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

Comparing AnyEvent-HTTP/HTTP.pm (file contents):
Revision 1.30 by root, Thu Oct 23 02:46:20 2008 UTC vs.
Revision 1.55 by root, Wed Jun 16 19:17:30 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 ();	48	use AnyEvent::Socket ();
49	use AnyEvent::Handle ();	49	use AnyEvent::Handle ();
50		50
51	use base Exporter::;	51	use base Exporter::;
52		52
53	our $VERSION = '1.05';	53	our $VERSION = '1.45';
54		54
55	our @EXPORT = qw(http_get http_post http_head http_request);	55	our @EXPORT = qw(http_get http_post http_head http_request);
56		56
57	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)";
58	our $MAX_RECURSE = 10;	58	our $MAX_RECURSE = 10;
59	our $MAX_PERSISTENT = 8;	59	our $MAX_PERSISTENT = 8;
60	our $PERSISTENT_TIMEOUT = 2;	60	our $PERSISTENT_TIMEOUT = 2;
61	our $TIMEOUT = 300;	61	our $TIMEOUT = 300;
62		62
63	# changing these is evil	63	# changing these is evil
64	our $MAX_PERSISTENT_PER_HOST = 2;	64	our $MAX_PERSISTENT_PER_HOST = 0;
65	our $MAX_PER_HOST = 4;	65	our $MAX_PER_HOST = 4;
66		66
67	our $PROXY;	67	our $PROXY;
68	our $ACTIVE = 0;	68	our $ACTIVE = 0;
69		69
…		…
94	When called in void context, nothing is returned. In other contexts,	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	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	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.	97	destroyed before the callbakc is called, the request will be cancelled.
98		98
99	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
100	(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
101	response headers as second argument.	101	second argument.
102		102
103	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
104	headers, the "pseudo-headers" C<HTTPVersion>, C<Status> and C<Reason>	104	headers, the "pseudo-headers" (uppercase to avoid clashing with possible
		105	response headers) C<HTTPVersion>, C<Status> and C<Reason> contain the
105	contain the three parts of the HTTP Status-Line of the same name. The	106	three parts of the HTTP Status-Line of the same name.
		107
106	pseudo-header C<URL> contains the original URL (which can differ from the	108	The pseudo-header C<URL> contains the actual URL (which can differ from
107	requested URL when following redirects).	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).
108		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	$headers->{Redirect}[1]{Redirect} >> will then contain the original
		119	response, and so on.
		120
109	If the server sends a header multiple lines, then their contents will be	121	If the server sends a header multiple times, then their contents will be
110	joined together with C<\x00>.	122	joined together with a comma (C<,>), as per the HTTP spec.
111		123
112	If an internal error occurs, such as not being able to resolve a hostname,	124	If an internal error occurs, such as not being able to resolve a hostname,
113	then C<$data> will be C<undef>, C<< $headers->{Status} >> will be C<599>	125	then C<$data> will be C<undef>, C<< $headers->{Status} >> will be C<59x>
114	and the C<Reason> pseudo-header will contain an error message.	126	(usually C<599>) and the C<Reason> pseudo-header will contain an error
		127	message.
115		128
116	A typical callback might look like this:	129	A typical callback might look like this:
117		130
118	sub {	131	sub {
119	my ($body, $hdr) = @_;	132	my ($body, $hdr) = @_;
…		…
137		150
138	=item headers => hashref	151	=item headers => hashref
139		152
140	The request headers to use. Currently, C<http_request> may provide its	153	The request headers to use. Currently, C<http_request> may provide its
141	own C<Host:>, C<Content-Length:>, C<Connection:> and C<Cookie:> headers	154	own C<Host:>, C<Content-Length:>, C<Connection:> and C<Cookie:> headers
142	and will provide defaults for C<User-Agent:> and C<Referer:>.	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).
143		158
144	=item timeout => $seconds	159	=item timeout => $seconds
145		160
146	The time-out to use for various stages - each connect attempt will reset	161	The time-out to use for various stages - each connect attempt will reset
147	the timeout, as will read or write activity. Default timeout is 5 minutes.	162	the timeout, as will read or write activity, i.e. this is not an overall
		163	timeout.
		164
		165	Default timeout is 5 minutes.
148		166
149	=item proxy => [$host, $port[, $scheme]] or undef	167	=item proxy => [$host, $port[, $scheme]] or undef
150		168
151	Use the given http proxy for all requests. If not specified, then the	169	Use the given http proxy for all requests. If not specified, then the
152	default proxy (as specified by C<$ENV{http_proxy}>) is used.	170	default proxy (as specified by C<$ENV{http_proxy}>) is used.
153		171
154	C<$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
155	HTTPS.	173	HTTPS.
156		174
157	=item body => $string	175	=item body => $string
158		176
159	The request body, usually empty. Will be-sent as-is (future versions of	177	The request body, usually empty. Will be-sent as-is (future versions of
…		…
165	based on the original netscape specification.	183	based on the original netscape specification.
166		184
167	The C<$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
168	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
169	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
170	recommended, as expire times are currently being ignored.	188	recommended, as expiry times are currently being ignored.
171		189
172	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
173	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
174	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
175	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
176	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.
177		278
178	=back	279	=back
179		280
180	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/
181		282
…		…
237	push @{ $CO_SLOT{$_[0]}[1] }, $_[1];	338	push @{ $CO_SLOT{$_[0]}[1] }, $_[1];
238		339
239	_slot_schedule $_[0];	340	_slot_schedule $_[0];
240	}	341	}
241		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
242	sub http_request($$@) {	349	sub http_request($$@) {
243	my $cb = pop;	350	my $cb = pop;
244	my ($method, $url, %arg) = @_;	351	my ($method, $url, %arg) = @_;
245		352
246	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";
247		357
248	$method = uc $method;	358	$method = uc $method;
249		359
250	if (my $hdr = $arg{headers}) {	360	if (my $hdr = $arg{headers}) {
251	while (my ($k, $v) = each %$hdr) {	361	while (my ($k, $v) = each %$hdr) {
252	$hdr{lc $k} = $v;	362	$hdr{lc $k} = $v;
253	}	363	}
254	}	364	}
255		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
256	my $recurse = exists $arg{recurse} ? delete $arg{recurse} : $MAX_RECURSE;	370	my $recurse = exists $arg{recurse} ? delete $arg{recurse} : $MAX_RECURSE;
257		371
258	return $cb->(undef, { Status => 599, Reason => "recursion limit reached", URL => $url })	372	return $cb->(undef, { Status => 599, Reason => "Too many redirections", @pseudo })
259	if $recurse < 0;	373	if $recurse < 0;
260		374
261	my $proxy = $arg{proxy} \|\| $PROXY;	375	my $proxy = $arg{proxy} \|\| $PROXY;
262	my $timeout = $arg{timeout} \|\| $TIMEOUT;	376	my $timeout = $arg{timeout} \|\| $TIMEOUT;
263		377
264	$hdr{"user-agent"} \|\|= $USERAGENT;
265
266	my ($scheme, $authority, $upath, $query, $fragment) =	378	my ($uscheme, $uauthority, $upath, $query, $fragment) =
267	$url =~ m\|(?:([^:/?#]+):)?(?://([^/?#]))?([^?#])(?:\?([^#]))?(?:#(.))?\|;	379	$url =~ m\|(?:([^:/?#]+):)?(?://([^/?#]))?([^?#])(?:\?([^#]))?(?:#(.))?\|;
268		380
269	$scheme = lc $scheme;	381	$uscheme = lc $uscheme;
270		382
271	my $uport = $scheme eq "http" ? 80	383	my $uport = $uscheme eq "http" ? 80
272	: $scheme eq "https" ? 443	384	: $uscheme eq "https" ? 443
273	: return $cb->(undef, { Status => 599, Reason => "only http and https URL schemes supported", URL => $url });	385	: return $cb->(undef, { Status => 599, Reason => "Only http and https URL schemes supported", @pseudo });
274		386
275	$hdr{referer} \|\|= "$scheme://$authority$upath"; # leave out fragment and query string, just a heuristic
276
277	$authority =~ /^(?: .*\@ )? ([^\@:]+) (?: : (\d+) )?$/x	387	$uauthority =~ /^(?: .*\@ )? ([^\@:]+) (?: : (\d+) )?$/x
278	or return $cb->(undef, { Status => 599, Reason => "unparsable URL", URL => $url });	388	or return $cb->(undef, { Status => 599, Reason => "Unparsable URL", @pseudo });
279		389
280	my $uhost = $1;	390	my $uhost = $1;
281	$uport = $2 if defined $2;	391	$uport = $2 if defined $2;
282		392
		393	$hdr{host} = defined $2 ? "$uhost:$2" : "$uhost"
		394	unless exists $hdr{host};
		395
283	$uhost =~ s/^\[(.*)\]$/$1/;	396	$uhost =~ s/^\[(.*)\]$/$1/;
284	$upath .= "?$query" if length $query;	397	$upath .= "?$query" if length $query;
285		398
286	$upath =~ s%^/?%/%;	399	$upath =~ s%^/?%/%;
287		400
288	# cookie processing	401	# cookie processing
289	if (my $jar = $arg{cookie_jar}) {	402	if (my $jar = $arg{cookie_jar}) {
290	%$jar = () if $jar->{version} < 1;	403	%$jar = () if $jar->{version} != 1;
291		404
292	my @cookie;	405	my @cookie;
293		406
294	while (my ($chost, $v) = each %$jar) {	407	while (my ($chost, $v) = each %$jar) {
295	if ($chost =~ /^\./) {	408	if ($chost =~ /^\./) {
…		…
302		415
303	while (my ($cpath, $v) = each %$v) {	416	while (my ($cpath, $v) = each %$v) {
304	next unless $cpath eq substr $upath, 0, length $cpath;	417	next unless $cpath eq substr $upath, 0, length $cpath;
305		418
306	while (my ($k, $v) = each %$v) {	419	while (my ($k, $v) = each %$v) {
307	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;
308	push @cookie, "$k=$v->{value}";	423	push @cookie, "$k=\"$value\"";
309	}	424	}
310	}	425	}
311	}	426	}
312		427
313	$hdr{cookie} = join "; ", @cookie	428	$hdr{cookie} = join "; ", @cookie
314	if @cookie;	429	if @cookie;
315	}	430	}
316		431
317	my ($rhost, $rport, $rpath); # request host, port, path	432	my ($rhost, $rport, $rscheme, $rpath); # request host, port, path
318		433
319	if ($proxy) {	434	if ($proxy) {
320	($rhost, $rport, $scheme) = @$proxy;	435	($rpath, $rhost, $rport, $rscheme) = ($url, @$proxy);
321	$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" && $rscheme eq "https";
322	} else {	442	} else {
323	($rhost, $rport, $rpath) = ($uhost, $uport, $upath);	443	($rhost, $rport, $rscheme, $rpath) = ($uhost, $uport, $uscheme, $upath);
324	$hdr{host} = $uhost;
325	}	444	}
326		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
327	$hdr{"content-length"} = length $arg{body};	450	$hdr{"content-length"} = length $arg{body}
		451	if length $arg{body} \|\| $method ne "GET";
328		452
329	my %state = (connect_guard => 1);	453	my %state = (connect_guard => 1);
330		454
331	_get_slot $uhost, sub {	455	_get_slot $uhost, sub {
332	$state{slot_guard} = shift;	456	$state{slot_guard} = shift;
333		457
334	return unless $state{connect_guard};	458	return unless $state{connect_guard};
335		459
336	$state{connect_guard} = AnyEvent::Socket::tcp_connect $rhost, $rport, sub {	460	$state{connect_guard} = AnyEvent::Socket::tcp_connect $rhost, $rport, sub {
337	$state{fh} = shift	461	$state{fh} = shift
		462	or do {
		463	my $err = "$!";
		464	%state = ();
338	or return $cb->(undef, { Status => 599, Reason => "$!", URL => $url });	465	return $cb->(undef, { Status => 599, Reason => $err, @pseudo });
		466	};
339		467
340	delete $state{connect_guard}; # reduce memory usage, save a tree	468	pop; # free memory, save a tree
		469
		470	return unless delete $state{connect_guard};
341		471
342	# get handle	472	# get handle
343	$state{handle} = new AnyEvent::Handle	473	$state{handle} = new AnyEvent::Handle
344	fh => $state{fh},	474	fh => $state{fh},
345	($scheme eq "https" ? (tls => "connect") : ());	475	timeout => $timeout,
		476	peername => $rhost,
		477	tls_ctx => $arg{tls_ctx};
346		478
347	# limit the number of persistent connections	479	# limit the number of persistent connections
		480	# keepalive not yet supported
348	if ($KA_COUNT{$_[1]} < $MAX_PERSISTENT_PER_HOST) {	481	if ($KA_COUNT{$_[1]} < $MAX_PERSISTENT_PER_HOST) {
349	++$KA_COUNT{$_[1]};	482	++$KA_COUNT{$_[1]};
350	$state{handle}{ka_count_guard} = AnyEvent::Util::guard { --$KA_COUNT{$_[1]} };	483	$state{handle}{ka_count_guard} = AnyEvent::Util::guard {
		484	--$KA_COUNT{$_[1]}
		485	};
351	$hdr{connection} = "keep-alive";	486	$hdr{connection} = "keep-alive";
352	delete $hdr{connection}; # keep-alive not yet supported
353	} else {	487	} else {
354	delete $hdr{connection};	488	delete $hdr{connection};
355	}	489	}
356		490
357	# (re-)configure handle	491	# (re-)configure handle
358	$state{handle}->timeout ($timeout);
359	$state{handle}->on_error (sub {	492	$state{handle}->on_error (sub {
360	my $errno = "$!";
361	%state = ();	493	%state = ();
362	$cb->(undef, { Status => 599, Reason => $errno, URL => $url });	494	$cb->(undef, { Status => 599, Reason => $_[2], @pseudo });
363	});	495	});
364	$state{handle}->on_eof (sub {	496	$state{handle}->on_eof (sub {
365	%state = ();	497	%state = ();
366	$cb->(undef, { Status => 599, Reason => "unexpected end-of-file", URL => $url });	498	$cb->(undef, { Status => 599, Reason => "Unexpected end-of-file", @pseudo });
367	});	499	});
368		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
369	# send request	507	# send request
370	$state{handle}->push_write (	508	$state{handle}->push_write (
371	"$method $rpath HTTP/1.0\015\012"	509	"$method $rpath HTTP/1.0\015\012"
372	. (join "", map "$_: $hdr{$_}\015\012", keys %hdr)	510	. (join "", map "\u$_: $hdr{$_}\015\012", grep defined $hdr{$_}, keys %hdr)
373	. "\015\012"	511	. "\015\012"
374	. (delete $arg{body})	512	. (delete $arg{body})
375	);
376
377	%hdr = (); # reduce memory usage, save a kitten
378
379	# status line
380	$state{handle}->push_read (line => qr/\015?\012/, sub {
381	$_[1] =~ /^HTTP\/([0-9\.]+) \s+ ([0-9]{3}) (?: \s+ ([^\015\012]*) )?/ix
382	or return (%state = (), $cb->(undef, { Status => 599, Reason => "invalid server response ($_[1])", URL => $url }));
383
384	my %hdr = ( # response headers
385	HTTPVersion => "\x00$1",
386	Status => "\x00$2",
387	Reason => "\x00$3",
388	URL => "\x00$url"
389	);	513	);
390		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
391	# headers, could be optimized a bit	531	# headers, could be optimized a bit
392	$state{handle}->unshift_read (line => qr/\015?\012\015?\012/, sub {	532	$state{handle}->unshift_read (line => $qr_nlnl, sub {
393	for ("$_[1]\012") {	533	for ("$_[1]") {
394	# we support spaces in field names, as lotus domino	534	y/\015//d; # weed out any \015, as they show up in the weirdest of places.
395	# creates them (actually spaces around seperators	535
396	# are strictly allowed in http, they are a security issue).	536	# things seen, not parsed:
		537	# p3pP="NON CUR OTPi OUR NOR UNI"
		538
397	$hdr{lc $1} .= "\x00$2"	539	$hdr{lc $1} .= ",$2"
398	while /\G	540	while /\G
399	([^:\000-\037]+):	541	([^:\000-\037]*):
400	[\011\040]*	542	[\011\040]*
401	((?: [^\015\012]+ \| \015?\012[\011\040] )*)	543	((?: [^\012]+ \| \012[\011\040] )*)
402	\015?\012	544	\012
403	/gxc;	545	/gxc;
404		546
405	/\G$/	547	/\G$/
406	or return (%state = (), $cb->(undef, { Status => 599, Reason => "garbled response headers", URL => $url }));	548	or return (%state = (), $cb->(undef, { Status => 599, Reason => "Garbled response headers", @pseudo }));
407	}
408
409	substr $_, 0, 1, ""
410	for values %hdr;
411
412	my $finish = sub {
413	# TODO: use destroy method, when/if available
414	#$state{handle}->destroy;
415	$state{handle}->on_eof (undef);
416	$state{handle}->on_error (undef);
417	%state = ();
418
419	# set-cookie processing
420	if ($arg{cookie_jar} && exists $hdr{"set-cookie"}) {
421	for (split /\x00/, $hdr{"set-cookie"}) {
422	my ($cookie, @arg) = split /;\s*/;
423	my ($name, $value) = split /=/, $cookie, 2;
424	my %kv = (value => $value, map { split /=/, $_, 2 } @arg);
425
426	my $cdom;
427	my $cpath = (delete $kv{path}) \|\| "/";
428
429	if (exists $kv{domain}) {
430	$cdom = delete $kv{domain};
431
432	$cdom =~ s/^\.?/./; # make sure it starts with a "."
433
434	next if $cdom =~ /\.$/;
435
436	# this is not rfc-like and not netscape-like. go figure.
437	my $ndots = $cdom =~ y/.//;
438	next if $ndots < ($cdom =~ /\.[^.][^.]\.[^.][^.]$/ ? 3 : 2);
439	} else {
440	$cdom = $uhost;
441	}
442
443	# store it
444	$arg{cookie_jar}{version} = 1;
445	$arg{cookie_jar}{$cdom}{$cpath}{$name} = \%kv;
446	}
447	}	549	}
448		550
		551	# remove the "," prefix we added to all headers above
		552	substr $_, 0, 1, ""
		553	for values %hdr;
		554
		555	# patch in all pseudo headers
		556	%hdr = (%hdr, @pseudo);
		557
		558	# redirect handling
449	# microsoft and other shitheads don't give a shit for following standards,	559	# microsoft and other shitheads don't give a shit for following standards,
450	# try to support some common forms of broken Location headers.	560	# try to support some common forms of broken Location headers.
451	if ($_[1]{location} !~ /^(?: $ \| [^:\/?\#]+ : )/x) {	561	if ($hdr{location} !~ /^(?: $ \| [^:\/?\#]+ : )/x) {
452	$_[1]{location} =~ s/^\.\/+//;	562	$hdr{location} =~ s/^\.\/+//;
453		563
454	my $url = "$scheme://$uhost:$uport";	564	my $url = "$rscheme://$uhost:$uport";
455		565
456	unless ($_[1]{location} =~ s/^\///) {	566	unless ($hdr{location} =~ s/^\///) {
457	$url .= $upath;	567	$url .= $upath;
458	$url =~ s/\/[^\/]*$//;	568	$url =~ s/\/[^\/]*$//;
459	}	569	}
460		570
461	$_[1]{location} = "$url/$_[1]{location}";	571	$hdr{location} = "$url/$hdr{location}";
462	}	572	}
463		573
464	if ($_[1]{Status} =~ /^30[12]$/ && $recurse && $method ne "POST") {	574	my $redirect;
		575
		576	if ($recurse) {
		577	my $status = $hdr{Status};
		578
		579	if (($status == 301 \|\| $status == 302) && $method ne "POST") {
465	# apparently, mozilla et al. just change POST to GET here	580	# apparently, mozilla et al. just change POST to GET here
466	# more research is needed before we do the same	581	# more research is needed before we do the same
467	http_request ($method, $_[1]{location}, %arg, recurse => $recurse - 1, $cb);	582	$redirect = 1;
468	} elsif ($_[1]{Status} == 303 && $recurse) {	583	} elsif ($status == 303) {
469	# even http/1.1 is unclear on how to mutate the method	584	# even http/1.1 is unclear on how to mutate the method
470	$method = "GET" unless $method eq "HEAD";	585	$method = "GET" unless $method eq "HEAD";
471	http_request ($method => $_[1]{location}, %arg, recurse => $recurse - 1, $cb);	586	$redirect = 1;
472	} elsif ($_[1]{Status} == 307 && $recurse && $method =~ /^(?:GET\|HEAD)$/) {	587	} elsif ($status == 307 && $method =~ /^(?:GET\|HEAD)$/) {
473	http_request ($method => $_[1]{location}, %arg, recurse => $recurse - 1, $cb);	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);
474	} else {	672	} else {
475	$cb->($_[0], $_[1]);	673	# body handling, four different code paths
		674	# for want_body_handle, on_body (2x), normal (2x)
		675	# we might read too much here, but it does not matter yet (no pers. connections)
		676	if (!$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	$_[0]->on_read (sub { });
		721	}
		722	}
476	}	723	}
477	};	724	});
		725	});
		726	};
478		727
479	if ($hdr{Status} =~ /^(?:1..\|204\|304)$/ or $method eq "HEAD") {	728	# now handle proxy-CONNECT method
480	$finish->(undef, \%hdr);	729	if ($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;
481	} else {	741	} else {
482	if (exists $hdr{"content-length"}) {	742	%state = ();
483	$_[0]->unshift_read (chunk => $hdr{"content-length"}, sub {	743	$cb->(undef, { Status => $2, Reason => $3, @pseudo });
484	# could cache persistent connection now
485	if ($hdr{connection} =~ /\bkeep-alive\b/i) {
486	# but we don't, due to misdesigns, this is annoyingly complex
487	};
488
489	$finish->($_[1], \%hdr);
490	});
491	} else {
492	# too bad, need to read until we get an error or EOF,
493	# no way to detect winged data.
494	$_[0]->on_error (sub {
495	$finish->($_[0]{rbuf}, \%hdr);
496	});
497	$_[0]->on_eof (undef);
498	$_[0]->on_read (sub { });
499	}
500	}	744	}
501	});	745	});
		746	} else {
		747	&$handle_actual_request;
502	});	748	}
503	}, sub {	749
504	$timeout	750	}, $arg{on_prepare} \|\| sub { $timeout };
505	};
506	};	751	};
507		752
508	defined wantarray && AnyEvent::Util::guard { %state = () }	753	defined wantarray && AnyEvent::Util::guard { %state = () }
509	}	754	}
510		755
…		…
524	&http_request	769	&http_request
525	}	770	}
526		771
527	=back	772	=back
528		773
		774	=head2 DNS CACHING
		775
		776	AnyEvent::HTTP uses the AnyEvent::Socket::tcp_connect function for
		777	the actual connection, which in turn uses AnyEvent::DNS to resolve
		778	hostnames. The latter is a simple stub resolver and does no caching
		779	on its own. If you want DNS caching, you currently have to provide
		780	your own default resolver (by storing a suitable resolver object in
		781	C<$AnyEvent::DNS::RESOLVER>).
		782
529	=head2 GLOBAL FUNCTIONS AND VARIABLES	783	=head2 GLOBAL FUNCTIONS AND VARIABLES
530		784
531	=over 4	785	=over 4
532		786
533	=item AnyEvent::HTTP::set_proxy "proxy-url"	787	=item AnyEvent::HTTP::set_proxy "proxy-url"
534		788
535	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
536	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>.
537		794
538	=item $AnyEvent::HTTP::MAX_RECURSE	795	=item $AnyEvent::HTTP::MAX_RECURSE
539		796
540	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>).
541		798
542	=item $AnyEvent::HTTP::USERAGENT	799	=item $AnyEvent::HTTP::USERAGENT
543		800
544	The default value for the C<User-Agent> header (the default is	801	The default value for the C<User-Agent> header (the default is
545	C<Mozilla/5.0 (compatible; AnyEvent::HTTP/$VERSION; +http://software.schmorp.de/pkg/AnyEvent)>).	802	C<Mozilla/5.0 (compatible; U; AnyEvent-HTTP/$VERSION; +http://software.schmorp.de/pkg/AnyEvent)>).
546		803
547	=item $AnyEvent::HTTP::MAX_PERSISTENT	804	=item $AnyEvent::HTTP::MAX_PER_HOST
548		805
549	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.
550		809
551	Not implemented currently.	810	The default value for this is C<4>, and it is highly advisable to not
552		811	increase it.
553	=item $AnyEvent::HTTP::PERSISTENT_TIMEOUT
554
555	The maximum time to cache a persistent connection, in seconds (default: 2).
556
557	Not implemented currently.
558		812
559	=item $AnyEvent::HTTP::ACTIVE	813	=item $AnyEvent::HTTP::ACTIVE
560		814
561	The number of active connections. This is not the number of currently	815	The number of active connections. This is not the number of currently
562	running requests, but the number of currently open and non-idle TCP	816	running requests, but the number of currently open and non-idle TCP
…		…
565	=back	819	=back
566		820
567	=cut	821	=cut
568		822
569	sub set_proxy($) {	823	sub set_proxy($) {
		824	if (length $_[0]) {
570	$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	}
571	}	831	}
572		832
573	# initialise proxy from environment	833	# initialise proxy from environment
		834	eval {
574	set_proxy $ENV{http_proxy};	835	set_proxy $ENV{http_proxy};
		836	};
575		837
576	=head1 SEE ALSO	838	=head1 SEE ALSO
577		839
578	L<AnyEvent>.	840	L<AnyEvent>.
579		841
580	=head1 AUTHOR	842	=head1 AUTHOR
581		843
582	Marc Lehmann <schmorp@schmorp.de>	844	Marc Lehmann <schmorp@schmorp.de>
583	http://home.schmorp.de/	845	http://home.schmorp.de/
584		846
		847	With many thanks to Дмитрий Шалашов, who provided countless
		848	testcases and bugreports.
		849
585	=cut	850	=cut
586		851
587	1	852	1
588		853

Diff Legend

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

Comparing AnyEvent-HTTP/HTTP.pm (file contents): Revision 1.30 by root, Thu Oct 23 02:46:20 2008 UTC vs. Revision 1.55 by root, Wed Jun 16 19:17:30 2010 UTC

Diff Legend

Comparing AnyEvent-HTTP/HTTP.pm (file contents):
Revision 1.30 by root, Thu Oct 23 02:46:20 2008 UTC vs.
Revision 1.55 by root, Wed Jun 16 19:17:30 2010 UTC