… | |
… | |
15 | This module is an L<AnyEvent> user, you need to make sure that you use and |
15 | This module is an L<AnyEvent> user, you need to make sure that you use and |
16 | run a supported event loop. |
16 | run a supported event loop. |
17 | |
17 | |
18 | This module implements a simple, stateless and non-blocking HTTP |
18 | This module implements a simple, stateless and non-blocking HTTP |
19 | client. It supports GET, POST and other request methods, cookies and more, |
19 | client. It supports GET, POST and other request methods, cookies and more, |
20 | all on a very low level. It can follow redirects supports proxies and |
20 | all on a very low level. It can follow redirects, supports proxies, and |
21 | automatically limits the number of connections to the values specified in |
21 | automatically limits the number of connections to the values specified in |
22 | the RFC. |
22 | the RFC. |
23 | |
23 | |
24 | It should generally be a "good client" that is enough for most HTTP |
24 | It should generally be a "good client" that is enough for most HTTP |
25 | tasks. Simple tasks should be simple, but complex tasks should still be |
25 | tasks. Simple tasks should be simple, but complex tasks should still be |
… | |
… | |
46 | use AnyEvent::Util (); |
46 | use AnyEvent::Util (); |
47 | use AnyEvent::Handle (); |
47 | use AnyEvent::Handle (); |
48 | |
48 | |
49 | use base Exporter::; |
49 | use base Exporter::; |
50 | |
50 | |
51 | our $VERSION = '1.5'; |
51 | our $VERSION = '2.13'; |
52 | |
52 | |
53 | our @EXPORT = qw(http_get http_post http_head http_request); |
53 | our @EXPORT = qw(http_get http_post http_head http_request); |
54 | |
54 | |
55 | our $USERAGENT = "Mozilla/5.0 (compatible; U; AnyEvent-HTTP/$VERSION; +http://software.schmorp.de/pkg/AnyEvent)"; |
55 | our $USERAGENT = "Mozilla/5.0 (compatible; U; AnyEvent-HTTP/$VERSION; +http://software.schmorp.de/pkg/AnyEvent)"; |
56 | our $MAX_RECURSE = 10; |
56 | our $MAX_RECURSE = 10; |
57 | our $MAX_PERSISTENT = 8; |
|
|
58 | our $PERSISTENT_TIMEOUT = 2; |
57 | our $PERSISTENT_TIMEOUT = 3; |
59 | our $TIMEOUT = 300; |
58 | our $TIMEOUT = 300; |
60 | |
59 | our $MAX_PER_HOST = 4; # changing this is evil |
61 | # changing these is evil |
|
|
62 | our $MAX_PERSISTENT_PER_HOST = 0; |
|
|
63 | our $MAX_PER_HOST = 4; |
|
|
64 | |
60 | |
65 | our $PROXY; |
61 | our $PROXY; |
66 | our $ACTIVE = 0; |
62 | our $ACTIVE = 0; |
67 | |
63 | |
68 | my %KA_COUNT; # number of open keep-alive connections per host |
64 | my %KA_CACHE; # indexed by uhost currently, points to [$handle...] array |
69 | my %CO_SLOT; # number of open connections, and wait queue, per host |
65 | my %CO_SLOT; # number of open connections, and wait queue, per host |
70 | |
66 | |
71 | =item http_get $url, key => value..., $cb->($data, $headers) |
67 | =item http_get $url, key => value..., $cb->($data, $headers) |
72 | |
68 | |
73 | Executes an HTTP-GET request. See the http_request function for details on |
69 | Executes an HTTP-GET request. See the http_request function for details on |
… | |
… | |
169 | C<Host:>, C<Content-Length:>, C<Connection:> and C<Cookie:> headers and |
165 | C<Host:>, C<Content-Length:>, C<Connection:> and C<Cookie:> headers and |
170 | will provide defaults at least for C<TE:>, C<Referer:> and C<User-Agent:> |
166 | will provide defaults at least for C<TE:>, C<Referer:> and C<User-Agent:> |
171 | (this can be suppressed by using C<undef> for these headers in which case |
167 | (this can be suppressed by using C<undef> for these headers in which case |
172 | they won't be sent at all). |
168 | they won't be sent at all). |
173 | |
169 | |
|
|
170 | You really should provide your own C<User-Agent:> header value that is |
|
|
171 | appropriate for your program - I wouldn't be surprised if the default |
|
|
172 | AnyEvent string gets blocked by webservers sooner or later. |
|
|
173 | |
|
|
174 | Also, make sure that your headers names and values do not contain any |
|
|
175 | embedded newlines. |
|
|
176 | |
174 | =item timeout => $seconds |
177 | =item timeout => $seconds |
175 | |
178 | |
176 | The time-out to use for various stages - each connect attempt will reset |
179 | The time-out to use for various stages - each connect attempt will reset |
177 | the timeout, as will read or write activity, i.e. this is not an overall |
180 | the timeout, as will read or write activity, i.e. this is not an overall |
178 | timeout. |
181 | timeout. |
179 | |
182 | |
180 | Default timeout is 5 minutes. |
183 | Default timeout is 5 minutes. |
181 | |
184 | |
182 | =item proxy => [$host, $port[, $scheme]] or undef |
185 | =item proxy => [$host, $port[, $scheme]] or undef |
183 | |
186 | |
184 | Use the given http proxy for all requests. If not specified, then the |
187 | Use the given http proxy for all requests, or no proxy if C<undef> is |
185 | default proxy (as specified by C<$ENV{http_proxy}>) is used. |
188 | used. |
186 | |
189 | |
187 | C<$scheme> must be either missing, C<http> for HTTP or C<https> for |
190 | C<$scheme> must be either missing or must be C<http> for HTTP. |
188 | HTTPS. |
191 | |
|
|
192 | If not specified, then the default proxy is used (see |
|
|
193 | C<AnyEvent::HTTP::set_proxy>). |
189 | |
194 | |
190 | =item body => $string |
195 | =item body => $string |
191 | |
196 | |
192 | The request body, usually empty. Will be sent as-is (future versions of |
197 | The request body, usually empty. Will be sent as-is (future versions of |
193 | this module might offer more options). |
198 | this module might offer more options). |
… | |
… | |
223 | verification, highest compatibility) and high-security (CA and common-name |
228 | verification, highest compatibility) and high-security (CA and common-name |
224 | verification) TLS context. |
229 | verification) TLS context. |
225 | |
230 | |
226 | The default for this option is C<low>, which could be interpreted as "give |
231 | The default for this option is C<low>, which could be interpreted as "give |
227 | me the page, no matter what". |
232 | me the page, no matter what". |
|
|
233 | |
|
|
234 | See also the C<sessionid> parameter. |
|
|
235 | |
|
|
236 | =item session => $string |
|
|
237 | |
|
|
238 | The module might reuse connections to the same host internally. Sometimes |
|
|
239 | (e.g. when using TLS), you do not want to reuse connections from other |
|
|
240 | sessions. This can be achieved by setting this parameter to some unique |
|
|
241 | ID (such as the address of an object storing your state data, or the TLS |
|
|
242 | context) - only connections using the same unique ID will be reused. |
228 | |
243 | |
229 | =item on_prepare => $callback->($fh) |
244 | =item on_prepare => $callback->($fh) |
230 | |
245 | |
231 | In rare cases you need to "tune" the socket before it is used to |
246 | In rare cases you need to "tune" the socket before it is used to |
232 | connect (for exmaple, to bind it on a given IP address). This parameter |
247 | connect (for exmaple, to bind it on a given IP address). This parameter |
… | |
… | |
302 | called. Instead of the C<$body> argument containing the body data, the |
317 | called. Instead of the C<$body> argument containing the body data, the |
303 | callback will receive the L<AnyEvent::Handle> object associated with the |
318 | callback will receive the L<AnyEvent::Handle> object associated with the |
304 | connection. In error cases, C<undef> will be passed. When there is no body |
319 | connection. In error cases, C<undef> will be passed. When there is no body |
305 | (e.g. status C<304>), the empty string will be passed. |
320 | (e.g. status C<304>), the empty string will be passed. |
306 | |
321 | |
307 | The handle object might or might not be in TLS mode, might be connected to |
322 | The handle object might or might not be in TLS mode, might be connected |
308 | a proxy, be a persistent connection etc., and configured in unspecified |
323 | to a proxy, be a persistent connection, use chunked transfer encoding |
309 | ways. The user is responsible for this handle (it will not be used by this |
324 | etc., and configured in unspecified ways. The user is responsible for this |
310 | module anymore). |
325 | handle (it will not be used by this module anymore). |
311 | |
326 | |
312 | This is useful with some push-type services, where, after the initial |
327 | This is useful with some push-type services, where, after the initial |
313 | headers, an interactive protocol is used (typical example would be the |
328 | headers, an interactive protocol is used (typical example would be the |
314 | push-style twitter API which starts a JSON/XML stream). |
329 | push-style twitter API which starts a JSON/XML stream). |
315 | |
330 | |
316 | If you think you need this, first have a look at C<on_body>, to see if |
331 | If you think you need this, first have a look at C<on_body>, to see if |
317 | that doesn't solve your problem in a better way. |
332 | that doesn't solve your problem in a better way. |
|
|
333 | |
|
|
334 | =item persistent => $boolean |
|
|
335 | |
|
|
336 | Try to create/reuse a persistent connection. When this flag is set |
|
|
337 | (default: true for idempotent requests, false for all others), then |
|
|
338 | C<http_request> tries to re-use an existing (previously-created) |
|
|
339 | persistent connection to the host and, failing that, tries to create a new |
|
|
340 | one. |
|
|
341 | |
|
|
342 | Requests failing in certain ways will be automatically retried once, which |
|
|
343 | is dangerous for non-idempotent requests, which is why it defaults to off |
|
|
344 | for them. The reason for this is because the bozos who designed HTTP/1.1 |
|
|
345 | made it impossible to distinguish between a fatal error and a normal |
|
|
346 | connection timeout, so you never know whether there was a problem with |
|
|
347 | your request or not. |
|
|
348 | |
|
|
349 | When reusing an existent connection, many parameters (such as TLS context) |
|
|
350 | will be ignored. See the C<session> parameter for a workaround. |
|
|
351 | |
|
|
352 | =item keepalive => $boolean |
|
|
353 | |
|
|
354 | Only used when C<persistent> is also true. This parameter decides whether |
|
|
355 | C<http_request> tries to handshake a HTTP/1.0-style keep-alive connection |
|
|
356 | (as opposed to only a HTTP/1.1 persistent connection). |
|
|
357 | |
|
|
358 | The default is true, except when using a proxy, in which case it defaults |
|
|
359 | to false, as HTTP/1.0 proxies cannot support this in a meaningful way. |
|
|
360 | |
|
|
361 | =item handle_params => { key => value ... } |
|
|
362 | |
|
|
363 | The key-value pairs in this hash will be passed to any L<AnyEvent::Handle> |
|
|
364 | constructor that is called - not all requests will create a handle, and |
|
|
365 | sometimes more than one is created, so this parameter is only good for |
|
|
366 | setting hints. |
|
|
367 | |
|
|
368 | Example: set the maximum read size to 4096, to potentially conserve memory |
|
|
369 | at the cost of speed. |
|
|
370 | |
|
|
371 | handle_params => { |
|
|
372 | max_read_size => 4096, |
|
|
373 | }, |
318 | |
374 | |
319 | =back |
375 | =back |
320 | |
376 | |
321 | Example: do a simple HTTP GET request for http://www.nethype.de/ and print |
377 | Example: do a simple HTTP GET request for http://www.nethype.de/ and print |
322 | the response body. |
378 | the response body. |
… | |
… | |
328 | |
384 | |
329 | Example: do a HTTP HEAD request on https://www.google.com/, use a |
385 | Example: do a HTTP HEAD request on https://www.google.com/, use a |
330 | timeout of 30 seconds. |
386 | timeout of 30 seconds. |
331 | |
387 | |
332 | http_request |
388 | http_request |
333 | GET => "https://www.google.com", |
389 | HEAD => "https://www.google.com", |
|
|
390 | headers => { "user-agent" => "MySearchClient 1.0" }, |
334 | timeout => 30, |
391 | timeout => 30, |
335 | sub { |
392 | sub { |
336 | my ($body, $hdr) = @_; |
393 | my ($body, $hdr) = @_; |
337 | use Data::Dumper; |
394 | use Data::Dumper; |
338 | print Dumper $hdr; |
395 | print Dumper $hdr; |
… | |
… | |
348 | }; |
405 | }; |
349 | |
406 | |
350 | undef $request; |
407 | undef $request; |
351 | |
408 | |
352 | =cut |
409 | =cut |
|
|
410 | |
|
|
411 | ############################################################################# |
|
|
412 | # wait queue/slots |
353 | |
413 | |
354 | sub _slot_schedule; |
414 | sub _slot_schedule; |
355 | sub _slot_schedule($) { |
415 | sub _slot_schedule($) { |
356 | my $host = shift; |
416 | my $host = shift; |
357 | |
417 | |
… | |
… | |
380 | |
440 | |
381 | _slot_schedule $_[0]; |
441 | _slot_schedule $_[0]; |
382 | } |
442 | } |
383 | |
443 | |
384 | ############################################################################# |
444 | ############################################################################# |
|
|
445 | # cookie handling |
385 | |
446 | |
386 | # expire cookies |
447 | # expire cookies |
387 | sub cookie_jar_expire($;$) { |
448 | sub cookie_jar_expire($;$) { |
388 | my ($jar, $session_end) = @_; |
449 | my ($jar, $session_end) = @_; |
389 | |
450 | |
… | |
… | |
413 | } |
474 | } |
414 | } |
475 | } |
415 | |
476 | |
416 | # extract cookies from jar |
477 | # extract cookies from jar |
417 | sub cookie_jar_extract($$$$) { |
478 | sub cookie_jar_extract($$$$) { |
418 | my ($jar, $uscheme, $uhost, $upath) = @_; |
479 | my ($jar, $scheme, $host, $path) = @_; |
419 | |
480 | |
420 | %$jar = () if $jar->{version} != 1; |
481 | %$jar = () if $jar->{version} != 1; |
421 | |
482 | |
422 | my @cookies; |
483 | my @cookies; |
423 | |
484 | |
424 | while (my ($chost, $paths) = each %$jar) { |
485 | while (my ($chost, $paths) = each %$jar) { |
425 | next unless ref $paths; |
486 | next unless ref $paths; |
426 | |
487 | |
427 | if ($chost =~ /^\./) { |
488 | if ($chost =~ /^\./) { |
428 | next unless $chost eq substr $uhost, -length $chost; |
489 | next unless $chost eq substr $host, -length $chost; |
429 | } elsif ($chost =~ /\./) { |
490 | } elsif ($chost =~ /\./) { |
430 | next unless $chost eq $uhost; |
491 | next unless $chost eq $host; |
431 | } else { |
492 | } else { |
432 | next; |
493 | next; |
433 | } |
494 | } |
434 | |
495 | |
435 | while (my ($cpath, $cookies) = each %$paths) { |
496 | while (my ($cpath, $cookies) = each %$paths) { |
436 | next unless $cpath eq substr $upath, 0, length $cpath; |
497 | next unless $cpath eq substr $path, 0, length $cpath; |
437 | |
498 | |
438 | while (my ($cookie, $kv) = each %$cookies) { |
499 | while (my ($cookie, $kv) = each %$cookies) { |
439 | next if $uscheme ne "https" && exists $kv->{secure}; |
500 | next if $scheme ne "https" && exists $kv->{secure}; |
440 | |
501 | |
441 | if (exists $kv->{_expires} and AE::now > $kv->{_expires}) { |
502 | if (exists $kv->{_expires} and AE::now > $kv->{_expires}) { |
442 | delete $cookies->{$cookie}; |
503 | delete $cookies->{$cookie}; |
443 | next; |
504 | next; |
444 | } |
505 | } |
… | |
… | |
458 | \@cookies |
519 | \@cookies |
459 | } |
520 | } |
460 | |
521 | |
461 | # parse set_cookie header into jar |
522 | # parse set_cookie header into jar |
462 | sub cookie_jar_set_cookie($$$$) { |
523 | sub cookie_jar_set_cookie($$$$) { |
463 | my ($jar, $set_cookie, $uhost, $date) = @_; |
524 | my ($jar, $set_cookie, $host, $date) = @_; |
464 | |
525 | |
465 | my $anow = int AE::now; |
526 | my $anow = int AE::now; |
466 | my $snow; # server-now |
527 | my $snow; # server-now |
467 | |
528 | |
468 | for ($set_cookie) { |
529 | for ($set_cookie) { |
… | |
… | |
474 | while ( |
535 | while ( |
475 | m{ |
536 | m{ |
476 | \G\s* |
537 | \G\s* |
477 | (?: |
538 | (?: |
478 | expires \s*=\s* ([A-Z][a-z][a-z]+,\ [^,;]+) |
539 | expires \s*=\s* ([A-Z][a-z][a-z]+,\ [^,;]+) |
479 | | ([^=;,[:space:]]+) (?: \s*=\s* (?: "((?:[^\\"]+|\\.)*)" | ([^=;,[:space:]]*) ) )? |
540 | | ([^=;,[:space:]]+) (?: \s*=\s* (?: "((?:[^\\"]+|\\.)*)" | ([^;,[:space:]]*) ) )? |
480 | ) |
541 | ) |
481 | }gcxsi |
542 | }gcxsi |
482 | ) { |
543 | ) { |
483 | my $name = $2; |
544 | my $name = $2; |
484 | my $value = $4; |
545 | my $value = $4; |
… | |
… | |
491 | # quoted |
552 | # quoted |
492 | $value = $3; |
553 | $value = $3; |
493 | $value =~ s/\\(.)/$1/gs; |
554 | $value =~ s/\\(.)/$1/gs; |
494 | } |
555 | } |
495 | |
556 | |
496 | push @kv, lc $name, $value; |
557 | push @kv, @kv ? lc $name : $name, $value; |
497 | |
558 | |
498 | last unless /\G\s*;/gc; |
559 | last unless /\G\s*;/gc; |
499 | } |
560 | } |
500 | |
561 | |
501 | last unless @kv; |
562 | last unless @kv; |
… | |
… | |
524 | |
585 | |
525 | # this is not rfc-like and not netscape-like. go figure. |
586 | # this is not rfc-like and not netscape-like. go figure. |
526 | my $ndots = $cdom =~ y/.//; |
587 | my $ndots = $cdom =~ y/.//; |
527 | next if $ndots < ($cdom =~ /\.[^.][^.]\.[^.][^.]$/ ? 3 : 2); |
588 | next if $ndots < ($cdom =~ /\.[^.][^.]\.[^.][^.]$/ ? 3 : 2); |
528 | } else { |
589 | } else { |
529 | $cdom = $uhost; |
590 | $cdom = $host; |
530 | } |
591 | } |
531 | |
592 | |
532 | # store it |
593 | # store it |
533 | $jar->{version} = 1; |
594 | $jar->{version} = 1; |
534 | $jar->{lc $cdom}{$cpath}{$name} = \%kv; |
595 | $jar->{lc $cdom}{$cpath}{$name} = \%kv; |
535 | |
596 | |
536 | redo if /\G\s*,/gc; |
597 | redo if /\G\s*,/gc; |
537 | } |
598 | } |
538 | } |
599 | } |
539 | |
600 | |
|
|
601 | ############################################################################# |
|
|
602 | # keepalive/persistent connection cache |
|
|
603 | |
|
|
604 | # fetch a connection from the keepalive cache |
|
|
605 | sub ka_fetch($) { |
|
|
606 | my $ka_key = shift; |
|
|
607 | |
|
|
608 | my $hdl = pop @{ $KA_CACHE{$ka_key} }; # currently we reuse the MOST RECENTLY USED connection |
|
|
609 | delete $KA_CACHE{$ka_key} |
|
|
610 | unless @{ $KA_CACHE{$ka_key} }; |
|
|
611 | |
|
|
612 | $hdl |
|
|
613 | } |
|
|
614 | |
|
|
615 | sub ka_store($$) { |
|
|
616 | my ($ka_key, $hdl) = @_; |
|
|
617 | |
|
|
618 | my $kaa = $KA_CACHE{$ka_key} ||= []; |
|
|
619 | |
|
|
620 | my $destroy = sub { |
|
|
621 | my @ka = grep $_ != $hdl, @{ $KA_CACHE{$ka_key} }; |
|
|
622 | |
|
|
623 | $hdl->destroy; |
|
|
624 | |
|
|
625 | @ka |
|
|
626 | ? $KA_CACHE{$ka_key} = \@ka |
|
|
627 | : delete $KA_CACHE{$ka_key}; |
|
|
628 | }; |
|
|
629 | |
|
|
630 | # on error etc., destroy |
|
|
631 | $hdl->on_error ($destroy); |
|
|
632 | $hdl->on_eof ($destroy); |
|
|
633 | $hdl->on_read ($destroy); |
|
|
634 | $hdl->timeout ($PERSISTENT_TIMEOUT); |
|
|
635 | |
|
|
636 | push @$kaa, $hdl; |
|
|
637 | shift @$kaa while @$kaa > $MAX_PER_HOST; |
|
|
638 | } |
|
|
639 | |
|
|
640 | ############################################################################# |
|
|
641 | # utilities |
|
|
642 | |
540 | # continue to parse $_ for headers and place them into the arg |
643 | # continue to parse $_ for headers and place them into the arg |
541 | sub parse_hdr() { |
644 | sub _parse_hdr() { |
542 | my %hdr; |
645 | my %hdr; |
543 | |
646 | |
544 | # things seen, not parsed: |
647 | # things seen, not parsed: |
545 | # p3pP="NON CUR OTPi OUR NOR UNI" |
648 | # p3pP="NON CUR OTPi OUR NOR UNI" |
546 | |
649 | |
… | |
… | |
560 | for values %hdr; |
663 | for values %hdr; |
561 | |
664 | |
562 | \%hdr |
665 | \%hdr |
563 | } |
666 | } |
564 | |
667 | |
|
|
668 | ############################################################################# |
|
|
669 | # http_get |
|
|
670 | |
565 | our $qr_nlnl = qr{(?<![^\012])\015?\012}; |
671 | our $qr_nlnl = qr{(?<![^\012])\015?\012}; |
566 | |
672 | |
567 | our $TLS_CTX_LOW = { cache => 1, sslv2 => 1 }; |
673 | our $TLS_CTX_LOW = { cache => 1, sslv2 => 1 }; |
568 | our $TLS_CTX_HIGH = { cache => 1, verify => 1, verify_peername => "https" }; |
674 | our $TLS_CTX_HIGH = { cache => 1, verify => 1, verify_peername => "https" }; |
|
|
675 | |
|
|
676 | # maybe it should just become a normal object :/ |
|
|
677 | |
|
|
678 | sub _destroy_state(\%) { |
|
|
679 | my ($state) = @_; |
|
|
680 | |
|
|
681 | $state->{handle}->destroy if $state->{handle}; |
|
|
682 | %$state = (); |
|
|
683 | } |
|
|
684 | |
|
|
685 | sub _error(\%$$) { |
|
|
686 | my ($state, $cb, $hdr) = @_; |
|
|
687 | |
|
|
688 | &_destroy_state ($state); |
|
|
689 | |
|
|
690 | $cb->(undef, $hdr); |
|
|
691 | () |
|
|
692 | } |
569 | |
693 | |
570 | sub http_request($$@) { |
694 | sub http_request($$@) { |
571 | my $cb = pop; |
695 | my $cb = pop; |
572 | my ($method, $url, %arg) = @_; |
696 | my ($method, $url, %arg) = @_; |
573 | |
697 | |
… | |
… | |
591 | my $recurse = exists $arg{recurse} ? delete $arg{recurse} : $MAX_RECURSE; |
715 | my $recurse = exists $arg{recurse} ? delete $arg{recurse} : $MAX_RECURSE; |
592 | |
716 | |
593 | return $cb->(undef, { @pseudo, Status => 599, Reason => "Too many redirections" }) |
717 | return $cb->(undef, { @pseudo, Status => 599, Reason => "Too many redirections" }) |
594 | if $recurse < 0; |
718 | if $recurse < 0; |
595 | |
719 | |
596 | my $proxy = $arg{proxy} || $PROXY; |
720 | my $proxy = exists $arg{proxy} ? $arg{proxy} : $PROXY; |
597 | my $timeout = $arg{timeout} || $TIMEOUT; |
721 | my $timeout = $arg{timeout} || $TIMEOUT; |
598 | |
722 | |
599 | my ($uscheme, $uauthority, $upath, $query, $fragment) = |
723 | my ($uscheme, $uauthority, $upath, $query, undef) = # ignore fragment |
600 | $url =~ m|(?:([^:/?#]+):)?(?://([^/?#]*))?([^?#]*)(?:(\?[^#]*))?(?:#(.*))?|; |
724 | $url =~ m|^([^:]+):(?://([^/?#]*))?([^?#]*)(?:(\?[^#]*))?(?:#(.*))?$|; |
601 | |
725 | |
602 | $uscheme = lc $uscheme; |
726 | $uscheme = lc $uscheme; |
603 | |
727 | |
604 | my $uport = $uscheme eq "http" ? 80 |
728 | my $uport = $uscheme eq "http" ? 80 |
605 | : $uscheme eq "https" ? 443 |
729 | : $uscheme eq "https" ? 443 |
… | |
… | |
649 | $hdr{"user-agent"} = $USERAGENT unless exists $hdr{"user-agent"}; |
773 | $hdr{"user-agent"} = $USERAGENT unless exists $hdr{"user-agent"}; |
650 | |
774 | |
651 | $hdr{"content-length"} = length $arg{body} |
775 | $hdr{"content-length"} = length $arg{body} |
652 | if length $arg{body} || $method ne "GET"; |
776 | if length $arg{body} || $method ne "GET"; |
653 | |
777 | |
654 | $hdr{connection} = "close Te"; #1.1 |
778 | my $idempotent = $method =~ /^(?:GET|HEAD|PUT|DELETE|OPTIONS|TRACE)$/; |
|
|
779 | |
|
|
780 | # default value for keepalive is true iff the request is for an idempotent method |
|
|
781 | my $persistent = exists $arg{persistent} ? !!$arg{persistent} : $idempotent; |
|
|
782 | my $keepalive = exists $arg{keepalive} ? !!$arg{keepalive} : !$proxy; |
|
|
783 | my $was_persistent; # true if this is actually a recycled connection |
|
|
784 | |
|
|
785 | # the key to use in the keepalive cache |
|
|
786 | my $ka_key = "$uscheme\x00$uhost\x00$uport\x00$arg{sessionid}"; |
|
|
787 | |
|
|
788 | $hdr{connection} = ($persistent ? $keepalive ? "keep-alive " : "" : "close ") . "Te"; #1.1 |
655 | $hdr{te} = "trailers" unless exists $hdr{te}; #1.1 |
789 | $hdr{te} = "trailers" unless exists $hdr{te}; #1.1 |
656 | |
790 | |
657 | my %state = (connect_guard => 1); |
791 | my %state = (connect_guard => 1); |
658 | |
792 | |
659 | _get_slot $uhost, sub { |
|
|
660 | $state{slot_guard} = shift; |
|
|
661 | |
|
|
662 | return unless $state{connect_guard}; |
|
|
663 | |
|
|
664 | my $ae_error = 595; # connecting |
793 | my $ae_error = 595; # connecting |
665 | |
794 | |
666 | # handle actual, non-tunneled, request |
795 | # handle actual, non-tunneled, request |
667 | my $handle_actual_request = sub { |
796 | my $handle_actual_request = sub { |
668 | $ae_error = 596; # request phase |
797 | $ae_error = 596; # request phase |
669 | |
798 | |
|
|
799 | my $hdl = $state{handle}; |
|
|
800 | |
670 | $state{handle}->starttls ("connect") if $uscheme eq "https" && !exists $state{handle}{tls}; |
801 | $hdl->starttls ("connect") if $uscheme eq "https" && !exists $hdl->{tls}; |
671 | |
802 | |
672 | # send request |
803 | # send request |
673 | $state{handle}->push_write ( |
804 | $hdl->push_write ( |
674 | "$method $rpath HTTP/1.1\015\012" |
805 | "$method $rpath HTTP/1.1\015\012" |
675 | . (join "", map "\u$_: $hdr{$_}\015\012", grep defined $hdr{$_}, keys %hdr) |
806 | . (join "", map "\u$_: $hdr{$_}\015\012", grep defined $hdr{$_}, keys %hdr) |
676 | . "\015\012" |
807 | . "\015\012" |
677 | . (delete $arg{body}) |
808 | . (delete $arg{body}) |
678 | ); |
809 | ); |
679 | |
810 | |
680 | # return if error occured during push_write() |
811 | # return if error occured during push_write() |
|
|
812 | return unless %state; |
|
|
813 | |
|
|
814 | # reduce memory usage, save a kitten, also re-use it for the response headers. |
|
|
815 | %hdr = (); |
|
|
816 | |
|
|
817 | # status line and headers |
|
|
818 | $state{read_response} = sub { |
681 | return unless %state; |
819 | return unless %state; |
682 | |
820 | |
683 | %hdr = (); # reduce memory usage, save a kitten, also make it possible to re-use |
|
|
684 | |
|
|
685 | # status line and headers |
|
|
686 | $state{read_response} = sub { |
|
|
687 | for ("$_[1]") { |
821 | for ("$_[1]") { |
688 | y/\015//d; # weed out any \015, as they show up in the weirdest of places. |
822 | y/\015//d; # weed out any \015, as they show up in the weirdest of places. |
689 | |
823 | |
690 | /^HTTP\/0*([0-9\.]+) \s+ ([0-9]{3}) (?: \s+ ([^\012]*) )? \012/gxci |
824 | /^HTTP\/0*([0-9\.]+) \s+ ([0-9]{3}) (?: \s+ ([^\012]*) )? \012/gxci |
691 | or return (%state = (), $cb->(undef, { @pseudo, Status => 599, Reason => "Invalid server response" })); |
825 | or return _error %state, $cb, { @pseudo, Status => 599, Reason => "Invalid server response" }; |
692 | |
826 | |
693 | # 100 Continue handling |
827 | # 100 Continue handling |
694 | # should not happen as we don't send expect: 100-continue, |
828 | # should not happen as we don't send expect: 100-continue, |
695 | # but we handle it just in case. |
829 | # but we handle it just in case. |
696 | # since we send the request body regardless, if we get an error |
830 | # since we send the request body regardless, if we get an error |
697 | # we are out of-sync, which we currently do NOT handle correctly. |
831 | # we are out of-sync, which we currently do NOT handle correctly. |
698 | return $state{handle}->push_read (line => $qr_nlnl, $state{read_response}) |
832 | return $state{handle}->push_read (line => $qr_nlnl, $state{read_response}) |
699 | if $2 eq 100; |
833 | if $2 eq 100; |
700 | |
834 | |
701 | push @pseudo, |
835 | push @pseudo, |
702 | HTTPVersion => $1, |
836 | HTTPVersion => $1, |
703 | Status => $2, |
837 | Status => $2, |
704 | Reason => $3, |
838 | Reason => $3, |
705 | ; |
839 | ; |
706 | |
840 | |
707 | my $hdr = parse_hdr |
841 | my $hdr = _parse_hdr |
708 | or return (%state = (), $cb->(undef, { @pseudo, Status => 599, Reason => "Garbled response headers" })); |
842 | or return _error %state, $cb, { @pseudo, Status => 599, Reason => "Garbled response headers" }; |
709 | |
843 | |
710 | %hdr = (%$hdr, @pseudo); |
844 | %hdr = (%$hdr, @pseudo); |
|
|
845 | } |
|
|
846 | |
|
|
847 | # redirect handling |
|
|
848 | # microsoft and other shitheads don't give a shit for following standards, |
|
|
849 | # try to support some common forms of broken Location headers. |
|
|
850 | if ($hdr{location} !~ /^(?: $ | [^:\/?\#]+ : )/x) { |
|
|
851 | $hdr{location} =~ s/^\.\/+//; |
|
|
852 | |
|
|
853 | my $url = "$rscheme://$uhost:$uport"; |
|
|
854 | |
|
|
855 | unless ($hdr{location} =~ s/^\///) { |
|
|
856 | $url .= $upath; |
|
|
857 | $url =~ s/\/[^\/]*$//; |
711 | } |
858 | } |
712 | |
859 | |
713 | # redirect handling |
|
|
714 | # microsoft and other shitheads don't give a shit for following standards, |
|
|
715 | # try to support some common forms of broken Location headers. |
|
|
716 | if ($hdr{location} !~ /^(?: $ | [^:\/?\#]+ : )/x) { |
|
|
717 | $hdr{location} =~ s/^\.\/+//; |
|
|
718 | |
|
|
719 | my $url = "$rscheme://$uhost:$uport"; |
|
|
720 | |
|
|
721 | unless ($hdr{location} =~ s/^\///) { |
|
|
722 | $url .= $upath; |
|
|
723 | $url =~ s/\/[^\/]*$//; |
|
|
724 | } |
|
|
725 | |
|
|
726 | $hdr{location} = "$url/$hdr{location}"; |
860 | $hdr{location} = "$url/$hdr{location}"; |
|
|
861 | } |
|
|
862 | |
|
|
863 | my $redirect; |
|
|
864 | |
|
|
865 | if ($recurse) { |
|
|
866 | my $status = $hdr{Status}; |
|
|
867 | |
|
|
868 | # industry standard is to redirect POST as GET for |
|
|
869 | # 301, 302 and 303, in contrast to HTTP/1.0 and 1.1. |
|
|
870 | # also, the UA should ask the user for 301 and 307 and POST, |
|
|
871 | # industry standard seems to be to simply follow. |
|
|
872 | # we go with the industry standard. |
|
|
873 | if ($status == 301 or $status == 302 or $status == 303) { |
|
|
874 | # HTTP/1.1 is unclear on how to mutate the method |
|
|
875 | $method = "GET" unless $method eq "HEAD"; |
|
|
876 | $redirect = 1; |
|
|
877 | } elsif ($status == 307) { |
|
|
878 | $redirect = 1; |
727 | } |
879 | } |
|
|
880 | } |
728 | |
881 | |
729 | my $redirect; |
882 | my $finish = sub { # ($data, $err_status, $err_reason[, $persistent]) |
730 | |
883 | if ($state{handle}) { |
731 | if ($recurse) { |
884 | # handle keepalive |
732 | my $status = $hdr{Status}; |
885 | if ( |
733 | |
886 | $persistent |
734 | # industry standard is to redirect POST as GET for |
887 | && $_[3] |
735 | # 301, 302 and 303, in contrast to HTTP/1.0 and 1.1. |
888 | && ($hdr{HTTPVersion} < 1.1 |
736 | # also, the UA should ask the user for 301 and 307 and POST, |
889 | ? $hdr{connection} =~ /\bkeep-?alive\b/i |
737 | # industry standard seems to be to simply follow. |
890 | : $hdr{connection} !~ /\bclose\b/i) |
738 | # we go with the industry standard. |
891 | ) { |
739 | if ($status == 301 or $status == 302 or $status == 303) { |
892 | ka_store $ka_key, delete $state{handle}; |
740 | # HTTP/1.1 is unclear on how to mutate the method |
893 | } else { |
741 | $method = "GET" unless $method eq "HEAD"; |
894 | # no keepalive, destroy the handle |
742 | $redirect = 1; |
895 | $state{handle}->destroy; |
743 | } elsif ($status == 307) { |
|
|
744 | $redirect = 1; |
|
|
745 | } |
896 | } |
746 | } |
897 | } |
747 | |
898 | |
748 | my $finish = sub { # ($data, $err_status, $err_reason[, $keepalive]) |
|
|
749 | my $may_keep_alive = $_[3]; |
|
|
750 | |
|
|
751 | $state{handle}->destroy if $state{handle}; |
|
|
752 | %state = (); |
899 | %state = (); |
753 | |
900 | |
754 | if (defined $_[1]) { |
901 | if (defined $_[1]) { |
755 | $hdr{OrigStatus} = $hdr{Status}; $hdr{Status} = $_[1]; |
902 | $hdr{OrigStatus} = $hdr{Status}; $hdr{Status} = $_[1]; |
756 | $hdr{OrigReason} = $hdr{Reason}; $hdr{Reason} = $_[2]; |
903 | $hdr{OrigReason} = $hdr{Reason}; $hdr{Reason} = $_[2]; |
757 | } |
904 | } |
758 | |
905 | |
759 | # set-cookie processing |
906 | # set-cookie processing |
760 | if ($arg{cookie_jar}) { |
907 | if ($arg{cookie_jar}) { |
761 | cookie_jar_set_cookie $arg{cookie_jar}, $hdr{"set-cookie"}, $uhost, $hdr{date}; |
908 | cookie_jar_set_cookie $arg{cookie_jar}, $hdr{"set-cookie"}, $uhost, $hdr{date}; |
762 | } |
909 | } |
763 | |
910 | |
764 | if ($redirect && exists $hdr{location}) { |
911 | if ($redirect && exists $hdr{location}) { |
765 | # we ignore any errors, as it is very common to receive |
912 | # we ignore any errors, as it is very common to receive |
766 | # Content-Length != 0 but no actual body |
913 | # Content-Length != 0 but no actual body |
767 | # we also access %hdr, as $_[1] might be an erro |
914 | # we also access %hdr, as $_[1] might be an erro |
|
|
915 | $state{recurse} = |
768 | http_request ( |
916 | http_request ( |
769 | $method => $hdr{location}, |
917 | $method => $hdr{location}, |
770 | %arg, |
918 | %arg, |
771 | recurse => $recurse - 1, |
919 | recurse => $recurse - 1, |
772 | Redirect => [$_[0], \%hdr], |
920 | Redirect => [$_[0], \%hdr], |
|
|
921 | sub { |
|
|
922 | %state = (); |
|
|
923 | &$cb |
|
|
924 | }, |
|
|
925 | ); |
|
|
926 | } else { |
|
|
927 | $cb->($_[0], \%hdr); |
|
|
928 | } |
|
|
929 | }; |
|
|
930 | |
|
|
931 | $ae_error = 597; # body phase |
|
|
932 | |
|
|
933 | my $chunked = $hdr{"transfer-encoding"} =~ /\bchunked\b/i; # not quite correct... |
|
|
934 | |
|
|
935 | my $len = $chunked ? undef : $hdr{"content-length"}; |
|
|
936 | |
|
|
937 | # body handling, many different code paths |
|
|
938 | # - no body expected |
|
|
939 | # - want_body_handle |
|
|
940 | # - te chunked |
|
|
941 | # - 2x length known (with or without on_body) |
|
|
942 | # - 2x length not known (with or without on_body) |
|
|
943 | if (!$redirect && $arg{on_header} && !$arg{on_header}(\%hdr)) { |
|
|
944 | $finish->(undef, 598 => "Request cancelled by on_header"); |
|
|
945 | } elsif ( |
|
|
946 | $hdr{Status} =~ /^(?:1..|204|205|304)$/ |
|
|
947 | or $method eq "HEAD" |
|
|
948 | or (defined $len && $len == 0) # == 0, not !, because "0 " is true |
|
|
949 | ) { |
|
|
950 | # no body |
|
|
951 | $finish->("", undef, undef, 1); |
|
|
952 | |
|
|
953 | } elsif (!$redirect && $arg{want_body_handle}) { |
|
|
954 | $_[0]->on_eof (undef); |
|
|
955 | $_[0]->on_error (undef); |
|
|
956 | $_[0]->on_read (undef); |
|
|
957 | |
|
|
958 | $finish->(delete $state{handle}); |
|
|
959 | |
|
|
960 | } elsif ($chunked) { |
|
|
961 | my $cl = 0; |
|
|
962 | my $body = ""; |
|
|
963 | my $on_body = $arg{on_body} || sub { $body .= shift; 1 }; |
|
|
964 | |
|
|
965 | $state{read_chunk} = sub { |
|
|
966 | $_[1] =~ /^([0-9a-fA-F]+)/ |
|
|
967 | or return $finish->(undef, $ae_error => "Garbled chunked transfer encoding"); |
|
|
968 | |
|
|
969 | my $len = hex $1; |
|
|
970 | |
|
|
971 | if ($len) { |
|
|
972 | $cl += $len; |
|
|
973 | |
|
|
974 | $_[0]->push_read (chunk => $len, sub { |
|
|
975 | $on_body->($_[1], \%hdr) |
|
|
976 | or return $finish->(undef, 598 => "Request cancelled by on_body"); |
|
|
977 | |
|
|
978 | $_[0]->push_read (line => sub { |
|
|
979 | length $_[1] |
|
|
980 | and return $finish->(undef, $ae_error => "Garbled chunked transfer encoding"); |
|
|
981 | $_[0]->push_read (line => $state{read_chunk}); |
773 | $cb); |
982 | }); |
|
|
983 | }); |
774 | } else { |
984 | } else { |
775 | $cb->($_[0], \%hdr); |
985 | $hdr{"content-length"} ||= $cl; |
|
|
986 | |
|
|
987 | $_[0]->push_read (line => $qr_nlnl, sub { |
|
|
988 | if (length $_[1]) { |
|
|
989 | for ("$_[1]") { |
|
|
990 | y/\015//d; # weed out any \015, as they show up in the weirdest of places. |
|
|
991 | |
|
|
992 | my $hdr = _parse_hdr |
|
|
993 | or return $finish->(undef, $ae_error => "Garbled response trailers"); |
|
|
994 | |
|
|
995 | %hdr = (%hdr, %$hdr); |
|
|
996 | } |
|
|
997 | } |
|
|
998 | |
|
|
999 | $finish->($body, undef, undef, 1); |
|
|
1000 | }); |
776 | } |
1001 | } |
777 | }; |
1002 | }; |
778 | |
1003 | |
779 | $ae_error = 597; # body phase |
|
|
780 | |
|
|
781 | my $len = $hdr{"content-length"}; |
|
|
782 | |
|
|
783 | # body handling, many different code paths |
|
|
784 | # - no body expected |
|
|
785 | # - want_body_handle |
|
|
786 | # - te chunked |
|
|
787 | # - 2x length known (with or without on_body) |
|
|
788 | # - 2x length not known (with or without on_body) |
|
|
789 | if (!$redirect && $arg{on_header} && !$arg{on_header}(\%hdr)) { |
|
|
790 | $finish->(undef, 598 => "Request cancelled by on_header"); |
|
|
791 | } elsif ( |
|
|
792 | $hdr{Status} =~ /^(?:1..|204|205|304)$/ |
|
|
793 | or $method eq "HEAD" |
|
|
794 | or (defined $len && $len == 0) # == 0, not !, because "0 " is true |
|
|
795 | ) { |
|
|
796 | # no body |
|
|
797 | $finish->("", undef, undef, 1); |
|
|
798 | |
|
|
799 | } elsif (!$redirect && $arg{want_body_handle}) { |
|
|
800 | $_[0]->on_eof (undef); |
|
|
801 | $_[0]->on_error (undef); |
|
|
802 | $_[0]->on_read (undef); |
|
|
803 | |
|
|
804 | $finish->(delete $state{handle}); |
|
|
805 | |
|
|
806 | } elsif ($hdr{"transfer-encoding"} =~ /\bchunked\b/i) { |
|
|
807 | my $cl = 0; |
|
|
808 | my $body = undef; |
|
|
809 | my $on_body = $arg{on_body} || sub { $body .= shift; 1 }; |
|
|
810 | |
|
|
811 | $state{read_chunk} = sub { |
|
|
812 | $_[1] =~ /^([0-9a-fA-F]+)/ |
|
|
813 | or $finish->(undef, $ae_error => "Garbled chunked transfer encoding"); |
|
|
814 | |
|
|
815 | my $len = hex $1; |
|
|
816 | |
|
|
817 | if ($len) { |
|
|
818 | $cl += $len; |
|
|
819 | |
|
|
820 | $_[0]->push_read (chunk => $len, sub { |
|
|
821 | $on_body->($_[1], \%hdr) |
|
|
822 | or return $finish->(undef, 598 => "Request cancelled by on_body"); |
|
|
823 | |
|
|
824 | $_[0]->push_read (line => sub { |
|
|
825 | length $_[1] |
|
|
826 | and return $finish->(undef, $ae_error => "Garbled chunked transfer encoding"); |
|
|
827 | $_[0]->push_read (line => $state{read_chunk}); |
|
|
828 | }); |
|
|
829 | }); |
|
|
830 | } else { |
|
|
831 | $hdr{"content-length"} ||= $cl; |
|
|
832 | |
|
|
833 | $_[0]->push_read (line => $qr_nlnl, sub { |
|
|
834 | if (length $_[1]) { |
|
|
835 | for ("$_[1]") { |
|
|
836 | y/\015//d; # weed out any \015, as they show up in the weirdest of places. |
|
|
837 | |
|
|
838 | my $hdr = parse_hdr |
|
|
839 | or return $finish->(undef, $ae_error => "Garbled response trailers"); |
|
|
840 | |
|
|
841 | %hdr = (%hdr, %$hdr); |
|
|
842 | } |
|
|
843 | } |
|
|
844 | |
|
|
845 | $finish->($body, undef, undef, 1); |
|
|
846 | }); |
|
|
847 | } |
|
|
848 | }; |
|
|
849 | |
|
|
850 | $_[0]->push_read (line => $state{read_chunk}); |
1004 | $_[0]->push_read (line => $state{read_chunk}); |
851 | |
1005 | |
852 | } elsif ($arg{on_body}) { |
1006 | } elsif ($arg{on_body}) { |
853 | if (defined $len) { |
1007 | if (defined $len) { |
854 | $_[0]->on_read (sub { |
1008 | $_[0]->on_read (sub { |
855 | $len -= length $_[0]{rbuf}; |
1009 | $len -= length $_[0]{rbuf}; |
856 | |
1010 | |
857 | $arg{on_body}(delete $_[0]{rbuf}, \%hdr) |
1011 | $arg{on_body}(delete $_[0]{rbuf}, \%hdr) |
858 | or return $finish->(undef, 598 => "Request cancelled by on_body"); |
1012 | or return $finish->(undef, 598 => "Request cancelled by on_body"); |
859 | |
1013 | |
860 | $len > 0 |
1014 | $len > 0 |
861 | or $finish->("", undef, undef, 1); |
1015 | or $finish->("", undef, undef, 1); |
862 | }); |
|
|
863 | } else { |
|
|
864 | $_[0]->on_eof (sub { |
|
|
865 | $finish->(""); |
|
|
866 | }); |
|
|
867 | $_[0]->on_read (sub { |
|
|
868 | $arg{on_body}(delete $_[0]{rbuf}, \%hdr) |
|
|
869 | or $finish->(undef, 598 => "Request cancelled by on_body"); |
|
|
870 | }); |
|
|
871 | } |
1016 | }); |
872 | } else { |
1017 | } else { |
873 | $_[0]->on_eof (undef); |
1018 | $_[0]->on_eof (sub { |
874 | |
1019 | $finish->(""); |
875 | if (defined $len) { |
1020 | }); |
876 | $_[0]->on_read (sub { |
1021 | $_[0]->on_read (sub { |
877 | $finish->((substr delete $_[0]{rbuf}, 0, $len, ""), undef, undef, 1) |
1022 | $arg{on_body}(delete $_[0]{rbuf}, \%hdr) |
878 | if $len <= length $_[0]{rbuf}; |
1023 | or $finish->(undef, 598 => "Request cancelled by on_body"); |
879 | }); |
|
|
880 | } else { |
|
|
881 | $_[0]->on_error (sub { |
|
|
882 | ($! == Errno::EPIPE || !$!) |
|
|
883 | ? $finish->(delete $_[0]{rbuf}) |
|
|
884 | : $finish->(undef, $ae_error => $_[2]); |
|
|
885 | }); |
|
|
886 | $_[0]->on_read (sub { }); |
|
|
887 | } |
1024 | }); |
888 | } |
1025 | } |
889 | }; |
|
|
890 | |
|
|
891 | $state{handle}->push_read (line => $qr_nlnl, $state{read_response}); |
|
|
892 | }; |
|
|
893 | |
|
|
894 | my $connect_cb = sub { |
|
|
895 | $state{fh} = shift |
|
|
896 | or do { |
|
|
897 | my $err = "$!"; |
|
|
898 | %state = (); |
|
|
899 | return $cb->(undef, { @pseudo, Status => $ae_error, Reason => $err }); |
|
|
900 | }; |
|
|
901 | |
|
|
902 | return unless delete $state{connect_guard}; |
|
|
903 | |
|
|
904 | # get handle |
|
|
905 | $state{handle} = new AnyEvent::Handle |
|
|
906 | fh => $state{fh}, |
|
|
907 | peername => $rhost, |
|
|
908 | tls_ctx => $arg{tls_ctx}, |
|
|
909 | # these need to be reconfigured on keepalive handles |
|
|
910 | timeout => $timeout, |
|
|
911 | on_error => sub { |
|
|
912 | %state = (); |
|
|
913 | $cb->(undef, { @pseudo, Status => $ae_error, Reason => $_[2] }); |
|
|
914 | }, |
|
|
915 | on_eof => sub { |
|
|
916 | %state = (); |
|
|
917 | $cb->(undef, { @pseudo, Status => $ae_error, Reason => "Unexpected end-of-file" }); |
|
|
918 | }, |
|
|
919 | ; |
|
|
920 | |
|
|
921 | # limit the number of persistent connections |
|
|
922 | # keepalive not yet supported |
|
|
923 | # if ($KA_COUNT{$_[1]} < $MAX_PERSISTENT_PER_HOST) { |
|
|
924 | # ++$KA_COUNT{$_[1]}; |
|
|
925 | # $state{handle}{ka_count_guard} = AnyEvent::Util::guard { |
|
|
926 | # --$KA_COUNT{$_[1]} |
|
|
927 | # }; |
|
|
928 | # $hdr{connection} = "keep-alive"; |
|
|
929 | # } |
|
|
930 | |
|
|
931 | $state{handle}->starttls ("connect") if $rscheme eq "https"; |
|
|
932 | |
|
|
933 | # now handle proxy-CONNECT method |
|
|
934 | if ($proxy && $uscheme eq "https") { |
|
|
935 | # oh dear, we have to wrap it into a connect request |
|
|
936 | |
|
|
937 | # maybe re-use $uauthority with patched port? |
|
|
938 | $state{handle}->push_write ("CONNECT $uhost:$uport HTTP/1.0\015\012\015\012"); |
|
|
939 | $state{handle}->push_read (line => $qr_nlnl, sub { |
|
|
940 | $_[1] =~ /^HTTP\/([0-9\.]+) \s+ ([0-9]{3}) (?: \s+ ([^\015\012]*) )?/ix |
|
|
941 | or return (%state = (), $cb->(undef, { @pseudo, Status => 599, Reason => "Invalid proxy connect response ($_[1])" })); |
|
|
942 | |
|
|
943 | if ($2 == 200) { |
|
|
944 | $rpath = $upath; |
|
|
945 | $handle_actual_request->(); |
|
|
946 | } else { |
|
|
947 | %state = (); |
|
|
948 | $cb->(undef, { @pseudo, Status => $2, Reason => $3 }); |
|
|
949 | } |
|
|
950 | }); |
|
|
951 | } else { |
1026 | } else { |
952 | $handle_actual_request->(); |
1027 | $_[0]->on_eof (undef); |
|
|
1028 | |
|
|
1029 | if (defined $len) { |
|
|
1030 | $_[0]->on_read (sub { |
|
|
1031 | $finish->((substr delete $_[0]{rbuf}, 0, $len, ""), undef, undef, 1) |
|
|
1032 | if $len <= length $_[0]{rbuf}; |
|
|
1033 | }); |
|
|
1034 | } else { |
|
|
1035 | $_[0]->on_error (sub { |
|
|
1036 | ($! == Errno::EPIPE || !$!) |
|
|
1037 | ? $finish->(delete $_[0]{rbuf}) |
|
|
1038 | : $finish->(undef, $ae_error => $_[2]); |
|
|
1039 | }); |
|
|
1040 | $_[0]->on_read (sub { }); |
|
|
1041 | } |
953 | } |
1042 | } |
954 | }; |
1043 | }; |
955 | |
1044 | |
956 | my $tcp_connect = $arg{tcp_connect} |
1045 | # if keepalive is enabled, then the server closing the connection |
957 | || do { require AnyEvent::Socket; \&AnyEvent::Socket::tcp_connect }; |
1046 | # before a response can happen legally - we retry on idempotent methods. |
|
|
1047 | if ($was_persistent && $idempotent) { |
|
|
1048 | my $old_eof = $hdl->{on_eof}; |
|
|
1049 | $hdl->{on_eof} = sub { |
|
|
1050 | _destroy_state %state; |
958 | |
1051 | |
959 | $state{connect_guard} = $tcp_connect->($rhost, $rport, $connect_cb, $arg{on_prepare} || sub { $timeout }); |
1052 | %state = (); |
|
|
1053 | $state{recurse} = |
|
|
1054 | http_request ( |
|
|
1055 | $method => $url, |
|
|
1056 | %arg, |
|
|
1057 | keepalive => 0, |
|
|
1058 | sub { |
|
|
1059 | %state = (); |
|
|
1060 | &$cb |
|
|
1061 | } |
|
|
1062 | ); |
|
|
1063 | }; |
|
|
1064 | $hdl->on_read (sub { |
|
|
1065 | return unless %state; |
|
|
1066 | |
|
|
1067 | # as soon as we receive something, a connection close |
|
|
1068 | # once more becomes a hard error |
|
|
1069 | $hdl->{on_eof} = $old_eof; |
|
|
1070 | $hdl->push_read (line => $qr_nlnl, $state{read_response}); |
|
|
1071 | }); |
|
|
1072 | } else { |
|
|
1073 | $hdl->push_read (line => $qr_nlnl, $state{read_response}); |
|
|
1074 | } |
960 | }; |
1075 | }; |
961 | |
1076 | |
|
|
1077 | my $prepare_handle = sub { |
|
|
1078 | my ($hdl) = $state{handle}; |
|
|
1079 | |
|
|
1080 | $hdl->on_error (sub { |
|
|
1081 | _error %state, $cb, { @pseudo, Status => $ae_error, Reason => $_[2] }; |
|
|
1082 | }); |
|
|
1083 | $hdl->on_eof (sub { |
|
|
1084 | _error %state, $cb, { @pseudo, Status => $ae_error, Reason => "Unexpected end-of-file" }; |
|
|
1085 | }); |
|
|
1086 | $hdl->timeout_reset; |
|
|
1087 | $hdl->timeout ($timeout); |
|
|
1088 | }; |
|
|
1089 | |
|
|
1090 | # connected to proxy (or origin server) |
|
|
1091 | my $connect_cb = sub { |
|
|
1092 | my $fh = shift |
|
|
1093 | or return _error %state, $cb, { @pseudo, Status => $ae_error, Reason => "$!" }; |
|
|
1094 | |
|
|
1095 | return unless delete $state{connect_guard}; |
|
|
1096 | |
|
|
1097 | # get handle |
|
|
1098 | $state{handle} = new AnyEvent::Handle |
|
|
1099 | %{ $arg{handle_params} }, |
|
|
1100 | fh => $fh, |
|
|
1101 | peername => $uhost, |
|
|
1102 | tls_ctx => $arg{tls_ctx}, |
|
|
1103 | ; |
|
|
1104 | |
|
|
1105 | $prepare_handle->(); |
|
|
1106 | |
|
|
1107 | #$state{handle}->starttls ("connect") if $rscheme eq "https"; |
|
|
1108 | |
|
|
1109 | # now handle proxy-CONNECT method |
|
|
1110 | if ($proxy && $uscheme eq "https") { |
|
|
1111 | # oh dear, we have to wrap it into a connect request |
|
|
1112 | |
|
|
1113 | # maybe re-use $uauthority with patched port? |
|
|
1114 | $state{handle}->push_write ("CONNECT $uhost:$uport HTTP/1.0\015\012\015\012"); |
|
|
1115 | $state{handle}->push_read (line => $qr_nlnl, sub { |
|
|
1116 | $_[1] =~ /^HTTP\/([0-9\.]+) \s+ ([0-9]{3}) (?: \s+ ([^\015\012]*) )?/ix |
|
|
1117 | or return _error %state, $cb, { @pseudo, Status => 599, Reason => "Invalid proxy connect response ($_[1])" }; |
|
|
1118 | |
|
|
1119 | if ($2 == 200) { |
|
|
1120 | $rpath = $upath; |
|
|
1121 | $handle_actual_request->(); |
|
|
1122 | } else { |
|
|
1123 | _error %state, $cb, { @pseudo, Status => $2, Reason => $3 }; |
|
|
1124 | } |
|
|
1125 | }); |
|
|
1126 | } else { |
|
|
1127 | $handle_actual_request->(); |
|
|
1128 | } |
|
|
1129 | }; |
|
|
1130 | |
|
|
1131 | _get_slot $uhost, sub { |
|
|
1132 | $state{slot_guard} = shift; |
|
|
1133 | |
|
|
1134 | return unless $state{connect_guard}; |
|
|
1135 | |
|
|
1136 | # try to use an existing keepalive connection, but only if we, ourselves, plan |
|
|
1137 | # on a keepalive request (in theory, this should be a separate config option). |
|
|
1138 | if ($persistent && $KA_CACHE{$ka_key}) { |
|
|
1139 | $was_persistent = 1; |
|
|
1140 | |
|
|
1141 | $state{handle} = ka_fetch $ka_key; |
|
|
1142 | $state{handle}->destroyed |
|
|
1143 | and die "AnyEvent::HTTP: unexpectedly got a destructed handle (1), please report.";#d# |
|
|
1144 | $prepare_handle->(); |
|
|
1145 | $state{handle}->destroyed |
|
|
1146 | and die "AnyEvent::HTTP: unexpectedly got a destructed handle (2), please report.";#d# |
|
|
1147 | $handle_actual_request->(); |
|
|
1148 | |
|
|
1149 | } else { |
|
|
1150 | my $tcp_connect = $arg{tcp_connect} |
|
|
1151 | || do { require AnyEvent::Socket; \&AnyEvent::Socket::tcp_connect }; |
|
|
1152 | |
|
|
1153 | $state{connect_guard} = $tcp_connect->($rhost, $rport, $connect_cb, $arg{on_prepare} || sub { $timeout }); |
|
|
1154 | } |
|
|
1155 | }; |
|
|
1156 | |
962 | defined wantarray && AnyEvent::Util::guard { %state = () } |
1157 | defined wantarray && AnyEvent::Util::guard { _destroy_state %state } |
963 | } |
1158 | } |
964 | |
1159 | |
965 | sub http_get($@) { |
1160 | sub http_get($@) { |
966 | unshift @_, "GET"; |
1161 | unshift @_, "GET"; |
967 | &http_request |
1162 | &http_request |
… | |
… | |
985 | AnyEvent::HTTP uses the AnyEvent::Socket::tcp_connect function for |
1180 | AnyEvent::HTTP uses the AnyEvent::Socket::tcp_connect function for |
986 | the actual connection, which in turn uses AnyEvent::DNS to resolve |
1181 | the actual connection, which in turn uses AnyEvent::DNS to resolve |
987 | hostnames. The latter is a simple stub resolver and does no caching |
1182 | hostnames. The latter is a simple stub resolver and does no caching |
988 | on its own. If you want DNS caching, you currently have to provide |
1183 | on its own. If you want DNS caching, you currently have to provide |
989 | your own default resolver (by storing a suitable resolver object in |
1184 | your own default resolver (by storing a suitable resolver object in |
990 | C<$AnyEvent::DNS::RESOLVER>). |
1185 | C<$AnyEvent::DNS::RESOLVER>) or your own C<tcp_connect> callback. |
991 | |
1186 | |
992 | =head2 GLOBAL FUNCTIONS AND VARIABLES |
1187 | =head2 GLOBAL FUNCTIONS AND VARIABLES |
993 | |
1188 | |
994 | =over 4 |
1189 | =over 4 |
995 | |
1190 | |
996 | =item AnyEvent::HTTP::set_proxy "proxy-url" |
1191 | =item AnyEvent::HTTP::set_proxy "proxy-url" |
997 | |
1192 | |
998 | Sets the default proxy server to use. The proxy-url must begin with a |
1193 | Sets the default proxy server to use. The proxy-url must begin with a |
999 | string of the form C<http://host:port> (optionally C<https:...>), croaks |
1194 | string of the form C<http://host:port>, croaks otherwise. |
1000 | otherwise. |
|
|
1001 | |
1195 | |
1002 | To clear an already-set proxy, use C<undef>. |
1196 | To clear an already-set proxy, use C<undef>. |
|
|
1197 | |
|
|
1198 | When AnyEvent::HTTP is laoded for the first time it will query the |
|
|
1199 | default proxy from the operating system, currently by looking at |
|
|
1200 | C<$ENV{http_proxy>}. |
1003 | |
1201 | |
1004 | =item AnyEvent::HTTP::cookie_jar_expire $jar[, $session_end] |
1202 | =item AnyEvent::HTTP::cookie_jar_expire $jar[, $session_end] |
1005 | |
1203 | |
1006 | Remove all cookies from the cookie jar that have been expired. If |
1204 | Remove all cookies from the cookie jar that have been expired. If |
1007 | C<$session_end> is given and true, then additionally remove all session |
1205 | C<$session_end> is given and true, then additionally remove all session |
… | |
… | |
1053 | |
1251 | |
1054 | =item $AnyEvent::HTTP::MAX_RECURSE |
1252 | =item $AnyEvent::HTTP::MAX_RECURSE |
1055 | |
1253 | |
1056 | The default value for the C<recurse> request parameter (default: C<10>). |
1254 | The default value for the C<recurse> request parameter (default: C<10>). |
1057 | |
1255 | |
|
|
1256 | =item $AnyEvent::HTTP::TIMEOUT |
|
|
1257 | |
|
|
1258 | The default timeout for conenction operations (default: C<300>). |
|
|
1259 | |
1058 | =item $AnyEvent::HTTP::USERAGENT |
1260 | =item $AnyEvent::HTTP::USERAGENT |
1059 | |
1261 | |
1060 | The default value for the C<User-Agent> header (the default is |
1262 | The default value for the C<User-Agent> header (the default is |
1061 | C<Mozilla/5.0 (compatible; U; AnyEvent-HTTP/$VERSION; +http://software.schmorp.de/pkg/AnyEvent)>). |
1263 | C<Mozilla/5.0 (compatible; U; AnyEvent-HTTP/$VERSION; +http://software.schmorp.de/pkg/AnyEvent)>). |
1062 | |
1264 | |
1063 | =item $AnyEvent::HTTP::MAX_PER_HOST |
1265 | =item $AnyEvent::HTTP::MAX_PER_HOST |
1064 | |
1266 | |
1065 | The maximum number of concurrent connections to the same host (identified |
1267 | The maximum number of concurrent connections to the same host (identified |
1066 | by the hostname). If the limit is exceeded, then the additional requests |
1268 | by the hostname). If the limit is exceeded, then the additional requests |
1067 | are queued until previous connections are closed. |
1269 | are queued until previous connections are closed. Both persistent and |
|
|
1270 | non-persistent connections are counted in this limit. |
1068 | |
1271 | |
1069 | The default value for this is C<4>, and it is highly advisable to not |
1272 | The default value for this is C<4>, and it is highly advisable to not |
1070 | increase it. |
1273 | increase it much. |
|
|
1274 | |
|
|
1275 | For comparison: the RFC's recommend 4 non-persistent or 2 persistent |
|
|
1276 | connections, older browsers used 2, newers (such as firefox 3) typically |
|
|
1277 | use 6, and Opera uses 8 because like, they have the fastest browser and |
|
|
1278 | give a shit for everybody else on the planet. |
|
|
1279 | |
|
|
1280 | =item $AnyEvent::HTTP::PERSISTENT_TIMEOUT |
|
|
1281 | |
|
|
1282 | The time after which idle persistent conenctions get closed by |
|
|
1283 | AnyEvent::HTTP (default: C<3>). |
1071 | |
1284 | |
1072 | =item $AnyEvent::HTTP::ACTIVE |
1285 | =item $AnyEvent::HTTP::ACTIVE |
1073 | |
1286 | |
1074 | The number of active connections. This is not the number of currently |
1287 | The number of active connections. This is not the number of currently |
1075 | running requests, but the number of currently open and non-idle TCP |
1288 | running requests, but the number of currently open and non-idle TCP |
1076 | connections. This number of can be useful for load-leveling. |
1289 | connections. This number can be useful for load-leveling. |
1077 | |
1290 | |
1078 | =back |
1291 | =back |
1079 | |
1292 | |
1080 | =cut |
1293 | =cut |
1081 | |
1294 | |
… | |
… | |
1124 | undef |
1337 | undef |
1125 | } |
1338 | } |
1126 | |
1339 | |
1127 | sub set_proxy($) { |
1340 | sub set_proxy($) { |
1128 | if (length $_[0]) { |
1341 | if (length $_[0]) { |
1129 | $_[0] =~ m%^(https?):// ([^:/]+) (?: : (\d*) )?%ix |
1342 | $_[0] =~ m%^(http):// ([^:/]+) (?: : (\d*) )?%ix |
1130 | or Carp::croak "$_[0]: invalid proxy URL"; |
1343 | or Carp::croak "$_[0]: invalid proxy URL"; |
1131 | $PROXY = [$2, $3 || 3128, $1] |
1344 | $PROXY = [$2, $3 || 3128, $1] |
1132 | } else { |
1345 | } else { |
1133 | undef $PROXY; |
1346 | undef $PROXY; |
1134 | } |
1347 | } |
… | |
… | |
1137 | # initialise proxy from environment |
1350 | # initialise proxy from environment |
1138 | eval { |
1351 | eval { |
1139 | set_proxy $ENV{http_proxy}; |
1352 | set_proxy $ENV{http_proxy}; |
1140 | }; |
1353 | }; |
1141 | |
1354 | |
|
|
1355 | =head2 SHOWCASE |
|
|
1356 | |
|
|
1357 | This section contaisn some more elaborate "real-world" examples or code |
|
|
1358 | snippets. |
|
|
1359 | |
|
|
1360 | =head2 HTTP/1.1 FILE DOWNLOAD |
|
|
1361 | |
|
|
1362 | Downloading files with HTTP can be quite tricky, especially when something |
|
|
1363 | goes wrong and you want to resume. |
|
|
1364 | |
|
|
1365 | Here is a function that initiates and resumes a download. It uses the |
|
|
1366 | last modified time to check for file content changes, and works with many |
|
|
1367 | HTTP/1.0 servers as well, and usually falls back to a complete re-download |
|
|
1368 | on older servers. |
|
|
1369 | |
|
|
1370 | It calls the completion callback with either C<undef>, which means a |
|
|
1371 | nonretryable error occured, C<0> when the download was partial and should |
|
|
1372 | be retried, and C<1> if it was successful. |
|
|
1373 | |
|
|
1374 | use AnyEvent::HTTP; |
|
|
1375 | |
|
|
1376 | sub download($$$) { |
|
|
1377 | my ($url, $file, $cb) = @_; |
|
|
1378 | |
|
|
1379 | open my $fh, "+<", $file |
|
|
1380 | or die "$file: $!"; |
|
|
1381 | |
|
|
1382 | my %hdr; |
|
|
1383 | my $ofs = 0; |
|
|
1384 | |
|
|
1385 | warn stat $fh; |
|
|
1386 | warn -s _; |
|
|
1387 | if (stat $fh and -s _) { |
|
|
1388 | $ofs = -s _; |
|
|
1389 | warn "-s is ", $ofs; |
|
|
1390 | $hdr{"if-unmodified-since"} = AnyEvent::HTTP::format_date +(stat _)[9]; |
|
|
1391 | $hdr{"range"} = "bytes=$ofs-"; |
|
|
1392 | } |
|
|
1393 | |
|
|
1394 | http_get $url, |
|
|
1395 | headers => \%hdr, |
|
|
1396 | on_header => sub { |
|
|
1397 | my ($hdr) = @_; |
|
|
1398 | |
|
|
1399 | if ($hdr->{Status} == 200 && $ofs) { |
|
|
1400 | # resume failed |
|
|
1401 | truncate $fh, $ofs = 0; |
|
|
1402 | } |
|
|
1403 | |
|
|
1404 | sysseek $fh, $ofs, 0; |
|
|
1405 | |
|
|
1406 | 1 |
|
|
1407 | }, |
|
|
1408 | on_body => sub { |
|
|
1409 | my ($data, $hdr) = @_; |
|
|
1410 | |
|
|
1411 | if ($hdr->{Status} =~ /^2/) { |
|
|
1412 | length $data == syswrite $fh, $data |
|
|
1413 | or return; # abort on write errors |
|
|
1414 | } |
|
|
1415 | |
|
|
1416 | 1 |
|
|
1417 | }, |
|
|
1418 | sub { |
|
|
1419 | my (undef, $hdr) = @_; |
|
|
1420 | |
|
|
1421 | my $status = $hdr->{Status}; |
|
|
1422 | |
|
|
1423 | if (my $time = AnyEvent::HTTP::parse_date $hdr->{"last-modified"}) { |
|
|
1424 | utime $fh, $time, $time; |
|
|
1425 | } |
|
|
1426 | |
|
|
1427 | if ($status == 200 || $status == 206 || $status == 416) { |
|
|
1428 | # download ok || resume ok || file already fully downloaded |
|
|
1429 | $cb->(1, $hdr); |
|
|
1430 | |
|
|
1431 | } elsif ($status == 412) { |
|
|
1432 | # file has changed while resuming, delete and retry |
|
|
1433 | unlink $file; |
|
|
1434 | $cb->(0, $hdr); |
|
|
1435 | |
|
|
1436 | } elsif ($status == 500 or $status == 503 or $status =~ /^59/) { |
|
|
1437 | # retry later |
|
|
1438 | $cb->(0, $hdr); |
|
|
1439 | |
|
|
1440 | } else { |
|
|
1441 | $cb->(undef, $hdr); |
|
|
1442 | } |
|
|
1443 | } |
|
|
1444 | ; |
|
|
1445 | } |
|
|
1446 | |
|
|
1447 | download "http://server/somelargefile", "/tmp/somelargefile", sub { |
|
|
1448 | if ($_[0]) { |
|
|
1449 | print "OK!\n"; |
|
|
1450 | } elsif (defined $_[0]) { |
|
|
1451 | print "please retry later\n"; |
|
|
1452 | } else { |
|
|
1453 | print "ERROR\n"; |
|
|
1454 | } |
|
|
1455 | }; |
|
|
1456 | |
1142 | =head2 SOCKS PROXIES |
1457 | =head3 SOCKS PROXIES |
1143 | |
1458 | |
1144 | Socks proxies are not directly supported by AnyEvent::HTTP. You can |
1459 | Socks proxies are not directly supported by AnyEvent::HTTP. You can |
1145 | compile your perl to support socks, or use an external program such as |
1460 | compile your perl to support socks, or use an external program such as |
1146 | F<socksify> (dante) or F<tsocks> to make your program use a socks proxy |
1461 | F<socksify> (dante) or F<tsocks> to make your program use a socks proxy |
1147 | transparently. |
1462 | transparently. |