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

Comparing AnyEvent-HTTP/HTTP.pm (file contents):
Revision 1.19 by elmex, Mon Jun 9 13:02:13 2008 UTC vs.
Revision 1.63 by root, Thu Dec 30 04:31:55 2010 UTC

…		…
39	package AnyEvent::HTTP;	39	package AnyEvent::HTTP;
40		40
41	use strict;	41	use strict;
42	no warnings;	42	no warnings;
43		43
44	use Carp;	44	use Errno ();
45		45
46	use AnyEvent ();	46	use AnyEvent 5.0 ();
47	use AnyEvent::Util ();	47	use AnyEvent::Util ();
48	use AnyEvent::Socket ();
49	use AnyEvent::Handle ();	48	use AnyEvent::Handle ();
50		49
51	use base Exporter::;	50	use base Exporter::;
52		51
53	our $VERSION = '1.01';	52	our $VERSION = '1.46';
54		53
55	our @EXPORT = qw(http_get http_post http_head http_request);	54	our @EXPORT = qw(http_get http_post http_head http_request);
56		55
57	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)";
58	our $MAX_RECURSE = 10;	57	our $MAX_RECURSE = 10;
59	our $MAX_PERSISTENT = 8;	58	our $MAX_PERSISTENT = 8;
60	our $PERSISTENT_TIMEOUT = 2;	59	our $PERSISTENT_TIMEOUT = 2;
61	our $TIMEOUT = 300;	60	our $TIMEOUT = 300;
62		61
63	# changing these is evil	62	# changing these is evil
64	our $MAX_PERSISTENT_PER_HOST = 2;	63	our $MAX_PERSISTENT_PER_HOST = 0;
65	our $MAX_PER_HOST = 4;	64	our $MAX_PER_HOST = 4;
66		65
67	our $PROXY;	66	our $PROXY;
68	our $ACTIVE = 0;	67	our $ACTIVE = 0;
69		68
…		…
71	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
72		71
73	=item http_get $url, key => value..., $cb->($data, $headers)	72	=item http_get $url, key => value..., $cb->($data, $headers)
74		73
75	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
76	additional parameters.	75	additional parameters and the return value.
77		76
78	=item http_head $url, key => value..., $cb->($data, $headers)	77	=item http_head $url, key => value..., $cb->($data, $headers)
79		78
80	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
81	additional parameters.	80	on additional parameters and the return value.
82		81
83	=item http_post $url, $body, key => value..., $cb->($data, $headers)	82	=item http_post $url, $body, key => value..., $cb->($data, $headers)
84		83
85	Executes an HTTP-POST request with a request body of C<$bod>. See the	84	Executes an HTTP-POST request with a request body of C<$body>. See the
86	http_request function for details on additional parameters.	85	http_request function for details on additional parameters and the return
		86	value.
87		87
88	=item http_request $method => $url, key => value..., $cb->($data, $headers)	88	=item http_request $method => $url, key => value..., $cb->($data, $headers)
89		89
90	Executes a HTTP request of type C<$method> (e.g. C<GET>, C<POST>). The URL	90	Executes a HTTP request of type C<$method> (e.g. C<GET>, C<POST>). The URL
91	must be an absolute http or https URL.	91	must be an absolute http or https URL.
92		92
		93	When called in void context, nothing is returned. In other contexts,
		94	C<http_request> returns a "cancellation guard" - you have to keep the
		95	object at least alive until the callback get called. If the object gets
		96	destroyed before the callback is called, the request will be cancelled.
		97
93	The callback will be called with the response data as first argument	98	The callback will be called with the response body data as first argument
94	(or C<undef> if it wasn't available due to errors), and a hash-ref with	99	(or C<undef> if an error occured), and a hash-ref with response headers as
95	response headers as second argument.	100	second argument.
96		101
97	All the headers in that hash are lowercased. In addition to the response	102	All the headers in that hash are lowercased. In addition to the response
98	headers, the three "pseudo-headers" C<HTTPVersion>, C<Status> and	103	headers, the "pseudo-headers" (uppercase to avoid clashing with possible
		104	response headers) C<HTTPVersion>, C<Status> and C<Reason> contain the
99	C<Reason> contain the three parts of the HTTP Status-Line of the same	105	three parts of the HTTP Status-Line of the same name.
		106
		107	The pseudo-header C<URL> contains the actual URL (which can differ from
		108	the requested URL when following redirects - for example, you might get
		109	an error that your URL scheme is not supported even though your URL is a
		110	valid http URL because it redirected to an ftp URL, in which case you can
		111	look at the URL pseudo header).
		112
		113	The pseudo-header C<Redirect> only exists when the request was a result
		114	of an internal redirect. In that case it is an array reference with
		115	the C<($data, $headers)> from the redirect response. Note that this
		116	response could in turn be the result of a redirect itself, and C<<
		117	$headers->{Redirect}[1]{Redirect} >> will then contain the original
		118	response, and so on.
		119
100	name. If the server sends a header multiple lines, then their contents	120	If the server sends a header multiple times, then their contents will be
101	will be joined together with C<\x00>.	121	joined together with a comma (C<,>), as per the HTTP spec.
102		122
103	If an internal error occurs, such as not being able to resolve a hostname,	123	If an internal error occurs, such as not being able to resolve a hostname,
104	then C<$data> will be C<undef>, C<< $headers->{Status} >> will be C<599>	124	then C<$data> will be C<undef>, C<< $headers->{Status} >> will be C<59x>
105	and the C<Reason> pseudo-header will contain an error message.	125	(usually C<599>) and the C<Reason> pseudo-header will contain an error
		126	message.
106		127
107	A typical callback might look like this:	128	A typical callback might look like this:
108		129
109	sub {	130	sub {
110	my ($body, $hdr) = @_;	131	my ($body, $hdr) = @_;
…		…
128		149
129	=item headers => hashref	150	=item headers => hashref
130		151
131	The request headers to use. Currently, C<http_request> may provide its	152	The request headers to use. Currently, C<http_request> may provide its
132	own C<Host:>, C<Content-Length:>, C<Connection:> and C<Cookie:> headers	153	own C<Host:>, C<Content-Length:>, C<Connection:> and C<Cookie:> headers
133	and will provide defaults for C<User-Agent:> and C<Referer:>.	154	and will provide defaults for C<User-Agent:> and C<Referer:> (this can be
		155	suppressed by using C<undef> for these headers in which case they won't be
		156	sent at all).
134		157
135	=item timeout => $seconds	158	=item timeout => $seconds
136		159
137	The time-out to use for various stages - each connect attempt will reset	160	The time-out to use for various stages - each connect attempt will reset
138	the timeout, as will read or write activity. Default timeout is 5 minutes.	161	the timeout, as will read or write activity, i.e. this is not an overall
		162	timeout.
		163
		164	Default timeout is 5 minutes.
139		165
140	=item proxy => [$host, $port[, $scheme]] or undef	166	=item proxy => [$host, $port[, $scheme]] or undef
141		167
142	Use the given http proxy for all requests. If not specified, then the	168	Use the given http proxy for all requests. If not specified, then the
143	default proxy (as specified by C<$ENV{http_proxy}>) is used.	169	default proxy (as specified by C<$ENV{http_proxy}>) is used.
144		170
145	C<$scheme> must be either missing or C<http> for HTTP, or C<https> for	171	C<$scheme> must be either missing, C<http> for HTTP or C<https> for
146	HTTPS.	172	HTTPS.
147		173
148	=item body => $string	174	=item body => $string
149		175
150	The request body, usually empty. Will be-sent as-is (future versions of	176	The request body, usually empty. Will be-sent as-is (future versions of
…		…
156	based on the original netscape specification.	182	based on the original netscape specification.
157		183
158	The C<$hash_ref> must be an (initially empty) hash reference which will	184	The C<$hash_ref> must be an (initially empty) hash reference which will
159	get updated automatically. It is possible to save the cookie_jar to	185	get updated automatically. It is possible to save the cookie_jar to
160	persistent storage with something like JSON or Storable, but this is not	186	persistent storage with something like JSON or Storable, but this is not
161	recommended, as expire times are currently being ignored.	187	recommended, as expiry times are currently being ignored.
162		188
163	Note that this cookie implementation is not of very high quality, nor	189	Note that this cookie implementation is not of very high quality, nor
164	meant to be complete. If you want complete cookie management you have to	190	meant to be complete. If you want complete cookie management you have to
165	do that on your own. C<cookie_jar> is meant as a quick fix to get some	191	do that on your own. C<cookie_jar> is meant as a quick fix to get some
166	cookie-using sites working. Cookies are a privacy disaster, do not use	192	cookie-using sites working. Cookies are a privacy disaster, do not use
167	them unless required to.	193	them unless required to.
		194
		195	=item tls_ctx => $scheme \| $tls_ctx
		196
		197	Specifies the AnyEvent::TLS context to be used for https connections. This
		198	parameter follows the same rules as the C<tls_ctx> parameter to
		199	L<AnyEvent::Handle>, but additionally, the two strings C<low> or
		200	C<high> can be specified, which give you a predefined low-security (no
		201	verification, highest compatibility) and high-security (CA and common-name
		202	verification) TLS context.
		203
		204	The default for this option is C<low>, which could be interpreted as "give
		205	me the page, no matter what".
		206
		207	=item on_prepare => $callback->($fh)
		208
		209	In rare cases you need to "tune" the socket before it is used to
		210	connect (for exmaple, to bind it on a given IP address). This parameter
		211	overrides the prepare callback passed to C<AnyEvent::Socket::tcp_connect>
		212	and behaves exactly the same way (e.g. it has to provide a
		213	timeout). See the description for the C<$prepare_cb> argument of
		214	C<AnyEvent::Socket::tcp_connect> for details.
		215
		216	=item tcp_connect => $callback->($host, $service, $connect_cb, $prepare_cb)
		217
		218	In even rarer cases you want total control over how AnyEvent::HTTP
		219	establishes connections. Normally it uses L<AnyEvent::Socket::tcp_connect>
		220	to do this, but you can provide your own C<tcp_connect> function -
		221	obviously, it has to follow the same calling conventions, except that it
		222	may always return a connection guard object.
		223
		224	There are probably lots of weird uses for this function, starting from
		225	tracing the hosts C<http_request> actually tries to connect, to (inexact
		226	but fast) host => IP address caching or even socks protocol support.
		227
		228	=item on_header => $callback->($headers)
		229
		230	When specified, this callback will be called with the header hash as soon
		231	as headers have been successfully received from the remote server (not on
		232	locally-generated errors).
		233
		234	It has to return either true (in which case AnyEvent::HTTP will continue),
		235	or false, in which case AnyEvent::HTTP will cancel the download (and call
		236	the finish callback with an error code of C<598>).
		237
		238	This callback is useful, among other things, to quickly reject unwanted
		239	content, which, if it is supposed to be rare, can be faster than first
		240	doing a C<HEAD> request.
		241
		242	Example: cancel the request unless the content-type is "text/html".
		243
		244	on_header => sub {
		245	$_[0]{"content-type"} =~ /^text\/html\s*(?:;\|$)/
		246	},
		247
		248	=item on_body => $callback->($partial_body, $headers)
		249
		250	When specified, all body data will be passed to this callback instead of
		251	to the completion callback. The completion callback will get the empty
		252	string instead of the body data.
		253
		254	It has to return either true (in which case AnyEvent::HTTP will continue),
		255	or false, in which case AnyEvent::HTTP will cancel the download (and call
		256	the completion callback with an error code of C<598>).
		257
		258	This callback is useful when the data is too large to be held in memory
		259	(so the callback writes it to a file) or when only some information should
		260	be extracted, or when the body should be processed incrementally.
		261
		262	It is usually preferred over doing your own body handling via
		263	C<want_body_handle>, but in case of streaming APIs, where HTTP is
		264	only used to create a connection, C<want_body_handle> is the better
		265	alternative, as it allows you to install your own event handler, reducing
		266	resource usage.
		267
		268	=item want_body_handle => $enable
		269
		270	When enabled (default is disabled), the behaviour of AnyEvent::HTTP
		271	changes considerably: after parsing the headers, and instead of
		272	downloading the body (if any), the completion callback will be
		273	called. Instead of the C<$body> argument containing the body data, the
		274	callback will receive the L<AnyEvent::Handle> object associated with the
		275	connection. In error cases, C<undef> will be passed. When there is no body
		276	(e.g. status C<304>), the empty string will be passed.
		277
		278	The handle object might or might not be in TLS mode, might be connected to
		279	a proxy, be a persistent connection etc., and configured in unspecified
		280	ways. The user is responsible for this handle (it will not be used by this
		281	module anymore).
		282
		283	This is useful with some push-type services, where, after the initial
		284	headers, an interactive protocol is used (typical example would be the
		285	push-style twitter API which starts a JSON/XML stream).
		286
		287	If you think you need this, first have a look at C<on_body>, to see if
		288	that doesn't solve your problem in a better way.
168		289
169	=back	290	=back
170		291
171	Example: make a simple HTTP GET request for http://www.nethype.de/	292	Example: make a simple HTTP GET request for http://www.nethype.de/
172		293
…		…
185	my ($body, $hdr) = @_;	306	my ($body, $hdr) = @_;
186	use Data::Dumper;	307	use Data::Dumper;
187	print Dumper $hdr;	308	print Dumper $hdr;
188	}	309	}
189	;	310	;
		311
		312	Example: make another simple HTTP GET request, but immediately try to
		313	cancel it.
		314
		315	my $request = http_request GET => "http://www.nethype.de/", sub {
		316	my ($body, $hdr) = @_;
		317	print "$body\n";
		318	};
		319
		320	undef $request;
190		321
191	=cut	322	=cut
192		323
193	sub _slot_schedule;	324	sub _slot_schedule;
194	sub _slot_schedule($) {	325	sub _slot_schedule($) {
…		…
218	push @{ $CO_SLOT{$_[0]}[1] }, $_[1];	349	push @{ $CO_SLOT{$_[0]}[1] }, $_[1];
219		350
220	_slot_schedule $_[0];	351	_slot_schedule $_[0];
221	}	352	}
222		353
		354	our $qr_nlnl = qr{(?<![^\012])\015?\012};
		355
		356	our $TLS_CTX_LOW = { cache => 1, sslv2 => 1 };
		357	our $TLS_CTX_HIGH = { cache => 1, verify => 1, verify_peername => "https" };
		358
223	sub http_request($$@) {	359	sub http_request($$@) {
224	my $cb = pop;	360	my $cb = pop;
225	my ($method, $url, %arg) = @_;	361	my ($method, $url, %arg) = @_;
226		362
227	my %hdr;	363	my %hdr;
		364
		365	$arg{tls_ctx} = $TLS_CTX_LOW if $arg{tls_ctx} eq "low" \|\| !exists $arg{tls_ctx};
		366	$arg{tls_ctx} = $TLS_CTX_HIGH if $arg{tls_ctx} eq "high";
228		367
229	$method = uc $method;	368	$method = uc $method;
230		369
231	if (my $hdr = $arg{headers}) {	370	if (my $hdr = $arg{headers}) {
232	while (my ($k, $v) = each %$hdr) {	371	while (my ($k, $v) = each %$hdr) {
233	$hdr{lc $k} = $v;	372	$hdr{lc $k} = $v;
234	}	373	}
235	}	374	}
236		375
		376	# pseudo headers for all subsequent responses
		377	my @pseudo = (URL => $url);
		378	push @pseudo, Redirect => delete $arg{Redirect} if exists $arg{Redirect};
		379
237	my $recurse = exists $arg{recurse} ? $arg{recurse} : $MAX_RECURSE;	380	my $recurse = exists $arg{recurse} ? delete $arg{recurse} : $MAX_RECURSE;
238		381
239	return $cb->(undef, { Status => 599, Reason => "recursion limit reached", URL => $url })	382	return $cb->(undef, { Status => 599, Reason => "Too many redirections", @pseudo })
240	if $recurse < 0;	383	if $recurse < 0;
241		384
242	my $proxy = $arg{proxy} \|\| $PROXY;	385	my $proxy = $arg{proxy} \|\| $PROXY;
243	my $timeout = $arg{timeout} \|\| $TIMEOUT;	386	my $timeout = $arg{timeout} \|\| $TIMEOUT;
244		387
245	$hdr{"user-agent"} \|\|= $USERAGENT;
246
247	my ($scheme, $authority, $upath, $query, $fragment) =	388	my ($uscheme, $uauthority, $upath, $query, $fragment) =
248	$url =~ m\|(?:([^:/?#]+):)?(?://([^/?#]))?([^?#])(?:\?([^#]))?(?:#(.))?\|;	389	$url =~ m\|(?:([^:/?#]+):)?(?://([^/?#]))?([^?#])(?:(\?[^#]))?(?:#(.))?\|;
249		390
250	$scheme = lc $scheme;	391	$uscheme = lc $uscheme;
251		392
252	my $uport = $scheme eq "http" ? 80	393	my $uport = $uscheme eq "http" ? 80
253	: $scheme eq "https" ? 443	394	: $uscheme eq "https" ? 443
254	: return $cb->(undef, { Status => 599, Reason => "only http and https URL schemes supported", URL => $url });	395	: return $cb->(undef, { Status => 599, Reason => "Only http and https URL schemes supported", @pseudo });
255		396
256	$hdr{referer} \|\|= "$scheme://$authority$upath"; # leave out fragment and query string, just a heuristic
257
258	$authority =~ /^(?: .*\@ )? ([^\@:]+) (?: : (\d+) )?$/x	397	$uauthority =~ /^(?: .*\@ )? ([^\@:]+) (?: : (\d+) )?$/x
259	or return $cb->(undef, { Status => 599, Reason => "unparsable URL", URL => $url });	398	or return $cb->(undef, { Status => 599, Reason => "Unparsable URL", @pseudo });
260		399
261	my $uhost = $1;	400	my $uhost = $1;
262	$uport = $2 if defined $2;	401	$uport = $2 if defined $2;
263		402
		403	$hdr{host} = defined $2 ? "$uhost:$2" : "$uhost"
		404	unless exists $hdr{host};
		405
264	$uhost =~ s/^\[(.*)\]$/$1/;	406	$uhost =~ s/^\[(.*)\]$/$1/;
265	$upath .= "?$query" if length $query;	407	$upath .= $query if length $query;
266		408
267	$upath =~ s%^/?%/%;	409	$upath =~ s%^/?%/%;
268		410
269	# cookie processing	411	# cookie processing
270	if (my $jar = $arg{cookie_jar}) {	412	if (my $jar = $arg{cookie_jar}) {
271	%$jar = () if $jar->{version} < 1;	413	%$jar = () if $jar->{version} != 1;
272		414
273	my @cookie;	415	my @cookie;
274		416
275	while (my ($chost, $v) = each %$jar) {	417	while (my ($chost, $v) = each %$jar) {
		418	if ($chost =~ /^\./) {
276	next unless $chost eq substr $uhost, -length $chost;	419	next unless $chost eq substr $uhost, -length $chost;
277	next unless $chost =~ /^\./;	420	} elsif ($chost =~ /\./) {
		421	next unless $chost eq $uhost;
		422	} else {
		423	next;
		424	}
278		425
279	while (my ($cpath, $v) = each %$v) {	426	while (my ($cpath, $v) = each %$v) {
280	next unless $cpath eq substr $upath, 0, length $cpath;	427	next unless $cpath eq substr $upath, 0, length $cpath;
281		428
282	while (my ($k, $v) = each %$v) {	429	while (my ($k, $v) = each %$v) {
283	next if $scheme ne "https" && exists $v->{secure};	430	next if $uscheme ne "https" && exists $v->{secure};
		431	my $value = $v->{value};
		432	$value =~ s/([\\"])/\\$1/g;
284	push @cookie, "$k=$v->{value}";	433	push @cookie, "$k=\"$value\"";
285	}	434	}
286	}	435	}
287	}	436	}
288		437
289	$hdr{cookie} = join "; ", @cookie	438	$hdr{cookie} = join "; ", @cookie
290	if @cookie;	439	if @cookie;
291	}	440	}
292		441
293	my ($rhost, $rport, $rpath); # request host, port, path	442	my ($rhost, $rport, $rscheme, $rpath); # request host, port, path
294		443
295	if ($proxy) {	444	if ($proxy) {
296	($rhost, $rport, $scheme) = @$proxy;	445	($rpath, $rhost, $rport, $rscheme) = ($url, @$proxy);
297	$rpath = $url;	446
		447	$rscheme = "http" unless defined $rscheme;
		448
		449	# don't support https requests over https-proxy transport,
		450	# can't be done with tls as spec'ed, unless you double-encrypt.
		451	$rscheme = "http" if $uscheme eq "https" && $rscheme eq "https";
298	} else {	452	} else {
299	($rhost, $rport, $rpath) = ($uhost, $uport, $upath);	453	($rhost, $rport, $rscheme, $rpath) = ($uhost, $uport, $uscheme, $upath);
300	$hdr{host} = $uhost;
301	}	454	}
302		455
		456	# leave out fragment and query string, just a heuristic
		457	$hdr{referer} \|\|= "$uscheme://$uauthority$upath" unless exists $hdr{referer};
		458	$hdr{"user-agent"} \|\|= $USERAGENT unless exists $hdr{"user-agent"};
		459
303	$hdr{"content-length"} = length $arg{body};	460	$hdr{"content-length"} = length $arg{body}
		461	if length $arg{body} \|\| $method ne "GET";
304		462
305	my %state = (connect_guard => 1);	463	my %state = (connect_guard => 1);
306		464
307	_get_slot $uhost, sub {	465	_get_slot $uhost, sub {
308	$state{slot_guard} = shift;	466	$state{slot_guard} = shift;
309		467
310	return unless $state{connect_guard};	468	return unless $state{connect_guard};
311		469
312	$state{connect_guard} = AnyEvent::Socket::tcp_connect $rhost, $rport, sub {	470	my $tcp_connect = $arg{tcp_connect}
		471	\|\| do { require AnyEvent::Socket; \&AnyEvent::Socket::tcp_connect };
		472
		473	$state{connect_guard} = $tcp_connect->(
		474	$rhost,
		475	$rport,
		476	sub {
313	$state{fh} = shift	477	$state{fh} = shift
314	or return $cb->(undef, { Status => 599, Reason => "$!", URL => $url });	478	or do {
315
316	delete $state{connect_guard}; # reduce memory usage, save a tree
317
318	# get handle
319	$state{handle} = new AnyEvent::Handle
320	fh => $state{fh},
321	($scheme eq "https" ? (tls => "connect") : ());
322
323	# limit the number of persistent connections
324	if ($KA_COUNT{$_[1]} < $MAX_PERSISTENT_PER_HOST) {
325	++$KA_COUNT{$_[1]};
326	$state{handle}{ka_count_guard} = AnyEvent::Util::guard { --$KA_COUNT{$_[1]} };
327	$hdr{connection} = "keep-alive";
328	delete $hdr{connection}; # keep-alive not yet supported
329	} else {
330	delete $hdr{connection};
331	}
332
333	# (re-)configure handle
334	$state{handle}->timeout ($timeout);
335	$state{handle}->on_error (sub {
336	my $errno = "$!";	479	my $err = "$!";
337	%state = ();
338	$cb->(undef, { Status => 599, Reason => $errno, URL => $url });
339	});
340	$state{handle}->on_eof (sub {
341	%state = ();
342	$cb->(undef, { Status => 599, Reason => "unexpected end-of-file", URL => $url });
343	});
344
345	# send request
346	$state{handle}->push_write (
347	"$method $rpath HTTP/1.0\015\012"
348	. (join "", map "$_: $hdr{$_}\015\012", keys %hdr)
349	. "\015\012"
350	. (delete $arg{body})
351	);
352
353	%hdr = (); # reduce memory usage, save a kitten
354
355	# status line
356	$state{handle}->push_read (line => qr/\015?\012/, sub {
357	$_[1] =~ /^HTTP\/([0-9\.]+) \s+ ([0-9]{3}) \s+ ([^\015\012]+)/ix
358	or return (%state = (), $cb->(undef, { Status => 599, Reason => "invalid server response ($_[1])", URL => $url }));
359
360	my %hdr = ( # response headers
361	HTTPVersion => "\x00$1",
362	Status => "\x00$2",
363	Reason => "\x00$3",
364	URL => "\x00$url"
365	);
366
367	# headers, could be optimized a bit
368	$state{handle}->unshift_read (line => qr/\015?\012\015?\012/, sub {
369	for ("$_[1]\012") {
370	# we support spaces in field names, as lotus domino
371	# creates them.
372	$hdr{lc $1} .= "\x00$2"
373	while /\G
374	([^:\000-\037]+):
375	[\011\040]*
376	((?: [^\015\012]+ \| \015?\012[\011\040] )*)
377	\015?\012
378	/gxc;
379
380	/\G$/
381	or return (%state = (), $cb->(undef, { Status => 599, Reason => "garbled response headers", URL => $url }));
382	}
383
384	substr $_, 0, 1, ""
385	for values %hdr;
386
387	my $finish = sub {
388	%state = ();	480	%state = ();
		481	return $cb->(undef, { Status => 599, Reason => $err, @pseudo });
		482	};
389		483
390	# set-cookie processing	484	pop; # free memory, save a tree
391	if ($arg{cookie_jar} && exists $hdr{"set-cookie"}) {
392	for (split /\x00/, $hdr{"set-cookie"}) {
393	my ($cookie, @arg) = split /;\s*/;
394	my ($name, $value) = split /=/, $cookie, 2;
395	my %kv = (value => $value, map { split /=/, $_, 2 } @arg);
396
397	my $cdom = (delete $kv{domain}) \|\| $uhost;
398	my $cpath = (delete $kv{path}) \|\| "/";
399
400	$cdom =~ s/^.?/./; # make sure it starts with a "."
401		485
402	next if $cdom =~ /\.$/;	486	return unless delete $state{connect_guard};
403		487
404	# this is not rfc-like and not netscape-like. go figure.	488	# get handle
405	my $ndots = $cdom =~ y/.//;	489	$state{handle} = new AnyEvent::Handle
406	next if $ndots < ($cdom =~ /\.[^.][^.]\.[^.][^.]$/ ? 3 : 2);	490	fh => $state{fh},
407		491	peername => $rhost,
		492	tls_ctx => $arg{tls_ctx},
		493	# these need to be reconfigured on keepalive handles
		494	timeout => $timeout,
		495	on_error => sub {
		496	%state = ();
		497	$cb->(undef, { Status => 599, Reason => $_[2], @pseudo });
		498	},
		499	on_eof => sub {
		500	%state = ();
		501	$cb->(undef, { Status => 599, Reason => "Unexpected end-of-file", @pseudo });
		502	},
		503	;
		504
		505	# limit the number of persistent connections
		506	# keepalive not yet supported
		507	# if ($KA_COUNT{$_[1]} < $MAX_PERSISTENT_PER_HOST) {
		508	# ++$KA_COUNT{$_[1]};
		509	# $state{handle}{ka_count_guard} = AnyEvent::Util::guard {
		510	# --$KA_COUNT{$_[1]}
		511	# };
		512	# $hdr{connection} = "keep-alive";
		513	# } else {
		514	delete $hdr{connection};
		515	# }
		516
		517	$state{handle}->starttls ("connect") if $rscheme eq "https";
		518
		519	# handle actual, non-tunneled, request
		520	my $handle_actual_request = sub {
		521	$state{handle}->starttls ("connect") if $uscheme eq "https" && !exists $state{handle}{tls};
		522
		523	# send request
		524	$state{handle}->push_write (
		525	"$method $rpath HTTP/1.0\015\012"
		526	. (join "", map "\u$_: $hdr{$_}\015\012", grep defined $hdr{$_}, keys %hdr)
		527	. "\015\012"
		528	. (delete $arg{body})
		529	);
		530
		531	# return if error occured during push_write()
		532	return unless %state;
		533
		534	%hdr = (); # reduce memory usage, save a kitten, also make it possible to re-use
		535
		536	# status line and headers
		537	$state{handle}->push_read (line => $qr_nlnl, sub {
		538	for ("$_[1]") {
		539	y/\015//d; # weed out any \015, as they show up in the weirdest of places.
		540
		541	/^HTTP\/([0-9\.]+) \s+ ([0-9]{3}) (?: \s+ ([^\015\012]*) )? \015?\012/igxc
		542	or return (%state = (), $cb->(undef, { Status => 599, Reason => "Invalid server response", @pseudo }));
		543
		544	push @pseudo,
		545	HTTPVersion => $1,
		546	Status => $2,
		547	Reason => $3,
		548	;
		549
		550	# things seen, not parsed:
		551	# p3pP="NON CUR OTPi OUR NOR UNI"
		552
		553	$hdr{lc $1} .= ",$2"
		554	while /\G
		555	([^:\000-\037]*):
		556	[\011\040]*
		557	((?: [^\012]+ \| \012[\011\040] )*)
		558	\012
		559	/gxc;
		560
		561	/\G$/
		562	or return (%state = (), $cb->(undef, { Status => 599, Reason => "Garbled response headers", @pseudo }));
		563	}
		564
		565	# remove the "," prefix we added to all headers above
		566	substr $_, 0, 1, ""
		567	for values %hdr;
		568
		569	# patch in all pseudo headers
		570	%hdr = (%hdr, @pseudo);
		571
		572	# redirect handling
		573	# microsoft and other shitheads don't give a shit for following standards,
		574	# try to support some common forms of broken Location headers.
		575	if ($hdr{location} !~ /^(?: $ \| [^:\/?\#]+ : )/x) {
		576	$hdr{location} =~ s/^\.\/+//;
		577
		578	my $url = "$rscheme://$uhost:$uport";
		579
		580	unless ($hdr{location} =~ s/^\///) {
		581	$url .= $upath;
		582	$url =~ s/\/[^\/]*$//;
		583	}
		584
		585	$hdr{location} = "$url/$hdr{location}";
		586	}
		587
		588	my $redirect;
		589
		590	if ($recurse) {
		591	my $status = $hdr{Status};
		592
		593	# industry standard is to redirect POST as GET for
		594	# 301, 302 and 303, in contrast to http/1.0 and 1.1.
		595	# also, the UA should ask the user for 301 and 307 and POST,
		596	# industry standard seems to be to simply follow.
		597	# we go with the industry standard.
		598	if ($status == 301 or $status == 302 or $status == 303) {
		599	# HTTP/1.1 is unclear on how to mutate the method
		600	$method = "GET" unless $method eq "HEAD";
408	# store it	601	$redirect = 1;
409	$arg{cookie_jar}{version} = 1;	602	} elsif ($status == 307) {
410	$arg{cookie_jar}{$cdom}{$cpath}{$name} = \%kv;	603	$redirect = 1;
411	}	604	}
412	}	605	}
413		606
414	if ($_[1]{Status} =~ /^30[12]$/ && $recurse) {	607	my $finish = sub {
415	# microsoft and other assholes don't give a shit for following standards,	608	$state{handle}->destroy if $state{handle};
416	# try to support a common form of broken Location header.	609	%state = ();
417	$_[1]{location} =~ s%^/%$scheme://$uhost:$uport/%;
418		610
419	http_request ($method, $_[1]{location}, %arg, recurse => $recurse - 1, $cb);	611	# set-cookie processing
		612	if ($arg{cookie_jar}) {
		613	for ($_[1]{"set-cookie"}) {
		614	# parse NAME=VALUE
		615	my @kv;
		616
		617	while (/\G\s* ([^=;,[:space:]]+) \s=\s (?: "((?:[^\\"]+\|\\.))" \| ([^=;,[:space:]]) )/gcxs) {
		618	my $name = $1;
		619	my $value = $3;
		620
		621	unless ($value) {
		622	$value = $2;
		623	$value =~ s/\\(.)/$1/gs;
		624	}
		625
		626	push @kv, $name => $value;
		627
		628	last unless /\G\s*;/gc;
		629	}
		630
		631	last unless @kv;
		632
		633	my $name = shift @kv;
		634	my %kv = (value => shift @kv, @kv);
		635
		636	my $cdom;
		637	my $cpath = (delete $kv{path}) \|\| "/";
		638
		639	if (exists $kv{domain}) {
		640	$cdom = delete $kv{domain};
		641
		642	$cdom =~ s/^\.?/./; # make sure it starts with a "."
		643
		644	next if $cdom =~ /\.$/;
		645
		646	# this is not rfc-like and not netscape-like. go figure.
		647	my $ndots = $cdom =~ y/.//;
		648	next if $ndots < ($cdom =~ /\.[^.][^.]\.[^.][^.]$/ ? 3 : 2);
		649	} else {
		650	$cdom = $uhost;
		651	}
		652
		653	# store it
		654	$arg{cookie_jar}{version} = 1;
		655	$arg{cookie_jar}{$cdom}{$cpath}{$name} = \%kv;
		656
		657	redo if /\G\s*,/gc;
		658	}
		659	}
		660
		661	if ($redirect && exists $hdr{location}) {
		662	# we ignore any errors, as it is very common to receive
		663	# Content-Length != 0 but no actual body
		664	# we also access %hdr, as $_[1] might be an erro
		665	http_request (
		666	$method => $hdr{location},
		667	%arg,
		668	recurse => $recurse - 1,
		669	Redirect => \@_,
		670	$cb);
		671	} else {
		672	$cb->($_[0], $_[1]);
		673	}
		674	};
		675
		676	my $len = $hdr{"content-length"};
		677
		678	if (!$redirect && $arg{on_header} && !$arg{on_header}(\%hdr)) {
		679	$finish->(undef, { Status => 598, Reason => "Request cancelled by on_header", @pseudo });
		680	} elsif (
		681	$hdr{Status} =~ /^(?:1..\|[23]04)$/
		682	or $method eq "HEAD"
		683	or (defined $len && !$len)
		684	) {
		685	# no body
		686	$finish->("", \%hdr);
420	} else {	687	} else {
421	$cb->($_[0], $_[1]);	688	# body handling, four different code paths
		689	# for want_body_handle, on_body (2x), normal (2x)
		690	# we might read too much here, but it does not matter yet (no pers. connections)
		691	if (!$redirect && $arg{want_body_handle}) {
		692	$_[0]->on_eof (undef);
		693	$_[0]->on_error (undef);
		694	$_[0]->on_read (undef);
		695
		696	$finish->(delete $state{handle}, \%hdr);
		697
		698	} elsif ($arg{on_body}) {
		699	$_[0]->on_error (sub { $finish->(undef, { Status => 599, Reason => $_[2], @pseudo }) });
		700	if ($len) {
		701	$_[0]->on_eof (undef);
		702	$_[0]->on_read (sub {
		703	$len -= length $_[0]{rbuf};
		704
		705	$arg{on_body}(delete $_[0]{rbuf}, \%hdr)
		706	or $finish->(undef, { Status => 598, Reason => "Request cancelled by on_body", @pseudo });
		707
		708	$len > 0
		709	or $finish->("", \%hdr);
		710	});
		711	} else {
		712	$_[0]->on_eof (sub {
		713	$finish->("", \%hdr);
		714	});
		715	$_[0]->on_read (sub {
		716	$arg{on_body}(delete $_[0]{rbuf}, \%hdr)
		717	or $finish->(undef, { Status => 598, Reason => "Request cancelled by on_body", @pseudo });
		718	});
		719	}
		720	} else {
		721	$_[0]->on_eof (undef);
		722
		723	if ($len) {
		724	$_[0]->on_error (sub { $finish->(undef, { Status => 599, Reason => $_[2], @pseudo }) });
		725	$_[0]->on_read (sub {
		726	$finish->((substr delete $_[0]{rbuf}, 0, $len, ""), \%hdr)
		727	if $len <= length $_[0]{rbuf};
		728	});
		729	} else {
		730	$_[0]->on_error (sub {
		731	($! == Errno::EPIPE \|\| !$!)
		732	? $finish->(delete $_[0]{rbuf}, \%hdr)
		733	: $finish->(undef, { Status => 599, Reason => $_[2], @pseudo });
		734	});
		735	$_[0]->on_read (sub { });
		736	}
		737	}
422	}	738	}
423	};	739	});
		740	};
424		741
425	if ($hdr{Status} =~ /^(?:1..\|204\|304)$/ or $method eq "HEAD") {	742	# now handle proxy-CONNECT method
426	$finish->(undef, \%hdr);	743	if ($proxy && $uscheme eq "https") {
427	} else {	744	# oh dear, we have to wrap it into a connect request
428	if (exists $hdr{"content-length"}) {
429	$_[0]->unshift_read (chunk => $hdr{"content-length"}, sub {
430	# could cache persistent connection now
431	if ($hdr{connection} =~ /\bkeep-alive\b/i) {
432	# but we don't, due to misdesigns, this is annoyingly complex
433	};
434		745
435	$finish->($_[1], \%hdr);	746	# maybe re-use $uauthority with patched port?
436	});	747	$state{handle}->push_write ("CONNECT $uhost:$uport HTTP/1.0\015\012Host: $uhost\015\012\015\012");
		748	$state{handle}->push_read (line => $qr_nlnl, sub {
		749	$_[1] =~ /^HTTP\/([0-9\.]+) \s+ ([0-9]{3}) (?: \s+ ([^\015\012]*) )?/ix
		750	or return (%state = (), $cb->(undef, { Status => 599, Reason => "Invalid proxy connect response ($_[1])", @pseudo }));
		751
		752	if ($2 == 200) {
		753	$rpath = $upath;
		754	&$handle_actual_request;
437	} else {	755	} else {
438	# too bad, need to read until we get an error or EOF,
439	# no way to detect winged data.
440	$_[0]->on_error (sub {
441	$finish->($_[0]{rbuf}, \%hdr);
442	});	756	%state = ();
443	$_[0]->on_eof (undef);	757	$cb->(undef, { Status => $2, Reason => $3, @pseudo });
444	$_[0]->on_read (sub { });
445	}	758	}
446	}	759	});
		760	} else {
		761	&$handle_actual_request;
447	});	762	}
		763
448	});	764	},
449	}, sub {	765	$arg{on_prepare} \|\| sub { $timeout }
450	$timeout
451	};	766	);
452	};	767	};
453		768
454	defined wantarray && AnyEvent::Util::guard { %state = () }	769	defined wantarray && AnyEvent::Util::guard { %state = () }
455	}	770	}
456		771
…		…
463	unshift @_, "HEAD";	778	unshift @_, "HEAD";
464	&http_request	779	&http_request
465	}	780	}
466		781
467	sub http_post($$@) {	782	sub http_post($$@) {
		783	my $url = shift;
468	unshift @_, "POST", "body";	784	unshift @_, "POST", $url, "body";
469	&http_request	785	&http_request
470	}	786	}
471		787
472	=back	788	=back
473		789
		790	=head2 DNS CACHING
		791
		792	AnyEvent::HTTP uses the AnyEvent::Socket::tcp_connect function for
		793	the actual connection, which in turn uses AnyEvent::DNS to resolve
		794	hostnames. The latter is a simple stub resolver and does no caching
		795	on its own. If you want DNS caching, you currently have to provide
		796	your own default resolver (by storing a suitable resolver object in
		797	C<$AnyEvent::DNS::RESOLVER>).
		798
474	=head2 GLOBAL FUNCTIONS AND VARIABLES	799	=head2 GLOBAL FUNCTIONS AND VARIABLES
475		800
476	=over 4	801	=over 4
477		802
478	=item AnyEvent::HTTP::set_proxy "proxy-url"	803	=item AnyEvent::HTTP::set_proxy "proxy-url"
479		804
480	Sets the default proxy server to use. The proxy-url must begin with a	805	Sets the default proxy server to use. The proxy-url must begin with a
481	string of the form C<http://host:port> (optionally C<https:...>).	806	string of the form C<http://host:port> (optionally C<https:...>), croaks
		807	otherwise.
		808
		809	To clear an already-set proxy, use C<undef>.
		810
		811	=item $date = AnyEvent::HTTP::format_date $timestamp
		812
		813	Takes a POSIX timestamp (seconds since the epoch) and formats it as a HTTP
		814	Date (RFC 2616).
		815
		816	=item $timestamp = AnyEvent::HTTP::parse_date $date
		817
		818	Takes a HTTP Date (RFC 2616) and returns the corresponding POSIX
		819	timestamp, or C<undef> if the date cannot be parsed.
482		820
483	=item $AnyEvent::HTTP::MAX_RECURSE	821	=item $AnyEvent::HTTP::MAX_RECURSE
484		822
485	The default value for the C<recurse> request parameter (default: C<10>).	823	The default value for the C<recurse> request parameter (default: C<10>).
486		824
487	=item $AnyEvent::HTTP::USERAGENT	825	=item $AnyEvent::HTTP::USERAGENT
488		826
489	The default value for the C<User-Agent> header (the default is	827	The default value for the C<User-Agent> header (the default is
490	C<Mozilla/5.0 (compatible; AnyEvent::HTTP/$VERSION; +http://software.schmorp.de/pkg/AnyEvent)>).	828	C<Mozilla/5.0 (compatible; U; AnyEvent-HTTP/$VERSION; +http://software.schmorp.de/pkg/AnyEvent)>).
491		829
492	=item $AnyEvent::HTTP::MAX_PERSISTENT	830	=item $AnyEvent::HTTP::MAX_PER_HOST
493		831
494	The maximum number of persistent connections to keep open (default: 8).	832	The maximum number of concurrent connections to the same host (identified
		833	by the hostname). If the limit is exceeded, then the additional requests
		834	are queued until previous connections are closed.
495		835
496	Not implemented currently.	836	The default value for this is C<4>, and it is highly advisable to not
497		837	increase it.
498	=item $AnyEvent::HTTP::PERSISTENT_TIMEOUT
499
500	The maximum time to cache a persistent connection, in seconds (default: 2).
501
502	Not implemented currently.
503		838
504	=item $AnyEvent::HTTP::ACTIVE	839	=item $AnyEvent::HTTP::ACTIVE
505		840
506	The number of active connections. This is not the number of currently	841	The number of active connections. This is not the number of currently
507	running requests, but the number of currently open and non-idle TCP	842	running requests, but the number of currently open and non-idle TCP
…		…
509		844
510	=back	845	=back
511		846
512	=cut	847	=cut
513		848
		849	our @month = qw(Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec);
		850	our @weekday = qw(Sun Mon Tue Wed Thu Fri Sat);
		851
		852	sub format_date($) {
		853	my ($time) = @_;
		854
		855	# RFC 822/1123 format
		856	my ($S, $M, $H, $mday, $mon, $year, $wday, $yday, undef) = gmtime $time;
		857
		858	sprintf "%s, %02d %s %04d %02d:%02d:%02d GMT",
		859	$weekday[$wday], $mday, $month[$mon], $year + 1900,
		860	$H, $M, $S;
		861	}
		862
		863	sub parse_date($) {
		864	my ($date) = @_;
		865
		866	my ($d, $m, $y, $H, $M, $S);
		867
		868	if ($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$/) {
		869	# RFC 822/1123, required by RFC 2616
		870	($d, $m, $y, $H, $M, $S) = ($1, $2, $3, $4, $5, $6);
		871
		872	} elsif ($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$/) {
		873	# RFC 850
		874	($d, $m, $y, $H, $M, $S) = ($1, $2, $3 < 69 ? $3 + 2000 : $3 + 1900, $4, $5, $6);
		875
		876	} elsif ($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])$/) {
		877	# ISO C's asctime
		878	($d, $m, $y, $H, $M, $S) = ($2, $1, $6, $3, $4, $5);
		879	}
		880	# other formats fail in the loop below
		881
		882	for (0..11) {
		883	if ($m eq $month[$_]) {
		884	require Time::Local;
		885	return Time::Local::timegm ($S, $M, $H, $d, $_, $y);
		886	}
		887	}
		888
		889	undef
		890	}
		891
514	sub set_proxy($) {	892	sub set_proxy($) {
		893	if (length $_[0]) {
515	$PROXY = [$2, $3 \|\| 3128, $1] if $_[0] =~ m%^(https?):// ([^:/]+) (?: : (\d*) )?%ix;	894	$_[0] =~ m%^(https?):// ([^:/]+) (?: : (\d*) )?%ix
		895	or Carp::croak "$_[0]: invalid proxy URL";
		896	$PROXY = [$2, $3 \|\| 3128, $1]
		897	} else {
		898	undef $PROXY;
		899	}
516	}	900	}
517		901
518	# initialise proxy from environment	902	# initialise proxy from environment
		903	eval {
519	set_proxy $ENV{http_proxy};	904	set_proxy $ENV{http_proxy};
		905	};
		906
		907	=head2 SOCKS PROXIES
		908
		909	Socks proxies are not directly supported by AnyEvent::HTTP. You can
		910	compile your perl to support socks, or use an external program such as
		911	F<socksify> (dante) or F<tsocks> to make your program use a socks proxy
		912	transparently.
		913
		914	Alternatively, for AnyEvent::HTTP only, you can use your own
		915	C<tcp_connect> function that does the proxy handshake - here is an example
		916	that works with socks4a proxies:
		917
		918	use Errno;
		919	use AnyEvent::Util;
		920	use AnyEvent::Socket;
		921	use AnyEvent::Handle;
		922
		923	# host, port and username of/for your socks4a proxy
		924	my $socks_host = "10.0.0.23";
		925	my $socks_port = 9050;
		926	my $socks_user = "";
		927
		928	sub socks4a_connect {
		929	my ($host, $port, $connect_cb, $prepare_cb) = @_;
		930
		931	my $hdl = new AnyEvent::Handle
		932	connect => [$socks_host, $socks_port],
		933	on_prepare => sub { $prepare_cb->($_[0]{fh}) },
		934	on_error => sub { $connect_cb->() },
		935	;
		936
		937	$hdl->push_write (pack "CCnNZZ", 4, 1, $port, 1, $socks_user, $host);
		938
		939	$hdl->push_read (chunk => 8, sub {
		940	my ($hdl, $chunk) = @_;
		941	my ($status, $port, $ipn) = unpack "xCna4", $chunk;
		942
		943	if ($status == 0x5a) {
		944	$connect_cb->($hdl->{fh}, (format_address $ipn) . ":$port");
		945	} else {
		946	$! = Errno::ENXIO; $connect_cb->();
		947	}
		948	});
		949
		950	$hdl
		951	}
		952
		953	Use C<socks4a_connect> instead of C<tcp_connect> when doing C<http_request>s,
		954	possibly after switching off other proxy types:
		955
		956	AnyEvent::HTTP::set_proxy undef; # usually you do not want other proxies
		957
		958	http_get 'http://www.google.com', tcp_connect => \&socks4a_connect, sub {
		959	my ($data, $headers) = @_;
		960	...
		961	};
520		962
521	=head1 SEE ALSO	963	=head1 SEE ALSO
522		964
523	L<AnyEvent>.	965	L<AnyEvent>.
524		966
525	=head1 AUTHOR	967	=head1 AUTHOR
526		968
527	Marc Lehmann <schmorp@schmorp.de>	969	Marc Lehmann <schmorp@schmorp.de>
528	http://home.schmorp.de/	970	http://home.schmorp.de/
529		971
		972	With many thanks to Дмитрий Шалашов, who provided countless
		973	testcases and bugreports.
		974
530	=cut	975	=cut
531		976
532	1	977	1
533		978

Diff Legend

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

Comparing AnyEvent-HTTP/HTTP.pm (file contents): Revision 1.19 by elmex, Mon Jun 9 13:02:13 2008 UTC vs. Revision 1.63 by root, Thu Dec 30 04:31:55 2010 UTC

Diff Legend

Comparing AnyEvent-HTTP/HTTP.pm (file contents):
Revision 1.19 by elmex, Mon Jun 9 13:02:13 2008 UTC vs.
Revision 1.63 by root, Thu Dec 30 04:31:55 2010 UTC