->push_write (pack "CCnNZZ", 4, 1, $port, 1, $socks_user, $host);
		1055
		1056	$hdl->push_read (chunk => 8, sub {
		1057	my ($hdl, $chunk) = @_;
		1058	my ($status, $port, $ipn) = unpack "xCna4", $chunk;
		1059
		1060	if ($status == 0x5a) {
		1061	$connect_cb->($hdl->{fh}, (format_address $ipn) . ":$port");
		1062	} else {
		1063	$! = Errno::ENXIO; $connect_cb->();
		1064	}
		1065	});
		1066
		1067	$hdl
		1068	}
		1069
		1070	Use C<socks4a_connect> instead of C<tcp_connect> when doing C<http_request>s,
		1071	possibly after switching off other proxy types:
		1072
		1073	AnyEvent::HTTP::set_proxy undef; # usually you do not want other proxies
		1074
		1075	http_get 'http://www.google.com', tcp_connect => \&socks4a_connect, sub {
		1076	my ($data, $headers) = @_;
		1077	...
		1078	};
501		1079
502	=head1 SEE ALSO	1080	=head1 SEE ALSO
503		1081
504	L<AnyEvent>.	1082	L<AnyEvent>.
505		1083
506	=head1 AUTHOR	1084	=head1 AUTHOR
507		1085
508	Marc Lehmann <schmorp@schmorp.de>	1086	Marc Lehmann <schmorp@schmorp.de>
509	http://home.schmorp.de/	1087	http://home.schmorp.de/
		1088
		1089	With many thanks to Дмитрий Шалашов, who provided countless
		1090	testcases and bugreports.
510		1091
511	=cut	1092	=cut
512		1093
513	1	1094	1
514		1095

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 1.70 by root, Fri Dec 31 20:31:47 2010 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 1.70 by root, Fri Dec 31 20:31:47 2010 UTC