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

Comparing AnyEvent-HTTP/HTTP.pm (file contents):
Revision 1.10 by root, Thu Jun 5 13:06:43 2008 UTC vs.
Revision 1.56 by root, Mon Sep 6 05:30:54 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.
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
…		…
20	package AnyEvent::HTTP;	39	package AnyEvent::HTTP;
21		40
22	use strict;	41	use strict;
23	no warnings;	42	no warnings;
24		43
25	use Carp;	44	use Errno ();
26		45
27	use AnyEvent ();	46	use AnyEvent 5.0 ();
28	use AnyEvent::Util ();	47	use AnyEvent::Util ();
29	use AnyEvent::Socket ();	48	use AnyEvent::Socket ();
30	use AnyEvent::Handle ();	49	use AnyEvent::Handle ();
31		50
32	use base Exporter::;	51	use base Exporter::;
33		52
34	our $VERSION = '1.0';	53	our $VERSION = '1.45';
35		54
36	our @EXPORT = qw(http_get http_request);	55	our @EXPORT = qw(http_get http_post http_head http_request);
37		56
38	our $USERAGENT = "Mozilla/5.0 (compatible; AnyEvent::HTTP/$VERSION; +http://software.schmorp.de/pkg/AnyEvent)";	57	our $USERAGENT = "Mozilla/5.0 (compatible; U; AnyEvent-HTTP/$VERSION; +http://software.schmorp.de/pkg/AnyEvent)";
39	our $MAX_RECURSE = 10;	58	our $MAX_RECURSE = 10;
40	our $MAX_PERSISTENT = 8;	59	our $MAX_PERSISTENT = 8;
41	our $PERSISTENT_TIMEOUT = 2;	60	our $PERSISTENT_TIMEOUT = 2;
42	our $TIMEOUT = 300;	61	our $TIMEOUT = 300;
43		62
44	# changing these is evil	63	# changing these is evil
45	our $MAX_PERSISTENT_PER_HOST = 2;	64	our $MAX_PERSISTENT_PER_HOST = 0;
46	our $MAX_PER_HOST = 4; # not respected yet :(	65	our $MAX_PER_HOST = 4;
47		66
48	our $PROXY;	67	our $PROXY;
		68	our $ACTIVE = 0;
49		69
50	my %KA_COUNT; # number of open keep-alive connections per host	70	my %KA_COUNT; # number of open keep-alive connections per host
		71	my %CO_SLOT; # number of open connections, and wait queue, per host
51		72
52	=item http_get $url, key => value..., $cb->($data, $headers)	73	=item http_get $url, key => value..., $cb->($data, $headers)
53		74
54	Executes an HTTP-GET request. See the http_request function for details on	75	Executes an HTTP-GET request. See the http_request function for details on
55	additional parameters.	76	additional parameters and the return value.
56		77
57	=item http_head $url, key => value..., $cb->($data, $headers)	78	=item http_head $url, key => value..., $cb->($data, $headers)
58		79
59	Executes an HTTP-HEAD request. See the http_request function for details on	80	Executes an HTTP-HEAD request. See the http_request function for details
60	additional parameters.	81	on additional parameters and the return value.
61		82
62	=item http_post $url, $body, key => value..., $cb->($data, $headers)	83	=item http_post $url, $body, key => value..., $cb->($data, $headers)
63		84
64	Executes an HTTP-POST request with a request body of C<$bod>. See the	85	Executes an HTTP-POST request with a request body of C<$body>. See the
65	http_request function for details on additional parameters.	86	http_request function for details on additional parameters and the return
		87	value.
66		88
67	=item http_request $method => $url, key => value..., $cb->($data, $headers)	89	=item http_request $method => $url, key => value..., $cb->($data, $headers)
68		90
69	Executes a HTTP request of type C<$method> (e.g. C<GET>, C<POST>). The URL	91	Executes a HTTP request of type C<$method> (e.g. C<GET>, C<POST>). The URL
70	must be an absolute http or https URL.	92	must be an absolute http or https URL.
71		93
		94	When called in void context, nothing is returned. In other contexts,
		95	C<http_request> returns a "cancellation guard" - you have to keep the
		96	object at least alive until the callback get called. If the object gets
		97	destroyed before the callbakc is called, the request will be cancelled.
		98
72	The callback will be called with the response data as first argument	99	The callback will be called with the response body data as first argument
73	(or C<undef> if it wasn't available due to errors), and a hash-ref with	100	(or C<undef> if an error occured), and a hash-ref with response headers as
74	response headers as second argument.	101	second argument.
75		102
76	All the headers in that hash are lowercased. In addition to the response	103	All the headers in that hash are lowercased. In addition to the response
77	headers, the three "pseudo-headers" C<HTTPVersion>, C<Status> and	104	headers, the "pseudo-headers" (uppercase to avoid clashing with possible
		105	response headers) C<HTTPVersion>, C<Status> and C<Reason> contain the
78	C<Reason> contain the three parts of the HTTP Status-Line of the same	106	three parts of the HTTP Status-Line of the same name.
		107
		108	The pseudo-header C<URL> contains the actual URL (which can differ from
		109	the requested URL when following redirects - for example, you might get
		110	an error that your URL scheme is not supported even though your URL is a
		111	valid http URL because it redirected to an ftp URL, in which case you can
		112	look at the URL pseudo header).
		113
		114	The pseudo-header C<Redirect> only exists when the request was a result
		115	of an internal redirect. In that case it is an array reference with
		116	the C<($data, $headers)> from the redirect response. Note that this
		117	response could in turn be the result of a redirect itself, and C<<
		118	$headers->{Redirect}[1]{Redirect} >> will then contain the original
		119	response, and so on.
		120
79	name. If the server sends a header multiple lines, then their contents	121	If the server sends a header multiple times, then their contents will be
80	will be joined together with C<\x00>.	122	joined together with a comma (C<,>), as per the HTTP spec.
81		123
82	If an internal error occurs, such as not being able to resolve a hostname,	124	If an internal error occurs, such as not being able to resolve a hostname,
83	then C<$data> will be C<undef>, C<< $headers->{Status} >> will be C<599>	125	then C<$data> will be C<undef>, C<< $headers->{Status} >> will be C<59x>
84	and the C<Reason> pseudo-header will contain an error message.	126	(usually C<599>) and the C<Reason> pseudo-header will contain an error
		127	message.
85		128
86	A typical callback might look like this:	129	A typical callback might look like this:
87		130
88	sub {	131	sub {
89	my ($body, $hdr) = @_;	132	my ($body, $hdr) = @_;
…		…
105	Whether to recurse requests or not, e.g. on redirects, authentication	148	Whether to recurse requests or not, e.g. on redirects, authentication
106	retries and so on, and how often to do so.	149	retries and so on, and how often to do so.
107		150
108	=item headers => hashref	151	=item headers => hashref
109		152
110	The request headers to use.	153	The request headers to use. Currently, C<http_request> may provide its
		154	own C<Host:>, C<Content-Length:>, C<Connection:> and C<Cookie:> headers
		155	and will provide defaults for C<User-Agent:> and C<Referer:> (this can be
		156	suppressed by using C<undef> for these headers in which case they won't be
		157	sent at all).
111		158
112	=item timeout => $seconds	159	=item timeout => $seconds
113		160
114	The time-out to use for various stages - each connect attempt will reset	161	The time-out to use for various stages - each connect attempt will reset
115	the timeout, as will read or write activity. Default timeout is 5 minutes.	162	the timeout, as will read or write activity, i.e. this is not an overall
		163	timeout.
		164
		165	Default timeout is 5 minutes.
116		166
117	=item proxy => [$host, $port[, $scheme]] or undef	167	=item proxy => [$host, $port[, $scheme]] or undef
118		168
119	Use the given http proxy for all requests. If not specified, then the	169	Use the given http proxy for all requests. If not specified, then the
120	default proxy (as specified by C<$ENV{http_proxy}>) is used.	170	default proxy (as specified by C<$ENV{http_proxy}>) is used.
121		171
122	C<$scheme> must be either missing or C<http> for HTTP, or C<https> for	172	C<$scheme> must be either missing, C<http> for HTTP or C<https> for
123	HTTPS.	173	HTTPS.
124		174
125	=item body => $string	175	=item body => $string
126		176
127	The request body, usually empty. Will be-sent as-is (future versions of	177	The request body, usually empty. Will be-sent as-is (future versions of
…		…
133	based on the original netscape specification.	183	based on the original netscape specification.
134		184
135	The C<$hash_ref> must be an (initially empty) hash reference which will	185	The C<$hash_ref> must be an (initially empty) hash reference which will
136	get updated automatically. It is possible to save the cookie_jar to	186	get updated automatically. It is possible to save the cookie_jar to
137	persistent storage with something like JSON or Storable, but this is not	187	persistent storage with something like JSON or Storable, but this is not
138	recommended, as expire times are currently being ignored.	188	recommended, as expiry times are currently being ignored.
139		189
140	Note that this cookie implementation is not of very high quality, nor	190	Note that this cookie implementation is not of very high quality, nor
141	meant to be complete. If you want complete cookie management you have to	191	meant to be complete. If you want complete cookie management you have to
142	do that on your own. C<cookie_jar> is meant as a quick fix to get some	192	do that on your own. C<cookie_jar> is meant as a quick fix to get some
143	cookie-using sites working. Cookies are a privacy disaster, do not use	193	cookie-using sites working. Cookies are a privacy disaster, do not use
144	them unless required to.	194	them unless required to.
		195
		196	=item tls_ctx => $scheme \| $tls_ctx
		197
		198	Specifies the AnyEvent::TLS context to be used for https connections. This
		199	parameter follows the same rules as the C<tls_ctx> parameter to
		200	L<AnyEvent::Handle>, but additionally, the two strings C<low> or
		201	C<high> can be specified, which give you a predefined low-security (no
		202	verification, highest compatibility) and high-security (CA and common-name
		203	verification) TLS context.
		204
		205	The default for this option is C<low>, which could be interpreted as "give
		206	me the page, no matter what".
		207
		208	=item on_prepare => $callback->($fh)
		209
		210	In rare cases you need to "tune" the socket before it is used to
		211	connect (for exmaple, to bind it on a given IP address). This parameter
		212	overrides the prepare callback passed to C<AnyEvent::Socket::tcp_connect>
		213	and behaves exactly the same way (e.g. it has to provide a
		214	timeout). See the description for the C<$prepare_cb> argument of
		215	C<AnyEvent::Socket::tcp_connect> for details.
		216
		217	=item on_header => $callback->($headers)
		218
		219	When specified, this callback will be called with the header hash as soon
		220	as headers have been successfully received from the remote server (not on
		221	locally-generated errors).
		222
		223	It has to return either true (in which case AnyEvent::HTTP will continue),
		224	or false, in which case AnyEvent::HTTP will cancel the download (and call
		225	the finish callback with an error code of C<598>).
		226
		227	This callback is useful, among other things, to quickly reject unwanted
		228	content, which, if it is supposed to be rare, can be faster than first
		229	doing a C<HEAD> request.
		230
		231	Example: cancel the request unless the content-type is "text/html".
		232
		233	on_header => sub {
		234	$_[0]{"content-type"} =~ /^text\/html\s*(?:;\|$)/
		235	},
		236
		237	=item on_body => $callback->($partial_body, $headers)
		238
		239	When specified, all body data will be passed to this callback instead of
		240	to the completion callback. The completion callback will get the empty
		241	string instead of the body data.
		242
		243	It has to return either true (in which case AnyEvent::HTTP will continue),
		244	or false, in which case AnyEvent::HTTP will cancel the download (and call
		245	the completion callback with an error code of C<598>).
		246
		247	This callback is useful when the data is too large to be held in memory
		248	(so the callback writes it to a file) or when only some information should
		249	be extracted, or when the body should be processed incrementally.
		250
		251	It is usually preferred over doing your own body handling via
		252	C<want_body_handle>, but in case of streaming APIs, where HTTP is
		253	only used to create a connection, C<want_body_handle> is the better
		254	alternative, as it allows you to install your own event handler, reducing
		255	resource usage.
		256
		257	=item want_body_handle => $enable
		258
		259	When enabled (default is disabled), the behaviour of AnyEvent::HTTP
		260	changes considerably: after parsing the headers, and instead of
		261	downloading the body (if any), the completion callback will be
		262	called. Instead of the C<$body> argument containing the body data, the
		263	callback will receive the L<AnyEvent::Handle> object associated with the
		264	connection. In error cases, C<undef> will be passed. When there is no body
		265	(e.g. status C<304>), the empty string will be passed.
		266
		267	The handle object might or might not be in TLS mode, might be connected to
		268	a proxy, be a persistent connection etc., and configured in unspecified
		269	ways. The user is responsible for this handle (it will not be used by this
		270	module anymore).
		271
		272	This is useful with some push-type services, where, after the initial
		273	headers, an interactive protocol is used (typical example would be the
		274	push-style twitter API which starts a JSON/XML stream).
		275
		276	If you think you need this, first have a look at C<on_body>, to see if
		277	that doesn't solve your problem in a better way.
145		278
146	=back	279	=back
147		280
148	Example: make a simple HTTP GET request for http://www.nethype.de/	281	Example: make a simple HTTP GET request for http://www.nethype.de/
149		282
…		…
163	use Data::Dumper;	296	use Data::Dumper;
164	print Dumper $hdr;	297	print Dumper $hdr;
165	}	298	}
166	;	299	;
167		300
		301	Example: make another simple HTTP GET request, but immediately try to
		302	cancel it.
		303
		304	my $request = http_request GET => "http://www.nethype.de/", sub {
		305	my ($body, $hdr) = @_;
		306	print "$body\n";
		307	};
		308
		309	undef $request;
		310
168	=cut	311	=cut
169		312
		313	sub _slot_schedule;
		314	sub _slot_schedule($) {
		315	my $host = shift;
		316
		317	while ($CO_SLOT{$host}[0] < $MAX_PER_HOST) {
		318	if (my $cb = shift @{ $CO_SLOT{$host}[1] }) {
		319	# somebody wants that slot
		320	++$CO_SLOT{$host}[0];
		321	++$ACTIVE;
		322
		323	$cb->(AnyEvent::Util::guard {
		324	--$ACTIVE;
		325	--$CO_SLOT{$host}[0];
		326	_slot_schedule $host;
		327	});
		328	} else {
		329	# nobody wants the slot, maybe we can forget about it
		330	delete $CO_SLOT{$host} unless $CO_SLOT{$host}[0];
		331	last;
		332	}
		333	}
		334	}
		335
		336	# wait for a free slot on host, call callback
		337	sub _get_slot($$) {
		338	push @{ $CO_SLOT{$_[0]}[1] }, $_[1];
		339
		340	_slot_schedule $_[0];
		341	}
		342
		343	our $qr_nl = qr{\015?\012};
		344	our $qr_nlnl = qr{(?<![^\012])\015?\012};
		345
		346	our $TLS_CTX_LOW = { cache => 1, sslv2 => 1 };
		347	our $TLS_CTX_HIGH = { cache => 1, verify => 1, verify_peername => "https" };
		348
170	sub http_request($$$;@) {	349	sub http_request($$@) {
171	my $cb = pop;	350	my $cb = pop;
172	my ($method, $url, %arg) = @_;	351	my ($method, $url, %arg) = @_;
173		352
174	my %hdr;	353	my %hdr;
		354
		355	$arg{tls_ctx} = $TLS_CTX_LOW if $arg{tls_ctx} eq "low" \|\| !exists $arg{tls_ctx};
		356	$arg{tls_ctx} = $TLS_CTX_HIGH if $arg{tls_ctx} eq "high";
175		357
176	$method = uc $method;	358	$method = uc $method;
177		359
178	if (my $hdr = $arg{headers}) {	360	if (my $hdr = $arg{headers}) {
179	while (my ($k, $v) = each %$hdr) {	361	while (my ($k, $v) = each %$hdr) {
180	$hdr{lc $k} = $v;	362	$hdr{lc $k} = $v;
181	}	363	}
182	}	364	}
183		365
		366	# pseudo headers for all subsequent responses
		367	my @pseudo = (URL => $url);
		368	push @pseudo, Redirect => delete $arg{Redirect} if exists $arg{Redirect};
		369
184	my $recurse = exists $arg{recurse} ? $arg{recurse} : $MAX_RECURSE;	370	my $recurse = exists $arg{recurse} ? delete $arg{recurse} : $MAX_RECURSE;
185		371
186	return $cb->(undef, { Status => 599, Reason => "recursion limit reached" })	372	return $cb->(undef, { Status => 599, Reason => "Too many redirections", @pseudo })
187	if $recurse < 0;	373	if $recurse < 0;
188		374
189	my $proxy = $arg{proxy} \|\| $PROXY;	375	my $proxy = $arg{proxy} \|\| $PROXY;
190	my $timeout = $arg{timeout} \|\| $TIMEOUT;	376	my $timeout = $arg{timeout} \|\| $TIMEOUT;
191		377
192	$hdr{"user-agent"} \|\|= $USERAGENT;
193
194	my ($scheme, $authority, $upath, $query, $fragment) =	378	my ($uscheme, $uauthority, $upath, $query, $fragment) =
195	$url =~ m\|(?:([^:/?#]+):)?(?://([^/?#]))?([^?#])(?:\?([^#]))?(?:#(.))?\|;	379	$url =~ m\|(?:([^:/?#]+):)?(?://([^/?#]))?([^?#])(?:(\?[^#]))?(?:#(.))?\|;
196		380
197	$scheme = lc $scheme;	381	$uscheme = lc $uscheme;
198		382
199	my $uport = $scheme eq "http" ? 80	383	my $uport = $uscheme eq "http" ? 80
200	: $scheme eq "https" ? 443	384	: $uscheme eq "https" ? 443
201	: return $cb->(undef, { Status => 599, Reason => "only http and https URL schemes supported" });	385	: return $cb->(undef, { Status => 599, Reason => "Only http and https URL schemes supported", @pseudo });
202		386
203	$authority =~ /^(?: .*\@ )? ([^\@:]+) (?: : (\d+) )?$/x	387	$uauthority =~ /^(?: .*\@ )? ([^\@:]+) (?: : (\d+) )?$/x
204	or return $cb->(undef, { Status => 599, Reason => "unparsable URL" });	388	or return $cb->(undef, { Status => 599, Reason => "Unparsable URL", @pseudo });
205		389
206	my $uhost = $1;	390	my $uhost = $1;
207	$uport = $2 if defined $2;	391	$uport = $2 if defined $2;
208		392
		393	$hdr{host} = defined $2 ? "$uhost:$2" : "$uhost"
		394	unless exists $hdr{host};
		395
209	$uhost =~ s/^\[(.*)\]$/$1/;	396	$uhost =~ s/^\[(.*)\]$/$1/;
210	$upath .= "?$query" if length $query;	397	$upath .= $query if length $query;
211		398
212	$upath =~ s%^/?%/%;	399	$upath =~ s%^/?%/%;
213		400
214	# cookie processing	401	# cookie processing
215	if (my $jar = $arg{cookie_jar}) {	402	if (my $jar = $arg{cookie_jar}) {
216	%$jar = () if $jar->{version} < 1;	403	%$jar = () if $jar->{version} != 1;
217		404
218	my @cookie;	405	my @cookie;
219		406
220	while (my ($chost, $v) = each %$jar) {	407	while (my ($chost, $v) = each %$jar) {
		408	if ($chost =~ /^\./) {
221	next unless $chost eq substr $uhost, -length $chost;	409	next unless $chost eq substr $uhost, -length $chost;
222	next unless $chost =~ /^\./;	410	} elsif ($chost =~ /\./) {
		411	next unless $chost eq $uhost;
		412	} else {
		413	next;
		414	}
223		415
224	while (my ($cpath, $v) = each %$v) {	416	while (my ($cpath, $v) = each %$v) {
225	next unless $cpath eq substr $upath, 0, length $cpath;	417	next unless $cpath eq substr $upath, 0, length $cpath;
226		418
227	while (my ($k, $v) = each %$v) {	419	while (my ($k, $v) = each %$v) {
228	next if $scheme ne "https" && exists $v->{secure};	420	next if $uscheme ne "https" && exists $v->{secure};
		421	my $value = $v->{value};
		422	$value =~ s/([\\"])/\\$1/g;
229	push @cookie, "$k=$v->{value}";	423	push @cookie, "$k=\"$value\"";
230	}	424	}
231	}	425	}
232	}	426	}
233		427
234	$hdr{cookie} = join "; ", @cookie	428	$hdr{cookie} = join "; ", @cookie
235	if @cookie;	429	if @cookie;
236	}	430	}
237		431
238	my ($rhost, $rport, $rpath); # request host, port, path	432	my ($rhost, $rport, $rscheme, $rpath); # request host, port, path
239		433
240	if ($proxy) {	434	if ($proxy) {
241	($rhost, $rport, $scheme) = @$proxy;	435	($rpath, $rhost, $rport, $rscheme) = ($url, @$proxy);
242	$rpath = $url;	436
		437	$rscheme = "http" unless defined $rscheme;
		438
		439	# don't support https requests over https-proxy transport,
		440	# can't be done with tls as spec'ed, unless you double-encrypt.
		441	$rscheme = "http" if $uscheme eq "https" && $rscheme eq "https";
243	} else {	442	} else {
244	($rhost, $rport, $rpath) = ($uhost, $uport, $upath);	443	($rhost, $rport, $rscheme, $rpath) = ($uhost, $uport, $uscheme, $upath);
245	$hdr{host} = $uhost;
246	}	444	}
247		445
		446	# leave out fragment and query string, just a heuristic
		447	$hdr{referer} \|\|= "$uscheme://$uauthority$upath" unless exists $hdr{referer};
		448	$hdr{"user-agent"} \|\|= $USERAGENT unless exists $hdr{"user-agent"};
		449
248	$hdr{"content-length"} = length $arg{body};	450	$hdr{"content-length"} = length $arg{body}
		451	if length $arg{body} \|\| $method ne "GET";
249		452
250	my %state;	453	my %state = (connect_guard => 1);
251		454
		455	_get_slot $uhost, sub {
		456	$state{slot_guard} = shift;
		457
		458	return unless $state{connect_guard};
		459
252	$state{connect_guard} = AnyEvent::Socket::tcp_connect $rhost, $rport, sub {	460	$state{connect_guard} = AnyEvent::Socket::tcp_connect $rhost, $rport, sub {
253	$state{fh} = shift	461	$state{fh} = shift
254	or return $cb->(undef, { Status => 599, Reason => "$!" });
255
256	delete $state{connect_guard}; # reduce memory usage, save a tree
257
258	# get handle
259	$state{handle} = new AnyEvent::Handle
260	fh => $state{fh},
261	($scheme eq "https" ? (tls => "connect") : ());
262
263	# limit the number of persistent connections
264	if ($KA_COUNT{$_[1]} < $MAX_PERSISTENT_PER_HOST) {
265	++$KA_COUNT{$_[1]};
266	$state{handle}{ka_count_guard} = AnyEvent::Util::guard { --$KA_COUNT{$_[1]} };
267	$hdr{connection} = "keep-alive";
268	delete $hdr{connection}; # keep-alive not yet supported
269	} else {
270	delete $hdr{connection};
271	}
272
273	# (re-)configure handle
274	$state{handle}->timeout ($timeout);
275	$state{handle}->on_error (sub {
276	%state = ();
277	$cb->(undef, { Status => 599, Reason => "$!" });
278	});
279	$state{handle}->on_eof (sub {
280	%state = ();
281	$cb->(undef, { Status => 599, Reason => "unexpected end-of-file" });
282	});
283
284	# send request
285	$state{handle}->push_write (
286	"$method $rpath HTTP/1.0\015\012"
287	. (join "", map "$_: $hdr{$_}\015\012", keys %hdr)
288	. "\015\012"
289	. (delete $arg{body})
290	);
291
292	%hdr = (); # reduce memory usage, save a kitten
293
294	# status line
295	$state{handle}->push_read (line => qr/\015?\012/, sub {
296	$_[1] =~ /^HTTP\/([0-9\.]+) \s+ ([0-9]{3}) \s+ ([^\015\012]+)/ix
297	or return (%state = (), $cb->(undef, { Status => 599, Reason => "invalid server response ($_[1])" }));
298
299	my %hdr = ( # response headers
300	HTTPVersion => "\x00$1",
301	Status => "\x00$2",
302	Reason => "\x00$3",
303	);
304
305	# headers, could be optimized a bit
306	$state{handle}->unshift_read (line => qr/\015?\012\015?\012/, sub {
307	for ("$_[1]\012") {
308	# we support spaces in field names, as lotus domino
309	# creates them.
310	$hdr{lc $1} .= "\x00$2"
311	while /\G
312	([^:\000-\037]+):
313	[\011\040]*
314	((?: [^\015\012]+ \| \015?\012[\011\040] )*)
315	\015?\012
316	/gxc;
317
318	/\G$/
319	or return (%state = (), $cb->(undef, { Status => 599, Reason => "garbled response headers" }));
320	}	462	or do {
321		463	my $err = "$!";
322	substr $_, 0, 1, ""
323	for values %hdr;
324
325	my $finish = sub {
326	%state = ();	464	%state = ();
		465	return $cb->(undef, { Status => 599, Reason => $err, @pseudo });
		466	};
327		467
328	# set-cookie processing	468	pop; # free memory, save a tree
329	if ($arg{cookie_jar} && exists $hdr{"set-cookie"}) {	469
330	for (split /\x00/, $hdr{"set-cookie"}) {	470	return unless delete $state{connect_guard};
331	my ($cookie, @arg) = split /;\s*/;	471
332	my ($name, $value) = split /=/, $cookie, 2;	472	# get handle
333	my %kv = (value => $value, map { split /=/, $_, 2 } @arg);	473	$state{handle} = new AnyEvent::Handle
334		474	fh => $state{fh},
335	my $cdom = (delete $kv{domain}) \|\| $uhost;	475	peername => $rhost,
336	my $cpath = (delete $kv{path}) \|\| "/";	476	tls_ctx => $arg{tls_ctx},
337		477	# these need to be reconfigured on keepalive handles
338	$cdom =~ s/^.?/./; # make sure it starts with a "."	478	timeout => $timeout,
339		479	on_error => sub {
340	my $ndots = $cdom =~ y/.//;	480	%state = ();
341	next if $ndots < ($cdom =~ /[^.]{3}$/ ? 2 : 3);	481	$cb->(undef, { Status => 599, Reason => $_[2], @pseudo });
342		482	},
343	# store it	483	on_eof => sub {
344	$arg{cookie_jar}{version} = 1;	484	%state = ();
345	$arg{cookie_jar}{$cdom}{$cpath}{$name} = \%kv;	485	$cb->(undef, { Status => 599, Reason => "Unexpected end-of-file", @pseudo });
		486	},
		487	;
		488
		489	# limit the number of persistent connections
		490	# keepalive not yet supported
		491	# if ($KA_COUNT{$_[1]} < $MAX_PERSISTENT_PER_HOST) {
		492	# ++$KA_COUNT{$_[1]};
		493	# $state{handle}{ka_count_guard} = AnyEvent::Util::guard {
		494	# --$KA_COUNT{$_[1]}
		495	# };
		496	# $hdr{connection} = "keep-alive";
		497	# } else {
		498	delete $hdr{connection};
		499	# }
		500
		501	$state{handle}->starttls ("connect") if $rscheme eq "https";
		502
		503	# handle actual, non-tunneled, request
		504	my $handle_actual_request = sub {
		505	$state{handle}->starttls ("connect") if $uscheme eq "https" && !exists $state{handle}{tls};
		506
		507	# send request
		508	$state{handle}->push_write (
		509	"$method $rpath HTTP/1.0\015\012"
		510	. (join "", map "\u$_: $hdr{$_}\015\012", grep defined $hdr{$_}, keys %hdr)
		511	. "\015\012"
		512	. (delete $arg{body})
		513	);
		514
		515	# return if error occured during push_write()
		516	return unless %state;
		517
		518	%hdr = (); # reduce memory usage, save a kitten, also make it possible to re-use
		519
		520	# status line
		521	$state{handle}->push_read (line => $qr_nl, sub {
		522	$_[1] =~ /^HTTP\/([0-9\.]+) \s+ ([0-9]{3}) (?: \s+ ([^\015\012]*) )?/ix
		523	or return (%state = (), $cb->(undef, { Status => 599, Reason => "Invalid server response ($_[1])", @pseudo }));
		524
		525	push @pseudo,
		526	HTTPVersion => $1,
		527	Status => $2,
		528	Reason => $3,
		529	;
		530
		531	# headers, could be optimized a bit
		532	$state{handle}->unshift_read (line => $qr_nlnl, sub {
		533	for ("$_[1]") {
		534	y/\015//d; # weed out any \015, as they show up in the weirdest of places.
		535
		536	# things seen, not parsed:
		537	# p3pP="NON CUR OTPi OUR NOR UNI"
		538
		539	$hdr{lc $1} .= ",$2"
		540	while /\G
		541	([^:\000-\037]*):
		542	[\011\040]*
		543	((?: [^\012]+ \| \012[\011\040] )*)
		544	\012
		545	/gxc;
		546
		547	/\G$/
		548	or return (%state = (), $cb->(undef, { Status => 599, Reason => "Garbled response headers", @pseudo }));
346	}	549	}
		550
		551	# remove the "," prefix we added to all headers above
		552	substr $_, 0, 1, ""
		553	for values %hdr;
		554
		555	# patch in all pseudo headers
		556	%hdr = (%hdr, @pseudo);
		557
		558	# redirect handling
		559	# microsoft and other shitheads don't give a shit for following standards,
		560	# try to support some common forms of broken Location headers.
		561	if ($hdr{location} !~ /^(?: $ \| [^:\/?\#]+ : )/x) {
		562	$hdr{location} =~ s/^\.\/+//;
		563
		564	my $url = "$rscheme://$uhost:$uport";
		565
		566	unless ($hdr{location} =~ s/^\///) {
		567	$url .= $upath;
		568	$url =~ s/\/[^\/]*$//;
		569	}
		570
		571	$hdr{location} = "$url/$hdr{location}";
		572	}
		573
		574	my $redirect;
		575
		576	if ($recurse) {
		577	my $status = $hdr{Status};
		578
		579	if (($status == 301 \|\| $status == 302) && $method ne "POST") {
		580	# apparently, mozilla et al. just change POST to GET here
		581	# more research is needed before we do the same
		582	$redirect = 1;
		583	} elsif ($status == 303) {
		584	# even http/1.1 is unclear on how to mutate the method
		585	$method = "GET" unless $method eq "HEAD";
		586	$redirect = 1;
		587	} elsif ($status == 307 && $method =~ /^(?:GET\|HEAD)$/) {
		588	$redirect = 1;
		589	}
		590	}
		591
		592	my $finish = sub {
		593	$state{handle}->destroy if $state{handle};
		594	%state = ();
		595
		596	# set-cookie processing
		597	if ($arg{cookie_jar}) {
		598	for ($_[1]{"set-cookie"}) {
		599	# parse NAME=VALUE
		600	my @kv;
		601
		602	while (/\G\s* ([^=;,[:space:]]+) \s=\s (?: "((?:[^\\"]+\|\\.))" \| ([^=;,[:space:]]) )/gcxs) {
		603	my $name = $1;
		604	my $value = $3;
		605
		606	unless ($value) {
		607	$value = $2;
		608	$value =~ s/\\(.)/$1/gs;
		609	}
		610
		611	push @kv, $name => $value;
		612
		613	last unless /\G\s*;/gc;
		614	}
		615
		616	last unless @kv;
		617
		618	my $name = shift @kv;
		619	my %kv = (value => shift @kv, @kv);
		620
		621	my $cdom;
		622	my $cpath = (delete $kv{path}) \|\| "/";
		623
		624	if (exists $kv{domain}) {
		625	$cdom = delete $kv{domain};
		626
		627	$cdom =~ s/^\.?/./; # make sure it starts with a "."
		628
		629	next if $cdom =~ /\.$/;
		630
		631	# this is not rfc-like and not netscape-like. go figure.
		632	my $ndots = $cdom =~ y/.//;
		633	next if $ndots < ($cdom =~ /\.[^.][^.]\.[^.][^.]$/ ? 3 : 2);
		634	} else {
		635	$cdom = $uhost;
		636	}
		637
		638	# store it
		639	$arg{cookie_jar}{version} = 1;
		640	$arg{cookie_jar}{$cdom}{$cpath}{$name} = \%kv;
		641
		642	redo if /\G\s*,/gc;
		643	}
		644	}
		645
		646	if ($redirect && exists $hdr{location}) {
		647	# we ignore any errors, as it is very common to receive
		648	# Content-Length != 0 but no actual body
		649	# we also access %hdr, as $_[1] might be an erro
		650	http_request (
		651	$method => $hdr{location},
		652	%arg,
		653	recurse => $recurse - 1,
		654	Redirect => \@_,
		655	$cb);
		656	} else {
		657	$cb->($_[0], $_[1]);
		658	}
		659	};
		660
		661	my $len = $hdr{"content-length"};
		662
		663	if (!$redirect && $arg{on_header} && !$arg{on_header}(\%hdr)) {
		664	$finish->(undef, { Status => 598, Reason => "Request cancelled by on_header", @pseudo });
		665	} elsif (
		666	$hdr{Status} =~ /^(?:1..\|[23]04)$/
		667	or $method eq "HEAD"
		668	or (defined $len && !$len)
		669	) {
		670	# no body
		671	$finish->("", \%hdr);
		672	} else {
		673	# body handling, four different code paths
		674	# for want_body_handle, on_body (2x), normal (2x)
		675	# we might read too much here, but it does not matter yet (no pers. connections)
		676	if (!$redirect && $arg{want_body_handle}) {
		677	$_[0]->on_eof (undef);
		678	$_[0]->on_error (undef);
		679	$_[0]->on_read (undef);
		680
		681	$finish->(delete $state{handle}, \%hdr);
		682
		683	} elsif ($arg{on_body}) {
		684	$_[0]->on_error (sub { $finish->(undef, { Status => 599, Reason => $_[2], @pseudo }) });
		685	if ($len) {
		686	$_[0]->on_eof (undef);
		687	$_[0]->on_read (sub {
		688	$len -= length $_[0]{rbuf};
		689
		690	$arg{on_body}(delete $_[0]{rbuf}, \%hdr)
		691	or $finish->(undef, { Status => 598, Reason => "Request cancelled by on_body", @pseudo });
		692
		693	$len > 0
		694	or $finish->("", \%hdr);
		695	});
		696	} else {
		697	$_[0]->on_eof (sub {
		698	$finish->("", \%hdr);
		699	});
		700	$_[0]->on_read (sub {
		701	$arg{on_body}(delete $_[0]{rbuf}, \%hdr)
		702	or $finish->(undef, { Status => 598, Reason => "Request cancelled by on_body", @pseudo });
		703	});
		704	}
		705	} else {
		706	$_[0]->on_eof (undef);
		707
		708	if ($len) {
		709	$_[0]->on_error (sub { $finish->(undef, { Status => 599, Reason => $_[2], @pseudo }) });
		710	$_[0]->on_read (sub {
		711	$finish->((substr delete $_[0]{rbuf}, 0, $len, ""), \%hdr)
		712	if $len <= length $_[0]{rbuf};
		713	});
		714	} else {
		715	$_[0]->on_error (sub {
		716	$! == Errno::EPIPE \|\| !$!
		717	? $finish->(delete $_[0]{rbuf}, \%hdr)
		718	: $finish->(undef, { Status => 599, Reason => $_[2], @pseudo });
		719	});
		720	$_[0]->on_read (sub { });
		721	}
		722	}
		723	}
		724	});
		725	});
		726	};
		727
		728	# now handle proxy-CONNECT method
		729	if ($proxy && $uscheme eq "https") {
		730	# oh dear, we have to wrap it into a connect request
		731
		732	# maybe re-use $uauthority with patched port?
		733	$state{handle}->push_write ("CONNECT $uhost:$uport HTTP/1.0\015\012Host: $uhost\015\012\015\012");
		734	$state{handle}->push_read (line => $qr_nlnl, sub {
		735	$_[1] =~ /^HTTP\/([0-9\.]+) \s+ ([0-9]{3}) (?: \s+ ([^\015\012]*) )?/ix
		736	or return (%state = (), $cb->(undef, { Status => 599, Reason => "Invalid proxy connect response ($_[1])", @pseudo }));
		737
		738	if ($2 == 200) {
		739	$rpath = $upath;
		740	&$handle_actual_request;
		741	} else {
		742	%state = ();
		743	$cb->(undef, { Status => $2, Reason => $3, @pseudo });
347	}	744	}
348
349	if ($_[1]{Status} =~ /^x30[12]$/ && $recurse) {
350	# microsoft and other assholes don't give a shit for following standards,
351	# try to support a common form of broken Location header.
352	$_[1]{location} =~ s%^/%$scheme://$uhost:$uport/%;
353
354	http_request ($method, $_[1]{location}, %arg, recurse => $recurse - 1, $cb);
355	} else {
356	$cb->($_[0], $_[1]);
357	}
358	};	745	});
359
360	if ($hdr{Status} =~ /^(?:1..\|204\|304)$/ or $method eq "HEAD") {
361	$finish->(undef, \%hdr);
362	} else {	746	} else {
363	if (exists $hdr{"content-length"}) {	747	&$handle_actual_request;
364	$_[0]->unshift_read (chunk => $hdr{"content-length"}, sub {
365	# could cache persistent connection now
366	if ($hdr{connection} =~ /\bkeep-alive\b/i) {
367	# but we don't, due to misdesigns, this is annoyingly complex
368	};
369
370	$finish->($_[1], \%hdr);
371	});
372	} else {
373	# too bad, need to read until we get an error or EOF,
374	# no way to detect winged data.
375	$_[0]->on_error (sub {
376	$finish->($_[0]{rbuf}, \%hdr);
377	});
378	$_[0]->on_eof (undef);
379	$_[0]->on_read (sub { });
380	}
381	}
382	});	748	}
383	});	749
384	}, sub {	750	}, $arg{on_prepare} \|\| sub { $timeout };
385	$timeout
386	};	751	};
387		752
388	defined wantarray && AnyEvent::Util::guard { %state = () }	753	defined wantarray && AnyEvent::Util::guard { %state = () }
389	}	754	}
390		755
391	sub http_get($$;@) {	756	sub http_get($@) {
392	unshift @_, "GET";	757	unshift @_, "GET";
393	&http_request	758	&http_request
394	}	759	}
395		760
396	sub http_head($$;@) {	761	sub http_head($@) {
397	unshift @_, "HEAD";	762	unshift @_, "HEAD";
398	&http_request	763	&http_request
399	}	764	}
400		765
401	sub http_post($$$;@) {	766	sub http_post($$@) {
		767	my $url = shift;
402	unshift @_, "POST", "body";	768	unshift @_, "POST", $url, "body";
403	&http_request	769	&http_request
404	}	770	}
405		771
406	=back	772	=back
407		773
		774	=head2 DNS CACHING
		775
		776	AnyEvent::HTTP uses the AnyEvent::Socket::tcp_connect function for
		777	the actual connection, which in turn uses AnyEvent::DNS to resolve
		778	hostnames. The latter is a simple stub resolver and does no caching
		779	on its own. If you want DNS caching, you currently have to provide
		780	your own default resolver (by storing a suitable resolver object in
		781	C<$AnyEvent::DNS::RESOLVER>).
		782
408	=head2 GLOBAL FUNCTIONS AND VARIABLES	783	=head2 GLOBAL FUNCTIONS AND VARIABLES
409		784
410	=over 4	785	=over 4
411		786
412	=item AnyEvent::HTTP::set_proxy "proxy-url"	787	=item AnyEvent::HTTP::set_proxy "proxy-url"
413		788
414	Sets the default proxy server to use. The proxy-url must begin with a	789	Sets the default proxy server to use. The proxy-url must begin with a
415	string of the form C<http://host:port> (optionally C<https:...>).	790	string of the form C<http://host:port> (optionally C<https:...>), croaks
		791	otherwise.
		792
		793	To clear an already-set proxy, use C<undef>.
416		794
417	=item $AnyEvent::HTTP::MAX_RECURSE	795	=item $AnyEvent::HTTP::MAX_RECURSE
418		796
419	The default value for the C<recurse> request parameter (default: C<10>).	797	The default value for the C<recurse> request parameter (default: C<10>).
420		798
421	=item $AnyEvent::HTTP::USERAGENT	799	=item $AnyEvent::HTTP::USERAGENT
422		800
423	The default value for the C<User-Agent> header (the default is	801	The default value for the C<User-Agent> header (the default is
424	C<Mozilla/5.0 (compatible; AnyEvent::HTTP/$VERSION; +http://software.schmorp.de/pkg/AnyEvent)>).	802	C<Mozilla/5.0 (compatible; U; AnyEvent-HTTP/$VERSION; +http://software.schmorp.de/pkg/AnyEvent)>).
425		803
426	=item $AnyEvent::HTTP::MAX_PERSISTENT	804	=item $AnyEvent::HTTP::MAX_PER_HOST
427		805
428	The maximum number of persistent connections to keep open (default: 8).	806	The maximum number of concurrent connections to the same host (identified
		807	by the hostname). If the limit is exceeded, then the additional requests
		808	are queued until previous connections are closed.
429		809
430	Not implemented currently.	810	The default value for this is C<4>, and it is highly advisable to not
		811	increase it.
431		812
432	=item $AnyEvent::HTTP::PERSISTENT_TIMEOUT	813	=item $AnyEvent::HTTP::ACTIVE
433		814
434	The maximum time to cache a persistent connection, in seconds (default: 2).	815	The number of active connections. This is not the number of currently
435		816	running requests, but the number of currently open and non-idle TCP
436	Not implemented currently.	817	connections. This number of can be useful for load-leveling.
437		818
438	=back	819	=back
439		820
440	=cut	821	=cut
441		822
442	sub set_proxy($) {	823	sub set_proxy($) {
		824	if (length $_[0]) {
443	$PROXY = [$2, $3 \|\| 3128, $1] if $_[0] =~ m%^(https?):// ([^:/]+) (?: : (\d*) )?%ix;	825	$_[0] =~ m%^(https?):// ([^:/]+) (?: : (\d*) )?%ix
		826	or Carp::croak "$_[0]: invalid proxy URL";
		827	$PROXY = [$2, $3 \|\| 3128, $1]
		828	} else {
		829	undef $PROXY;
		830	}
444	}	831	}
445		832
446	# initialise proxy from environment	833	# initialise proxy from environment
		834	eval {
447	set_proxy $ENV{http_proxy};	835	set_proxy $ENV{http_proxy};
		836	};
448		837
449	=head1 SEE ALSO	838	=head1 SEE ALSO
450		839
451	L<AnyEvent>.	840	L<AnyEvent>.
452		841
453	=head1 AUTHOR	842	=head1 AUTHOR
454		843
455	Marc Lehmann <schmorp@schmorp.de>	844	Marc Lehmann <schmorp@schmorp.de>
456	http://home.schmorp.de/	845	http://home.schmorp.de/
		846
		847	With many thanks to Дмитрий Шалашов, who provided countless
		848	testcases and bugreports.
457		849
458	=cut	850	=cut
459		851
460	1	852	1
461		853

Diff Legend

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

Comparing AnyEvent-HTTP/HTTP.pm (file contents): Revision 1.10 by root, Thu Jun 5 13:06:43 2008 UTC vs. Revision 1.56 by root, Mon Sep 6 05:30:54 2010 UTC

Diff Legend

Comparing AnyEvent-HTTP/HTTP.pm (file contents):
Revision 1.10 by root, Thu Jun 5 13:06:43 2008 UTC vs.
Revision 1.56 by root, Mon Sep 6 05:30:54 2010 UTC