… | |
… | |
41 | use strict; |
41 | use strict; |
42 | no warnings; |
42 | no warnings; |
43 | |
43 | |
44 | use Errno (); |
44 | use Errno (); |
45 | |
45 | |
46 | use AnyEvent 4.452 (); |
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 = '1.12'; |
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 $USERAGENT = "Mozilla/5.0 (compatible; U; AnyEvent-HTTP/$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 $ACTIVE = 0; |
69 | |
69 | |
… | |
… | |
99 | The callback will be called with the response body data as first argument |
99 | The callback will be called with the response body data as first argument |
100 | (or C<undef> if an error occured), and a hash-ref with response headers as |
100 | (or C<undef> if an error occured), and a hash-ref with response headers as |
101 | second argument. |
101 | second argument. |
102 | |
102 | |
103 | All the headers in that hash are lowercased. In addition to the response |
103 | All the headers in that hash are lowercased. In addition to the response |
104 | headers, the "pseudo-headers" C<HTTPVersion>, C<Status> and C<Reason> |
104 | headers, the "pseudo-headers" (uppercase to avoid clashing with possible |
|
|
105 | response headers) C<HTTPVersion>, C<Status> and C<Reason> contain the |
105 | contain the three parts of the HTTP Status-Line of the same name. The |
106 | three parts of the HTTP Status-Line of the same name. |
|
|
107 | |
106 | pseudo-header C<URL> contains the original URL (which can differ from the |
108 | The pseudo-header C<URL> contains the actual URL (which can differ from |
107 | requested URL when following redirects). |
109 | the requested URL when following redirects - for example, you might get |
|
|
110 | an error that your URL scheme is not supported even though your URL is a |
|
|
111 | valid http URL because it redirected to an ftp URL, in which case you can |
|
|
112 | look at the URL pseudo header). |
|
|
113 | |
|
|
114 | The pseudo-header C<Redirect> only exists when the request was a result |
|
|
115 | of an internal redirect. In that case it is an array reference with |
|
|
116 | the C<($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. |
108 | |
120 | |
109 | If the server sends a header multiple times, then their contents will be |
121 | If the server sends a header multiple times, then their contents will be |
110 | joined together with a comma (C<,>), as per the HTTP spec. |
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, |
… | |
… | |
138 | |
150 | |
139 | =item headers => hashref |
151 | =item headers => hashref |
140 | |
152 | |
141 | 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 |
142 | 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 |
143 | 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). |
144 | |
158 | |
145 | =item timeout => $seconds |
159 | =item timeout => $seconds |
146 | |
160 | |
147 | The time-out to use for various stages - each connect attempt will reset |
161 | The time-out to use for various stages - each connect attempt will reset |
148 | the timeout, as will read or write activity. Default timeout is 5 minutes. |
162 | the timeout, as will read or write activity, i.e. this is not an overall |
|
|
163 | timeout. |
|
|
164 | |
|
|
165 | Default timeout is 5 minutes. |
149 | |
166 | |
150 | =item proxy => [$host, $port[, $scheme]] or undef |
167 | =item proxy => [$host, $port[, $scheme]] or undef |
151 | |
168 | |
152 | Use the given http proxy for all requests. If not specified, then the |
169 | Use the given http proxy for all requests. If not specified, then the |
153 | default proxy (as specified by C<$ENV{http_proxy}>) is used. |
170 | default proxy (as specified by C<$ENV{http_proxy}>) is used. |
154 | |
171 | |
155 | C<$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 |
156 | HTTPS. |
173 | HTTPS. |
157 | |
174 | |
158 | =item body => $string |
175 | =item body => $string |
159 | |
176 | |
160 | 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 |
… | |
… | |
186 | verification) TLS context. |
203 | verification) TLS context. |
187 | |
204 | |
188 | The default for this option is C<low>, which could be interpreted as "give |
205 | The default for this option is C<low>, which could be interpreted as "give |
189 | me the page, no matter what". |
206 | me the page, no matter what". |
190 | |
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 | |
191 | =item on_header => $callback->($headers) |
217 | =item on_header => $callback->($headers) |
192 | |
218 | |
193 | When specified, this callback will be called with the header hash as soon |
219 | When specified, this callback will be called with the header hash as soon |
194 | as headers have been successfully received from the remote server (not on |
220 | as headers have been successfully received from the remote server (not on |
195 | locally-generated errors). |
221 | locally-generated errors). |
… | |
… | |
221 | This callback is useful when the data is too large to be held in memory |
247 | This callback is useful when the data is too large to be held in memory |
222 | (so the callback writes it to a file) or when only some information should |
248 | (so the callback writes it to a file) or when only some information should |
223 | be extracted, or when the body should be processed incrementally. |
249 | be extracted, or when the body should be processed incrementally. |
224 | |
250 | |
225 | It is usually preferred over doing your own body handling via |
251 | It is usually preferred over doing your own body handling via |
226 | C<want_body_handle>. |
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. |
227 | |
256 | |
228 | =item want_body_handle => $enable |
257 | =item want_body_handle => $enable |
229 | |
258 | |
230 | When enabled (default is disabled), the behaviour of AnyEvent::HTTP |
259 | When enabled (default is disabled), the behaviour of AnyEvent::HTTP |
231 | changes considerably: after parsing the headers, and instead of |
260 | changes considerably: after parsing the headers, and instead of |
… | |
… | |
243 | This is useful with some push-type services, where, after the initial |
272 | This is useful with some push-type services, where, after the initial |
244 | headers, an interactive protocol is used (typical example would be the |
273 | headers, an interactive protocol is used (typical example would be the |
245 | push-style twitter API which starts a JSON/XML stream). |
274 | push-style twitter API which starts a JSON/XML stream). |
246 | |
275 | |
247 | If you think you need this, first have a look at C<on_body>, to see if |
276 | If you think you need this, first have a look at C<on_body>, to see if |
248 | that doesn'T solve your problem in a better way. |
277 | that doesn't solve your problem in a better way. |
249 | |
278 | |
250 | =back |
279 | =back |
251 | |
280 | |
252 | 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/ |
253 | |
282 | |
… | |
… | |
309 | push @{ $CO_SLOT{$_[0]}[1] }, $_[1]; |
338 | push @{ $CO_SLOT{$_[0]}[1] }, $_[1]; |
310 | |
339 | |
311 | _slot_schedule $_[0]; |
340 | _slot_schedule $_[0]; |
312 | } |
341 | } |
313 | |
342 | |
314 | our $qr_nl = qr<\015?\012>; |
343 | our $qr_nl = qr{\015?\012}; |
315 | our $qr_nlnl = qr<\015?\012\015?\012>; |
344 | our $qr_nlnl = qr{(?<![^\012])\015?\012}; |
316 | |
345 | |
317 | our $TLS_CTX_LOW = { cache => 1, sslv2 => 1 }; |
346 | our $TLS_CTX_LOW = { cache => 1, sslv2 => 1 }; |
318 | our $TLS_CTX_HIGH = { cache => 1, verify => 1, verify_peername => "https" }; |
347 | our $TLS_CTX_HIGH = { cache => 1, verify => 1, verify_peername => "https" }; |
319 | |
348 | |
320 | sub http_request($$@) { |
349 | sub http_request($$@) { |
… | |
… | |
332 | while (my ($k, $v) = each %$hdr) { |
361 | while (my ($k, $v) = each %$hdr) { |
333 | $hdr{lc $k} = $v; |
362 | $hdr{lc $k} = $v; |
334 | } |
363 | } |
335 | } |
364 | } |
336 | |
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 | |
337 | my $recurse = exists $arg{recurse} ? delete $arg{recurse} : $MAX_RECURSE; |
370 | my $recurse = exists $arg{recurse} ? delete $arg{recurse} : $MAX_RECURSE; |
338 | |
371 | |
339 | return $cb->(undef, { Status => 599, Reason => "Too many redirections", URL => $url }) |
372 | return $cb->(undef, { Status => 599, Reason => "Too many redirections", @pseudo }) |
340 | if $recurse < 0; |
373 | if $recurse < 0; |
341 | |
374 | |
342 | my $proxy = $arg{proxy} || $PROXY; |
375 | my $proxy = $arg{proxy} || $PROXY; |
343 | my $timeout = $arg{timeout} || $TIMEOUT; |
376 | my $timeout = $arg{timeout} || $TIMEOUT; |
344 | |
377 | |
… | |
… | |
347 | |
380 | |
348 | $uscheme = lc $uscheme; |
381 | $uscheme = lc $uscheme; |
349 | |
382 | |
350 | my $uport = $uscheme eq "http" ? 80 |
383 | my $uport = $uscheme eq "http" ? 80 |
351 | : $uscheme eq "https" ? 443 |
384 | : $uscheme eq "https" ? 443 |
352 | : 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 }); |
353 | |
386 | |
354 | $uauthority =~ /^(?: .*\@ )? ([^\@:]+) (?: : (\d+) )?$/x |
387 | $uauthority =~ /^(?: .*\@ )? ([^\@:]+) (?: : (\d+) )?$/x |
355 | or return $cb->(undef, { Status => 599, Reason => "Unparsable URL", URL => $url }); |
388 | or return $cb->(undef, { Status => 599, Reason => "Unparsable URL", @pseudo }); |
356 | |
389 | |
357 | my $uhost = $1; |
390 | my $uhost = $1; |
358 | $uport = $2 if defined $2; |
391 | $uport = $2 if defined $2; |
|
|
392 | |
|
|
393 | $hdr{host} = defined $2 ? "$uhost:$2" : "$uhost" |
|
|
394 | unless exists $hdr{host}; |
359 | |
395 | |
360 | $uhost =~ s/^\[(.*)\]$/$1/; |
396 | $uhost =~ s/^\[(.*)\]$/$1/; |
361 | $upath .= "?$query" if length $query; |
397 | $upath .= "?$query" if length $query; |
362 | |
398 | |
363 | $upath =~ s%^/?%/%; |
399 | $upath =~ s%^/?%/%; |
… | |
… | |
396 | my ($rhost, $rport, $rscheme, $rpath); # request host, port, path |
432 | my ($rhost, $rport, $rscheme, $rpath); # request host, port, path |
397 | |
433 | |
398 | if ($proxy) { |
434 | if ($proxy) { |
399 | ($rpath, $rhost, $rport, $rscheme) = ($url, @$proxy); |
435 | ($rpath, $rhost, $rport, $rscheme) = ($url, @$proxy); |
400 | |
436 | |
|
|
437 | $rscheme = "http" unless defined $rscheme; |
|
|
438 | |
401 | # don't support https requests over https-proxy transport, |
439 | # don't support https requests over https-proxy transport, |
402 | # can't be done with tls as spec'ed, unless you double-encrypt. |
440 | # can't be done with tls as spec'ed, unless you double-encrypt. |
403 | $rscheme = "http" if $uscheme eq "https" && $rscheme eq "https"; |
441 | $rscheme = "http" if $uscheme eq "https" && $rscheme eq "https"; |
404 | } else { |
442 | } else { |
405 | ($rhost, $rport, $rscheme, $rpath) = ($uhost, $uport, $uscheme, $upath); |
443 | ($rhost, $rport, $rscheme, $rpath) = ($uhost, $uport, $uscheme, $upath); |
406 | } |
444 | } |
407 | |
445 | |
408 | $hdr{"user-agent"} ||= $USERAGENT; |
446 | # leave out fragment and query string, just a heuristic |
409 | $hdr{referer} ||= "$uscheme://$uauthority$upath"; # 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"}; |
410 | |
449 | |
411 | $hdr{host} = "$uhost:$uport"; |
|
|
412 | $hdr{"content-length"} = length $arg{body}; |
450 | $hdr{"content-length"} = length $arg{body} |
|
|
451 | if length $arg{body} || $method ne "GET"; |
413 | |
452 | |
414 | my %state = (connect_guard => 1); |
453 | my %state = (connect_guard => 1); |
415 | |
454 | |
416 | _get_slot $uhost, sub { |
455 | _get_slot $uhost, sub { |
417 | $state{slot_guard} = shift; |
456 | $state{slot_guard} = shift; |
418 | |
457 | |
419 | return unless $state{connect_guard}; |
458 | return unless $state{connect_guard}; |
420 | |
459 | |
421 | $state{connect_guard} = AnyEvent::Socket::tcp_connect $rhost, $rport, sub { |
460 | $state{connect_guard} = AnyEvent::Socket::tcp_connect $rhost, $rport, sub { |
422 | $state{fh} = shift |
461 | $state{fh} = shift |
|
|
462 | or do { |
|
|
463 | my $err = "$!"; |
|
|
464 | %state = (); |
423 | or return (%state = (), $cb->(undef, { Status => 599, Reason => "$!", URL => $url })); |
465 | return $cb->(undef, { Status => 599, Reason => $err, @pseudo }); |
|
|
466 | }; |
|
|
467 | |
424 | pop; # free memory, save a tree |
468 | pop; # free memory, save a tree |
425 | |
469 | |
426 | return unless delete $state{connect_guard}; |
470 | return unless delete $state{connect_guard}; |
427 | |
471 | |
428 | # get handle |
472 | # get handle |
… | |
… | |
445 | } |
489 | } |
446 | |
490 | |
447 | # (re-)configure handle |
491 | # (re-)configure handle |
448 | $state{handle}->on_error (sub { |
492 | $state{handle}->on_error (sub { |
449 | %state = (); |
493 | %state = (); |
450 | $cb->(undef, { Status => 599, Reason => $_[2], URL => $url }); |
494 | $cb->(undef, { Status => 599, Reason => $_[2], @pseudo }); |
451 | }); |
495 | }); |
452 | $state{handle}->on_eof (sub { |
496 | $state{handle}->on_eof (sub { |
453 | %state = (); |
497 | %state = (); |
454 | $cb->(undef, { Status => 599, Reason => "Unexpected end-of-file", URL => $url }); |
498 | $cb->(undef, { Status => 599, Reason => "Unexpected end-of-file", @pseudo }); |
455 | }); |
499 | }); |
456 | |
500 | |
457 | $state{handle}->starttls ("connect") if $rscheme eq "https"; |
501 | $state{handle}->starttls ("connect") if $rscheme eq "https"; |
458 | |
502 | |
459 | # handle actual, non-tunneled, request |
503 | # handle actual, non-tunneled, request |
… | |
… | |
461 | $state{handle}->starttls ("connect") if $uscheme eq "https" && !exists $state{handle}{tls}; |
505 | $state{handle}->starttls ("connect") if $uscheme eq "https" && !exists $state{handle}{tls}; |
462 | |
506 | |
463 | # send request |
507 | # send request |
464 | $state{handle}->push_write ( |
508 | $state{handle}->push_write ( |
465 | "$method $rpath HTTP/1.0\015\012" |
509 | "$method $rpath HTTP/1.0\015\012" |
466 | . (join "", map "\u$_: $hdr{$_}\015\012", keys %hdr) |
510 | . (join "", map "\u$_: $hdr{$_}\015\012", grep defined $hdr{$_}, keys %hdr) |
467 | . "\015\012" |
511 | . "\015\012" |
468 | . (delete $arg{body}) |
512 | . (delete $arg{body}) |
469 | ); |
513 | ); |
470 | |
514 | |
|
|
515 | # return if error occured during push_write() |
|
|
516 | return unless %state; |
|
|
517 | |
471 | %hdr = (); # reduce memory usage, save a kitten |
518 | %hdr = (); # reduce memory usage, save a kitten, also make it possible to re-use |
472 | |
519 | |
473 | # status line |
520 | # status line |
474 | $state{handle}->push_read (line => $qr_nl, sub { |
521 | $state{handle}->push_read (line => $qr_nl, sub { |
475 | $_[1] =~ /^HTTP\/([0-9\.]+) \s+ ([0-9]{3}) (?: \s+ ([^\015\012]*) )?/ix |
522 | $_[1] =~ /^HTTP\/([0-9\.]+) \s+ ([0-9]{3}) (?: \s+ ([^\015\012]*) )?/ix |
476 | or return (%state = (), $cb->(undef, { Status => 599, Reason => "Invalid server response ($_[1])", URL => $url })); |
523 | or return (%state = (), $cb->(undef, { Status => 599, Reason => "Invalid server response ($_[1])", @pseudo })); |
477 | |
524 | |
478 | my %hdr = ( # response headers |
525 | push @pseudo, |
479 | HTTPVersion => ",$1", |
526 | HTTPVersion => $1, |
480 | Status => ",$2", |
527 | Status => $2, |
481 | Reason => ",$3", |
528 | Reason => $3, |
482 | URL => ",$url" |
|
|
483 | ); |
529 | ; |
484 | |
530 | |
485 | # headers, could be optimized a bit |
531 | # headers, could be optimized a bit |
486 | $state{handle}->unshift_read (line => $qr_nlnl, sub { |
532 | $state{handle}->unshift_read (line => $qr_nlnl, sub { |
487 | for ("$_[1]\012") { |
533 | for ("$_[1]") { |
488 | y/\015//d; # weed out any \015, as they show up in the weirdest of places. |
534 | y/\015//d; # weed out any \015, as they show up in the weirdest of places. |
489 | |
535 | |
490 | # things seen, not parsed: |
536 | # things seen, not parsed: |
491 | # p3pP="NON CUR OTPi OUR NOR UNI" |
537 | # p3pP="NON CUR OTPi OUR NOR UNI" |
492 | |
538 | |
493 | $hdr{lc $1} .= ",$2" |
539 | $hdr{lc $1} .= ",$2" |
494 | while /\G |
540 | while /\G |
495 | ([^:\000-\037]+): |
541 | ([^:\000-\037]*): |
496 | [\011\040]* |
542 | [\011\040]* |
497 | ((?: [^\012]+ | \012[\011\040] )*) |
543 | ((?: [^\012]+ | \012[\011\040] )*) |
498 | \012 |
544 | \012 |
499 | /gxc; |
545 | /gxc; |
500 | |
546 | |
501 | /\G$/ |
547 | /\G$/ |
502 | 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 })); |
503 | } |
549 | } |
504 | |
550 | |
|
|
551 | # remove the "," prefix we added to all headers above |
505 | substr $_, 0, 1, "" |
552 | substr $_, 0, 1, "" |
506 | for values %hdr; |
553 | for values %hdr; |
|
|
554 | |
|
|
555 | # patch in all pseudo headers |
|
|
556 | %hdr = (%hdr, @pseudo); |
507 | |
557 | |
508 | # redirect handling |
558 | # redirect handling |
509 | # microsoft and other shitheads don't give a shit for following standards, |
559 | # microsoft and other shitheads don't give a shit for following standards, |
510 | # try to support some common forms of broken Location headers. |
560 | # try to support some common forms of broken Location headers. |
511 | if ($hdr{location} !~ /^(?: $ | [^:\/?\#]+ : )/x) { |
561 | if ($hdr{location} !~ /^(?: $ | [^:\/?\#]+ : )/x) { |
… | |
… | |
522 | } |
572 | } |
523 | |
573 | |
524 | my $redirect; |
574 | my $redirect; |
525 | |
575 | |
526 | if ($recurse) { |
576 | if ($recurse) { |
|
|
577 | my $status = $hdr{Status}; |
|
|
578 | |
527 | if ($hdr{Status} =~ /^30[12]$/ && $method ne "POST") { |
579 | if (($status == 301 || $status == 302) && $method ne "POST") { |
528 | # apparently, mozilla et al. just change POST to GET here |
580 | # apparently, mozilla et al. just change POST to GET here |
529 | # more research is needed before we do the same |
581 | # more research is needed before we do the same |
530 | $redirect = 1; |
582 | $redirect = 1; |
531 | } elsif ($hdr{Status} == 303) { |
583 | } elsif ($status == 303) { |
532 | # 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 |
533 | $method = "GET" unless $method eq "HEAD"; |
585 | $method = "GET" unless $method eq "HEAD"; |
534 | $redirect = 1; |
586 | $redirect = 1; |
535 | } elsif ($hdr{Status} == 307 && $method =~ /^(?:GET|HEAD)$/) { |
587 | } elsif ($status == 307 && $method =~ /^(?:GET|HEAD)$/) { |
536 | $redirect = 1; |
588 | $redirect = 1; |
537 | } |
589 | } |
538 | } |
590 | } |
539 | |
591 | |
540 | my $finish = sub { |
592 | my $finish = sub { |
… | |
… | |
589 | |
641 | |
590 | redo if /\G\s*,/gc; |
642 | redo if /\G\s*,/gc; |
591 | } |
643 | } |
592 | } |
644 | } |
593 | |
645 | |
594 | if ($redirect) { |
646 | if ($redirect && exists $hdr{location}) { |
595 | # we ignore any errors, as it is very common to receive |
647 | # we ignore any errors, as it is very common to receive |
596 | # Content-Length != 0 but no actual body |
648 | # Content-Length != 0 but no actual body |
597 | # we also access %hdr, as $_[1] might be an erro |
649 | # we also access %hdr, as $_[1] might be an erro |
598 | http_request ($method => $hdr{location}, %arg, recurse => $recurse - 1, $cb); |
650 | http_request ( |
|
|
651 | $method => $hdr{location}, |
|
|
652 | %arg, |
|
|
653 | recurse => $recurse - 1, |
|
|
654 | Redirect => \@_, |
|
|
655 | $cb); |
599 | } else { |
656 | } else { |
600 | $cb->($_[0], $_[1]); |
657 | $cb->($_[0], $_[1]); |
601 | } |
658 | } |
602 | }; |
659 | }; |
603 | |
660 | |
604 | my $len = $hdr{"content-length"}; |
661 | my $len = $hdr{"content-length"}; |
605 | |
662 | |
606 | if (!$redirect && $arg{on_header} && !$arg{on_header}(\%hdr)) { |
663 | if (!$redirect && $arg{on_header} && !$arg{on_header}(\%hdr)) { |
607 | $finish->(undef, { Status => 598, Reason => "Request cancelled by on_header", URL => $url }); |
664 | $finish->(undef, { Status => 598, Reason => "Request cancelled by on_header", @pseudo }); |
608 | } elsif ( |
665 | } elsif ( |
609 | $hdr{Status} =~ /^(?:1..|[23]04)$/ |
666 | $hdr{Status} =~ /^(?:1..|[23]04)$/ |
610 | or $method eq "HEAD" |
667 | or $method eq "HEAD" |
611 | or (defined $len && !$len) |
668 | or (defined $len && !$len) |
612 | ) { |
669 | ) { |
… | |
… | |
622 | $_[0]->on_read (undef); |
679 | $_[0]->on_read (undef); |
623 | |
680 | |
624 | $finish->(delete $state{handle}, \%hdr); |
681 | $finish->(delete $state{handle}, \%hdr); |
625 | |
682 | |
626 | } elsif ($arg{on_body}) { |
683 | } elsif ($arg{on_body}) { |
627 | $_[0]->on_error (sub { $finish->(undef, { Status => 599, Reason => $_[2], URL => $url }) }); |
684 | $_[0]->on_error (sub { $finish->(undef, { Status => 599, Reason => $_[2], @pseudo }) }); |
628 | if ($len) { |
685 | if ($len) { |
629 | $_[0]->on_eof (undef); |
686 | $_[0]->on_eof (undef); |
630 | $_[0]->on_read (sub { |
687 | $_[0]->on_read (sub { |
631 | $len -= length $_[0]{rbuf}; |
688 | $len -= length $_[0]{rbuf}; |
632 | |
689 | |
633 | $arg{on_body}(delete $_[0]{rbuf}, \%hdr) |
690 | $arg{on_body}(delete $_[0]{rbuf}, \%hdr) |
634 | or $finish->(undef, { Status => 598, Reason => "Request cancelled by on_body", URL => $url }); |
691 | or $finish->(undef, { Status => 598, Reason => "Request cancelled by on_body", @pseudo }); |
635 | |
692 | |
636 | $len > 0 |
693 | $len > 0 |
637 | or $finish->("", \%hdr); |
694 | or $finish->("", \%hdr); |
638 | }); |
695 | }); |
639 | } else { |
696 | } else { |
640 | $_[0]->on_eof (sub { |
697 | $_[0]->on_eof (sub { |
641 | $finish->("", \%hdr); |
698 | $finish->("", \%hdr); |
642 | }); |
699 | }); |
643 | $_[0]->on_read (sub { |
700 | $_[0]->on_read (sub { |
644 | $arg{on_body}(delete $_[0]{rbuf}, \%hdr) |
701 | $arg{on_body}(delete $_[0]{rbuf}, \%hdr) |
645 | or $finish->(undef, { Status => 598, Reason => "Request cancelled by on_body", URL => $url }); |
702 | or $finish->(undef, { Status => 598, Reason => "Request cancelled by on_body", @pseudo }); |
646 | }); |
703 | }); |
647 | } |
704 | } |
648 | } else { |
705 | } else { |
649 | $_[0]->on_eof (undef); |
706 | $_[0]->on_eof (undef); |
650 | |
707 | |
651 | if ($len) { |
708 | if ($len) { |
652 | $_[0]->on_error (sub { $finish->(undef, { Status => 599, Reason => $_[2], URL => $url }) }); |
709 | $_[0]->on_error (sub { $finish->(undef, { Status => 599, Reason => $_[2], @pseudo }) }); |
653 | $_[0]->on_read (sub { |
710 | $_[0]->on_read (sub { |
654 | $finish->((substr delete $_[0]{rbuf}, 0, $len, ""), \%hdr) |
711 | $finish->((substr delete $_[0]{rbuf}, 0, $len, ""), \%hdr) |
655 | if $len <= length $_[0]{rbuf}; |
712 | if $len <= length $_[0]{rbuf}; |
656 | }); |
713 | }); |
657 | } else { |
714 | } else { |
658 | $_[0]->on_error (sub { |
715 | $_[0]->on_error (sub { |
659 | $! == Errno::EPIPE |
716 | $! == Errno::EPIPE || !$! |
660 | ? $finish->(delete $_[0]{rbuf}, \%hdr) |
717 | ? $finish->(delete $_[0]{rbuf}, \%hdr) |
661 | : $finish->(undef, { Status => 599, Reason => $_[2], URL => $url }); |
718 | : $finish->(undef, { Status => 599, Reason => $_[2], @pseudo }); |
662 | }); |
719 | }); |
663 | $_[0]->on_read (sub { }); |
720 | $_[0]->on_read (sub { }); |
664 | } |
721 | } |
665 | } |
722 | } |
666 | } |
723 | } |
… | |
… | |
674 | |
731 | |
675 | # maybe re-use $uauthority with patched port? |
732 | # maybe re-use $uauthority with patched port? |
676 | $state{handle}->push_write ("CONNECT $uhost:$uport HTTP/1.0\015\012Host: $uhost\015\012\015\012"); |
733 | $state{handle}->push_write ("CONNECT $uhost:$uport HTTP/1.0\015\012Host: $uhost\015\012\015\012"); |
677 | $state{handle}->push_read (line => $qr_nlnl, sub { |
734 | $state{handle}->push_read (line => $qr_nlnl, sub { |
678 | $_[1] =~ /^HTTP\/([0-9\.]+) \s+ ([0-9]{3}) (?: \s+ ([^\015\012]*) )?/ix |
735 | $_[1] =~ /^HTTP\/([0-9\.]+) \s+ ([0-9]{3}) (?: \s+ ([^\015\012]*) )?/ix |
679 | or return (%state = (), $cb->(undef, { Status => 599, Reason => "Invalid proxy connect response ($_[1])", URL => $url })); |
736 | or return (%state = (), $cb->(undef, { Status => 599, Reason => "Invalid proxy connect response ($_[1])", @pseudo })); |
680 | |
737 | |
681 | if ($2 == 200) { |
738 | if ($2 == 200) { |
682 | $rpath = $upath; |
739 | $rpath = $upath; |
683 | &$handle_actual_request; |
740 | &$handle_actual_request; |
684 | } else { |
741 | } else { |
685 | %state = (); |
742 | %state = (); |
686 | $cb->(undef, { Status => $2, Reason => $3, URL => $url }); |
743 | $cb->(undef, { Status => $2, Reason => $3, @pseudo }); |
687 | } |
744 | } |
688 | }); |
745 | }); |
689 | } else { |
746 | } else { |
690 | &$handle_actual_request; |
747 | &$handle_actual_request; |
691 | } |
748 | } |
692 | |
749 | |
693 | }, sub { |
750 | }, $arg{on_prepare} || sub { $timeout }; |
694 | $timeout |
|
|
695 | }; |
|
|
696 | }; |
751 | }; |
697 | |
752 | |
698 | defined wantarray && AnyEvent::Util::guard { %state = () } |
753 | defined wantarray && AnyEvent::Util::guard { %state = () } |
699 | } |
754 | } |
700 | |
755 | |
… | |
… | |
714 | &http_request |
769 | &http_request |
715 | } |
770 | } |
716 | |
771 | |
717 | =back |
772 | =back |
718 | |
773 | |
|
|
774 | =head2 DNS CACHING |
|
|
775 | |
|
|
776 | AnyEvent::HTTP uses the AnyEvent::Socket::tcp_connect function for |
|
|
777 | the actual connection, which in turn uses AnyEvent::DNS to resolve |
|
|
778 | hostnames. The latter is a simple stub resolver and does no caching |
|
|
779 | on its own. If you want DNS caching, you currently have to provide |
|
|
780 | your own default resolver (by storing a suitable resolver object in |
|
|
781 | C<$AnyEvent::DNS::RESOLVER>). |
|
|
782 | |
719 | =head2 GLOBAL FUNCTIONS AND VARIABLES |
783 | =head2 GLOBAL FUNCTIONS AND VARIABLES |
720 | |
784 | |
721 | =over 4 |
785 | =over 4 |
722 | |
786 | |
723 | =item AnyEvent::HTTP::set_proxy "proxy-url" |
787 | =item AnyEvent::HTTP::set_proxy "proxy-url" |
724 | |
788 | |
725 | 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 |
726 | 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>. |
727 | |
794 | |
728 | =item $AnyEvent::HTTP::MAX_RECURSE |
795 | =item $AnyEvent::HTTP::MAX_RECURSE |
729 | |
796 | |
730 | 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>). |
731 | |
798 | |
732 | =item $AnyEvent::HTTP::USERAGENT |
799 | =item $AnyEvent::HTTP::USERAGENT |
733 | |
800 | |
734 | The default value for the C<User-Agent> header (the default is |
801 | The default value for the C<User-Agent> header (the default is |
735 | C<Mozilla/5.0 (compatible; U; AnyEvent-HTTP/$VERSION; +http://software.schmorp.de/pkg/AnyEvent)>). |
802 | C<Mozilla/5.0 (compatible; U; AnyEvent-HTTP/$VERSION; +http://software.schmorp.de/pkg/AnyEvent)>). |
736 | |
803 | |
737 | =item $AnyEvent::HTTP::MAX_PERSISTENT |
804 | =item $AnyEvent::HTTP::MAX_PER_HOST |
738 | |
805 | |
739 | 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. |
740 | |
809 | |
741 | Not implemented currently. |
810 | The default value for this is C<4>, and it is highly advisable to not |
742 | |
811 | increase it. |
743 | =item $AnyEvent::HTTP::PERSISTENT_TIMEOUT |
|
|
744 | |
|
|
745 | The maximum time to cache a persistent connection, in seconds (default: 2). |
|
|
746 | |
|
|
747 | Not implemented currently. |
|
|
748 | |
812 | |
749 | =item $AnyEvent::HTTP::ACTIVE |
813 | =item $AnyEvent::HTTP::ACTIVE |
750 | |
814 | |
751 | The number of active connections. This is not the number of currently |
815 | The number of active connections. This is not the number of currently |
752 | running requests, but the number of currently open and non-idle TCP |
816 | running requests, but the number of currently open and non-idle TCP |
… | |
… | |
755 | =back |
819 | =back |
756 | |
820 | |
757 | =cut |
821 | =cut |
758 | |
822 | |
759 | sub set_proxy($) { |
823 | sub set_proxy($) { |
|
|
824 | if (length $_[0]) { |
760 | $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 | } |
761 | } |
831 | } |
762 | |
832 | |
763 | # initialise proxy from environment |
833 | # initialise proxy from environment |
|
|
834 | eval { |
764 | set_proxy $ENV{http_proxy}; |
835 | set_proxy $ENV{http_proxy}; |
|
|
836 | }; |
765 | |
837 | |
766 | =head1 SEE ALSO |
838 | =head1 SEE ALSO |
767 | |
839 | |
768 | L<AnyEvent>. |
840 | L<AnyEvent>. |
769 | |
841 | |