| … | |
… | |
| 36 | |
36 | |
| 37 | =cut |
37 | =cut |
| 38 | |
38 | |
| 39 | package AnyEvent::HTTP; |
39 | package AnyEvent::HTTP; |
| 40 | |
40 | |
| 41 | use strict; |
41 | use common::sense; |
| 42 | no warnings; |
|
|
| 43 | |
42 | |
| 44 | use Errno (); |
43 | use Errno (); |
| 45 | |
44 | |
| 46 | use AnyEvent 5.0 (); |
45 | use AnyEvent 5.0 (); |
| 47 | use AnyEvent::Util (); |
46 | use AnyEvent::Util (); |
| … | |
… | |
| 58 | our $MAX_PERSISTENT = 8; |
57 | our $MAX_PERSISTENT = 8; |
| 59 | our $PERSISTENT_TIMEOUT = 2; |
58 | our $PERSISTENT_TIMEOUT = 2; |
| 60 | our $TIMEOUT = 300; |
59 | our $TIMEOUT = 300; |
| 61 | |
60 | |
| 62 | # changing these is evil |
61 | # changing these is evil |
| 63 | our $MAX_PERSISTENT_PER_HOST = 0; |
62 | our $MAX_PERSISTENT_PER_HOST = 2; |
| 64 | our $MAX_PER_HOST = 4; |
63 | our $MAX_PER_HOST = 4; |
| 65 | |
64 | |
| 66 | our $PROXY; |
65 | our $PROXY; |
| 67 | our $ACTIVE = 0; |
66 | our $ACTIVE = 0; |
| 68 | |
67 | |
| … | |
… | |
| 122 | |
121 | |
| 123 | If the server sends a header multiple times, then their contents will be |
122 | If the server sends a header multiple times, then their contents will be |
| 124 | joined together with a comma (C<,>), as per the HTTP spec. |
123 | joined together with a comma (C<,>), as per the HTTP spec. |
| 125 | |
124 | |
| 126 | If an internal error occurs, such as not being able to resolve a hostname, |
125 | If an internal error occurs, such as not being able to resolve a hostname, |
| 127 | then C<$data> will be C<undef>, C<< $headers->{Status} >> will be C<59x> |
126 | then C<$data> will be C<undef>, C<< $headers->{Status} >> will be |
| 128 | (usually C<599>) and the C<Reason> pseudo-header will contain an error |
127 | C<590>-C<599> and the C<Reason> pseudo-header will contain an error |
| 129 | message. |
128 | message. Currently the following status codes are used: |
|
|
129 | |
|
|
130 | =over 4 |
|
|
131 | |
|
|
132 | =item 595 - errors during connection etsbalishment, proxy handshake. |
|
|
133 | |
|
|
134 | =item 596 - errors during TLS negotiation, request sending and header processing. |
|
|
135 | |
|
|
136 | =item 597 - errors during body receiving or processing. |
|
|
137 | |
|
|
138 | =item 598 - user aborted request via C<on_header> or C<on_body>. |
|
|
139 | |
|
|
140 | =item 599 - other, usually nonretryable, errors (garbled URL etc.). |
|
|
141 | |
|
|
142 | =back |
| 130 | |
143 | |
| 131 | A typical callback might look like this: |
144 | A typical callback might look like this: |
| 132 | |
145 | |
| 133 | sub { |
146 | sub { |
| 134 | my ($body, $hdr) = @_; |
147 | my ($body, $hdr) = @_; |
| … | |
… | |
| 182 | =item cookie_jar => $hash_ref |
195 | =item cookie_jar => $hash_ref |
| 183 | |
196 | |
| 184 | Passing this parameter enables (simplified) cookie-processing, loosely |
197 | Passing this parameter enables (simplified) cookie-processing, loosely |
| 185 | based on the original netscape specification. |
198 | based on the original netscape specification. |
| 186 | |
199 | |
| 187 | The C<$hash_ref> must be an (initially empty) hash reference which will |
200 | The C<$hash_ref> must be an (initially empty) hash reference which |
| 188 | get updated automatically. It is possible to save the cookie jar to |
201 | will get updated automatically. It is possible to save the cookie jar |
| 189 | persistent storage with something like JSON or Storable, but this is not |
202 | to persistent storage with something like JSON or Storable - see the |
| 190 | recommended, as session-only cookies might survive longer than expected. |
203 | C<AnyEvent::HTTP::cookie_jar_expire> function if you wish to remove |
|
|
204 | expired or session-only cookies, and also for documentation on the format |
|
|
205 | of the cookie jar. |
| 191 | |
206 | |
| 192 | Note that this cookie implementation is not meant to be complete. If |
207 | Note that this cookie implementation is not meant to be complete. If |
| 193 | you want complete cookie management you have to do that on your |
208 | you want complete cookie management you have to do that on your |
| 194 | own. C<cookie_jar> is meant as a quick fix to get some cookie-using sites |
209 | own. C<cookie_jar> is meant as a quick fix to get most cookie-using sites |
| 195 | working. Cookies are a privacy disaster, do not use them unless required |
210 | working. Cookies are a privacy disaster, do not use them unless required |
| 196 | to. |
211 | to. |
| 197 | |
212 | |
| 198 | When cookie processing is enabled, the C<Cookie:> and C<Set-Cookie:> |
213 | When cookie processing is enabled, the C<Cookie:> and C<Set-Cookie:> |
| 199 | headers will be set and handled by this module, otherwise they will be |
214 | headers will be set and handled by this module, otherwise they will be |
| … | |
… | |
| 364 | push @{ $CO_SLOT{$_[0]}[1] }, $_[1]; |
379 | push @{ $CO_SLOT{$_[0]}[1] }, $_[1]; |
| 365 | |
380 | |
| 366 | _slot_schedule $_[0]; |
381 | _slot_schedule $_[0]; |
| 367 | } |
382 | } |
| 368 | |
383 | |
|
|
384 | ############################################################################# |
|
|
385 | |
|
|
386 | # expire cookies |
|
|
387 | sub cookie_jar_expire($;$) { |
|
|
388 | my ($jar, $session_end) = @_; |
|
|
389 | |
|
|
390 | %$jar = () if $jar->{version} != 1; |
|
|
391 | |
|
|
392 | my $anow = AE::now; |
|
|
393 | |
|
|
394 | while (my ($chost, $paths) = each %$jar) { |
|
|
395 | next unless ref $paths; |
|
|
396 | |
|
|
397 | while (my ($cpath, $cookies) = each %$paths) { |
|
|
398 | while (my ($cookie, $kv) = each %$cookies) { |
|
|
399 | if (exists $kv->{_expires}) { |
|
|
400 | delete $cookies->{$cookie} |
|
|
401 | if $anow > $kv->{_expires}; |
|
|
402 | } elsif ($session_end) { |
|
|
403 | delete $cookies->{$cookie}; |
|
|
404 | } |
|
|
405 | } |
|
|
406 | |
|
|
407 | delete $paths->{$cpath} |
|
|
408 | unless %$cookies; |
|
|
409 | } |
|
|
410 | |
|
|
411 | delete $jar->{$chost} |
|
|
412 | unless %$paths; |
|
|
413 | } |
|
|
414 | } |
|
|
415 | |
|
|
416 | # extract cookies from jar |
|
|
417 | sub cookie_jar_extract($$$$) { |
|
|
418 | my ($jar, $uscheme, $uhost, $upath) = @_; |
|
|
419 | |
|
|
420 | %$jar = () if $jar->{version} != 1; |
|
|
421 | |
|
|
422 | my @cookies; |
|
|
423 | |
|
|
424 | while (my ($chost, $paths) = each %$jar) { |
|
|
425 | next unless ref $paths; |
|
|
426 | |
|
|
427 | if ($chost =~ /^\./) { |
|
|
428 | next unless $chost eq substr $uhost, -length $chost; |
|
|
429 | } elsif ($chost =~ /\./) { |
|
|
430 | next unless $chost eq $uhost; |
|
|
431 | } else { |
|
|
432 | next; |
|
|
433 | } |
|
|
434 | |
|
|
435 | while (my ($cpath, $cookies) = each %$paths) { |
|
|
436 | next unless $cpath eq substr $upath, 0, length $cpath; |
|
|
437 | |
|
|
438 | while (my ($cookie, $kv) = each %$cookies) { |
|
|
439 | next if $uscheme ne "https" && exists $kv->{secure}; |
|
|
440 | |
|
|
441 | if (exists $kv->{_expires} and AE::now > $kv->{_expires}) { |
|
|
442 | delete $cookies->{$cookie}; |
|
|
443 | next; |
|
|
444 | } |
|
|
445 | |
|
|
446 | my $value = $kv->{value}; |
|
|
447 | |
|
|
448 | if ($value =~ /[=;,[:space:]]/) { |
|
|
449 | $value =~ s/([\\"])/\\$1/g; |
|
|
450 | $value = "\"$value\""; |
|
|
451 | } |
|
|
452 | |
|
|
453 | push @cookies, "$cookie=$value"; |
|
|
454 | } |
|
|
455 | } |
|
|
456 | } |
|
|
457 | |
|
|
458 | \@cookies |
|
|
459 | } |
|
|
460 | |
|
|
461 | # parse set_cookie header into jar |
|
|
462 | sub cookie_jar_set_cookie($$$$) { |
|
|
463 | my ($jar, $set_cookie, $uhost, $date) = @_; |
|
|
464 | |
|
|
465 | my $anow = int AE::now; |
|
|
466 | my $snow; # server-now |
|
|
467 | |
|
|
468 | for ($set_cookie) { |
|
|
469 | # parse NAME=VALUE |
|
|
470 | my @kv; |
|
|
471 | |
|
|
472 | # expires is not http-compliant in the original cookie-spec, |
|
|
473 | # we support the official date format and some extensions |
|
|
474 | while ( |
|
|
475 | m{ |
|
|
476 | \G\s* |
|
|
477 | (?: |
|
|
478 | expires \s*=\s* ([A-Z][a-z][a-z]+,\ [^,;]+) |
|
|
479 | | ([^=;,[:space:]]+) (?: \s*=\s* (?: "((?:[^\\"]+|\\.)*)" | ([^=;,[:space:]]*) ) )? |
|
|
480 | ) |
|
|
481 | }gcxsi |
|
|
482 | ) { |
|
|
483 | my $name = $2; |
|
|
484 | my $value = $4; |
|
|
485 | |
|
|
486 | if (defined $1) { |
|
|
487 | # expires |
|
|
488 | $name = "expires"; |
|
|
489 | $value = $1; |
|
|
490 | } elsif (defined $3) { |
|
|
491 | # quoted |
|
|
492 | $value = $3; |
|
|
493 | $value =~ s/\\(.)/$1/gs; |
|
|
494 | } |
|
|
495 | |
|
|
496 | push @kv, lc $name, $value; |
|
|
497 | |
|
|
498 | last unless /\G\s*;/gc; |
|
|
499 | } |
|
|
500 | |
|
|
501 | last unless @kv; |
|
|
502 | |
|
|
503 | my $name = shift @kv; |
|
|
504 | my %kv = (value => shift @kv, @kv); |
|
|
505 | |
|
|
506 | if (exists $kv{"max-age"}) { |
|
|
507 | $kv{_expires} = $anow + delete $kv{"max-age"}; |
|
|
508 | } elsif (exists $kv{expires}) { |
|
|
509 | $snow ||= parse_date ($date) || $anow; |
|
|
510 | $kv{_expires} = $anow + (parse_date (delete $kv{expires}) - $snow); |
|
|
511 | } else { |
|
|
512 | delete $kv{_expires}; |
|
|
513 | } |
|
|
514 | |
|
|
515 | my $cdom; |
|
|
516 | my $cpath = (delete $kv{path}) || "/"; |
|
|
517 | |
|
|
518 | if (exists $kv{domain}) { |
|
|
519 | $cdom = delete $kv{domain}; |
|
|
520 | |
|
|
521 | $cdom =~ s/^\.?/./; # make sure it starts with a "." |
|
|
522 | |
|
|
523 | next if $cdom =~ /\.$/; |
|
|
524 | |
|
|
525 | # this is not rfc-like and not netscape-like. go figure. |
|
|
526 | my $ndots = $cdom =~ y/.//; |
|
|
527 | next if $ndots < ($cdom =~ /\.[^.][^.]\.[^.][^.]$/ ? 3 : 2); |
|
|
528 | } else { |
|
|
529 | $cdom = $uhost; |
|
|
530 | } |
|
|
531 | |
|
|
532 | # store it |
|
|
533 | $jar->{version} = 1; |
|
|
534 | $jar->{lc $cdom}{$cpath}{$name} = \%kv; |
|
|
535 | |
|
|
536 | redo if /\G\s*,/gc; |
|
|
537 | } |
|
|
538 | } |
|
|
539 | |
| 369 | # continue to parse $_ for headers and place them into the arg |
540 | # continue to parse $_ for headers and place them into the arg |
| 370 | sub parse_hdr() { |
541 | sub parse_hdr() { |
| 371 | my %hdr; |
542 | my %hdr; |
| 372 | |
543 | |
| 373 | # things seen, not parsed: |
544 | # things seen, not parsed: |
| … | |
… | |
| 435 | : return $cb->(undef, { @pseudo, Status => 599, Reason => "Only http and https URL schemes supported" }); |
606 | : return $cb->(undef, { @pseudo, Status => 599, Reason => "Only http and https URL schemes supported" }); |
| 436 | |
607 | |
| 437 | $uauthority =~ /^(?: .*\@ )? ([^\@:]+) (?: : (\d+) )?$/x |
608 | $uauthority =~ /^(?: .*\@ )? ([^\@:]+) (?: : (\d+) )?$/x |
| 438 | or return $cb->(undef, { @pseudo, Status => 599, Reason => "Unparsable URL" }); |
609 | or return $cb->(undef, { @pseudo, Status => 599, Reason => "Unparsable URL" }); |
| 439 | |
610 | |
| 440 | my $uhost = $1; |
611 | my $uhost = lc $1; |
| 441 | $uport = $2 if defined $2; |
612 | $uport = $2 if defined $2; |
| 442 | |
613 | |
| 443 | $hdr{host} = defined $2 ? "$uhost:$2" : "$uhost" |
614 | $hdr{host} = defined $2 ? "$uhost:$2" : "$uhost" |
| 444 | unless exists $hdr{host}; |
615 | unless exists $hdr{host}; |
| 445 | |
616 | |
| … | |
… | |
| 448 | |
619 | |
| 449 | $upath =~ s%^/?%/%; |
620 | $upath =~ s%^/?%/%; |
| 450 | |
621 | |
| 451 | # cookie processing |
622 | # cookie processing |
| 452 | if (my $jar = $arg{cookie_jar}) { |
623 | if (my $jar = $arg{cookie_jar}) { |
| 453 | %$jar = () if $jar->{version} != 1; |
624 | my $cookies = cookie_jar_extract $jar, $uscheme, $uhost, $upath; |
| 454 | |
|
|
| 455 | my @cookie; |
|
|
| 456 | |
|
|
| 457 | while (my ($chost, $paths) = each %$jar) { |
|
|
| 458 | if ($chost =~ /^\./) { |
|
|
| 459 | next unless $chost eq substr $uhost, -length $chost; |
|
|
| 460 | } elsif ($chost =~ /\./) { |
|
|
| 461 | next unless $chost eq $uhost; |
|
|
| 462 | } else { |
|
|
| 463 | next; |
|
|
| 464 | } |
|
|
| 465 | |
625 | |
| 466 | while (my ($cpath, $cookies) = each %$paths) { |
|
|
| 467 | next unless $cpath eq substr $upath, 0, length $cpath; |
|
|
| 468 | |
|
|
| 469 | while (my ($cookie, $kv) = each %$cookies) { |
|
|
| 470 | next if $uscheme ne "https" && exists $kv->{secure}; |
|
|
| 471 | |
|
|
| 472 | if (exists $kv->{expires}) { |
|
|
| 473 | if (AE::now > parse_date ($kv->{expires})) { |
|
|
| 474 | delete $cookies->{$cookie}; |
|
|
| 475 | next; |
|
|
| 476 | } |
|
|
| 477 | } |
|
|
| 478 | |
|
|
| 479 | my $value = $kv->{value}; |
|
|
| 480 | $value =~ s/([\\"])/\\$1/g; |
|
|
| 481 | push @cookie, "$cookie=\"$value\""; |
|
|
| 482 | } |
|
|
| 483 | } |
|
|
| 484 | } |
|
|
| 485 | |
|
|
| 486 | $hdr{cookie} = join "; ", @cookie |
626 | $hdr{cookie} = join "; ", @$cookies |
| 487 | if @cookie; |
627 | if @$cookies; |
| 488 | } |
628 | } |
| 489 | |
629 | |
| 490 | my ($rhost, $rport, $rscheme, $rpath); # request host, port, path |
630 | my ($rhost, $rport, $rscheme, $rpath); # request host, port, path |
| 491 | |
631 | |
| 492 | if ($proxy) { |
632 | if ($proxy) { |
| … | |
… | |
| 495 | $rscheme = "http" unless defined $rscheme; |
635 | $rscheme = "http" unless defined $rscheme; |
| 496 | |
636 | |
| 497 | # don't support https requests over https-proxy transport, |
637 | # don't support https requests over https-proxy transport, |
| 498 | # can't be done with tls as spec'ed, unless you double-encrypt. |
638 | # can't be done with tls as spec'ed, unless you double-encrypt. |
| 499 | $rscheme = "http" if $uscheme eq "https" && $rscheme eq "https"; |
639 | $rscheme = "http" if $uscheme eq "https" && $rscheme eq "https"; |
|
|
640 | |
|
|
641 | $rhost = lc $rhost; |
|
|
642 | $rscheme = lc $rscheme; |
| 500 | } else { |
643 | } else { |
| 501 | ($rhost, $rport, $rscheme, $rpath) = ($uhost, $uport, $uscheme, $upath); |
644 | ($rhost, $rport, $rscheme, $rpath) = ($uhost, $uport, $uscheme, $upath); |
| 502 | } |
645 | } |
| 503 | |
646 | |
| 504 | # leave out fragment and query string, just a heuristic |
647 | # leave out fragment and query string, just a heuristic |
| … | |
… | |
| 506 | $hdr{"user-agent"} = $USERAGENT unless exists $hdr{"user-agent"}; |
649 | $hdr{"user-agent"} = $USERAGENT unless exists $hdr{"user-agent"}; |
| 507 | |
650 | |
| 508 | $hdr{"content-length"} = length $arg{body} |
651 | $hdr{"content-length"} = length $arg{body} |
| 509 | if length $arg{body} || $method ne "GET"; |
652 | if length $arg{body} || $method ne "GET"; |
| 510 | |
653 | |
| 511 | $hdr{connection} = "close TE"; #1.1 |
654 | my $idempotent = $method =~ /^(?:GET|HEAD|PUT|DELETE|OPTIONS|TRACE)$/; |
|
|
655 | |
|
|
656 | # default value for keepalive is true iff the request is for an idempotent method |
|
|
657 | my $keepalive = exists $arg{keepalive} |
|
|
658 | ? $arg{keepalive}*1 |
|
|
659 | : $idempotent ? $PERSISTENT_TIMEOUT : 0; |
|
|
660 | |
|
|
661 | $hdr{connection} = ($keepalive ? "" : "close ") . "Te"; #1.1 |
| 512 | $hdr{te} = "trailers" unless exists $hdr{te}; #1.1 |
662 | $hdr{te} = "trailers" unless exists $hdr{te}; #1.1 |
| 513 | |
663 | |
| 514 | my %state = (connect_guard => 1); |
664 | my %state = (connect_guard => 1); |
| 515 | |
665 | |
| 516 | _get_slot $uhost, sub { |
666 | my $ae_error = 595; # connecting |
| 517 | $state{slot_guard} = shift; |
|
|
| 518 | |
667 | |
|
|
668 | # handle actual, non-tunneled, request |
|
|
669 | my $handle_actual_request = sub { |
|
|
670 | $ae_error = 596; # request phase |
|
|
671 | |
|
|
672 | $state{handle}->starttls ("connect") if $uscheme eq "https" && !exists $state{handle}{tls}; |
|
|
673 | |
|
|
674 | # send request |
|
|
675 | $state{handle}->push_write ( |
|
|
676 | "$method $rpath HTTP/1.1\015\012" |
|
|
677 | . (join "", map "\u$_: $hdr{$_}\015\012", grep defined $hdr{$_}, keys %hdr) |
|
|
678 | . "\015\012" |
|
|
679 | . (delete $arg{body}) |
|
|
680 | ); |
|
|
681 | |
|
|
682 | # return if error occured during push_write() |
| 519 | return unless $state{connect_guard}; |
683 | return unless %state; |
| 520 | |
684 | |
| 521 | my $connect_cb = sub { |
685 | # reduce memory usage, save a kitten, also re-use it for the response headers. |
| 522 | $state{fh} = shift |
686 | %hdr = (); |
|
|
687 | |
|
|
688 | # status line and headers |
|
|
689 | $state{read_response} = sub { |
|
|
690 | for ("$_[1]") { |
|
|
691 | y/\015//d; # weed out any \015, as they show up in the weirdest of places. |
|
|
692 | |
|
|
693 | /^HTTP\/0*([0-9\.]+) \s+ ([0-9]{3}) (?: \s+ ([^\012]*) )? \012/gxci |
|
|
694 | or return (%state = (), $cb->(undef, { @pseudo, Status => 599, Reason => "Invalid server response" })); |
|
|
695 | |
|
|
696 | # 100 Continue handling |
|
|
697 | # should not happen as we don't send expect: 100-continue, |
|
|
698 | # but we handle it just in case. |
|
|
699 | # since we send the request body regardless, if we get an error |
|
|
700 | # we are out of-sync, which we currently do NOT handle correctly. |
|
|
701 | return $state{handle}->push_read (line => $qr_nlnl, $state{read_response}) |
|
|
702 | if $2 eq 100; |
|
|
703 | |
|
|
704 | push @pseudo, |
|
|
705 | HTTPVersion => $1, |
|
|
706 | Status => $2, |
|
|
707 | Reason => $3, |
| 523 | or do { |
708 | ; |
| 524 | my $err = "$!"; |
709 | |
|
|
710 | my $hdr = parse_hdr |
|
|
711 | or return (%state = (), $cb->(undef, { @pseudo, Status => 599, Reason => "Garbled response headers" })); |
|
|
712 | |
|
|
713 | %hdr = (%$hdr, @pseudo); |
|
|
714 | } |
|
|
715 | |
|
|
716 | # redirect handling |
|
|
717 | # microsoft and other shitheads don't give a shit for following standards, |
|
|
718 | # try to support some common forms of broken Location headers. |
|
|
719 | if ($hdr{location} !~ /^(?: $ | [^:\/?\#]+ : )/x) { |
|
|
720 | $hdr{location} =~ s/^\.\/+//; |
|
|
721 | |
|
|
722 | my $url = "$rscheme://$uhost:$uport"; |
|
|
723 | |
|
|
724 | unless ($hdr{location} =~ s/^\///) { |
|
|
725 | $url .= $upath; |
|
|
726 | $url =~ s/\/[^\/]*$//; |
|
|
727 | } |
|
|
728 | |
|
|
729 | $hdr{location} = "$url/$hdr{location}"; |
|
|
730 | } |
|
|
731 | |
|
|
732 | my $redirect; |
|
|
733 | |
|
|
734 | if ($recurse) { |
|
|
735 | my $status = $hdr{Status}; |
|
|
736 | |
|
|
737 | # industry standard is to redirect POST as GET for |
|
|
738 | # 301, 302 and 303, in contrast to HTTP/1.0 and 1.1. |
|
|
739 | # also, the UA should ask the user for 301 and 307 and POST, |
|
|
740 | # industry standard seems to be to simply follow. |
|
|
741 | # we go with the industry standard. |
|
|
742 | if ($status == 301 or $status == 302 or $status == 303) { |
|
|
743 | # HTTP/1.1 is unclear on how to mutate the method |
|
|
744 | $method = "GET" unless $method eq "HEAD"; |
|
|
745 | $redirect = 1; |
|
|
746 | } elsif ($status == 307) { |
|
|
747 | $redirect = 1; |
|
|
748 | } |
|
|
749 | } |
|
|
750 | |
|
|
751 | my $finish = sub { # ($data, $err_status, $err_reason[, $keepalive]) |
|
|
752 | my $may_keep_alive = $_[3]; |
|
|
753 | |
|
|
754 | $state{handle}->destroy if $state{handle}; |
| 525 | %state = (); |
755 | %state = (); |
| 526 | return $cb->(undef, { @pseudo, Status => 599, Reason => $err }); |
756 | |
|
|
757 | if (defined $_[1]) { |
|
|
758 | $hdr{OrigStatus} = $hdr{Status}; $hdr{Status} = $_[1]; |
|
|
759 | $hdr{OrigReason} = $hdr{Reason}; $hdr{Reason} = $_[2]; |
|
|
760 | } |
|
|
761 | |
|
|
762 | # set-cookie processing |
|
|
763 | if ($arg{cookie_jar}) { |
|
|
764 | cookie_jar_set_cookie $arg{cookie_jar}, $hdr{"set-cookie"}, $uhost, $hdr{date}; |
|
|
765 | } |
|
|
766 | |
|
|
767 | if ($redirect && exists $hdr{location}) { |
|
|
768 | # we ignore any errors, as it is very common to receive |
|
|
769 | # Content-Length != 0 but no actual body |
|
|
770 | # we also access %hdr, as $_[1] might be an erro |
|
|
771 | http_request ( |
|
|
772 | $method => $hdr{location}, |
|
|
773 | %arg, |
|
|
774 | recurse => $recurse - 1, |
|
|
775 | Redirect => [$_[0], \%hdr], |
|
|
776 | $cb); |
|
|
777 | } else { |
|
|
778 | $cb->($_[0], \%hdr); |
|
|
779 | } |
|
|
780 | }; |
|
|
781 | |
|
|
782 | $ae_error = 597; # body phase |
|
|
783 | |
|
|
784 | my $len = $hdr{"content-length"}; |
|
|
785 | |
|
|
786 | # body handling, many different code paths |
|
|
787 | # - no body expected |
|
|
788 | # - want_body_handle |
|
|
789 | # - te chunked |
|
|
790 | # - 2x length known (with or without on_body) |
|
|
791 | # - 2x length not known (with or without on_body) |
|
|
792 | if (!$redirect && $arg{on_header} && !$arg{on_header}(\%hdr)) { |
|
|
793 | $finish->(undef, 598 => "Request cancelled by on_header"); |
|
|
794 | } elsif ( |
|
|
795 | $hdr{Status} =~ /^(?:1..|204|205|304)$/ |
|
|
796 | or $method eq "HEAD" |
|
|
797 | or (defined $len && $len == 0) # == 0, not !, because "0 " is true |
|
|
798 | ) { |
|
|
799 | # no body |
|
|
800 | $finish->("", undef, undef, 1); |
|
|
801 | |
|
|
802 | } elsif (!$redirect && $arg{want_body_handle}) { |
|
|
803 | $_[0]->on_eof (undef); |
|
|
804 | $_[0]->on_error (undef); |
|
|
805 | $_[0]->on_read (undef); |
|
|
806 | |
|
|
807 | $finish->(delete $state{handle}); |
|
|
808 | |
|
|
809 | } elsif ($hdr{"transfer-encoding"} =~ /\bchunked\b/i) { |
|
|
810 | my $cl = 0; |
|
|
811 | my $body = undef; |
|
|
812 | my $on_body = $arg{on_body} || sub { $body .= shift; 1 }; |
|
|
813 | |
|
|
814 | $state{read_chunk} = sub { |
|
|
815 | $_[1] =~ /^([0-9a-fA-F]+)/ |
|
|
816 | or $finish->(undef, $ae_error => "Garbled chunked transfer encoding"); |
|
|
817 | |
|
|
818 | my $len = hex $1; |
|
|
819 | |
|
|
820 | if ($len) { |
|
|
821 | $cl += $len; |
|
|
822 | |
|
|
823 | $_[0]->push_read (chunk => $len, sub { |
|
|
824 | $on_body->($_[1], \%hdr) |
|
|
825 | or return $finish->(undef, 598 => "Request cancelled by on_body"); |
|
|
826 | |
|
|
827 | $_[0]->push_read (line => sub { |
|
|
828 | length $_[1] |
|
|
829 | and return $finish->(undef, $ae_error => "Garbled chunked transfer encoding"); |
|
|
830 | $_[0]->push_read (line => $state{read_chunk}); |
|
|
831 | }); |
|
|
832 | }); |
|
|
833 | } else { |
|
|
834 | $hdr{"content-length"} ||= $cl; |
|
|
835 | |
|
|
836 | $_[0]->push_read (line => $qr_nlnl, sub { |
|
|
837 | if (length $_[1]) { |
|
|
838 | for ("$_[1]") { |
|
|
839 | y/\015//d; # weed out any \015, as they show up in the weirdest of places. |
|
|
840 | |
|
|
841 | my $hdr = parse_hdr |
|
|
842 | or return $finish->(undef, $ae_error => "Garbled response trailers"); |
|
|
843 | |
|
|
844 | %hdr = (%hdr, %$hdr); |
|
|
845 | } |
|
|
846 | } |
|
|
847 | |
|
|
848 | $finish->($body, undef, undef, 1); |
|
|
849 | }); |
|
|
850 | } |
| 527 | }; |
851 | }; |
| 528 | |
852 | |
| 529 | pop; # free memory, save a tree |
853 | $_[0]->push_read (line => $state{read_chunk}); |
| 530 | |
854 | |
|
|
855 | } elsif ($arg{on_body}) { |
|
|
856 | if (defined $len) { |
|
|
857 | $_[0]->on_read (sub { |
|
|
858 | $len -= length $_[0]{rbuf}; |
|
|
859 | |
|
|
860 | $arg{on_body}(delete $_[0]{rbuf}, \%hdr) |
|
|
861 | or return $finish->(undef, 598 => "Request cancelled by on_body"); |
|
|
862 | |
|
|
863 | $len > 0 |
|
|
864 | or $finish->("", undef, undef, 1); |
|
|
865 | }); |
|
|
866 | } else { |
|
|
867 | $_[0]->on_eof (sub { |
|
|
868 | $finish->(""); |
|
|
869 | }); |
|
|
870 | $_[0]->on_read (sub { |
|
|
871 | $arg{on_body}(delete $_[0]{rbuf}, \%hdr) |
|
|
872 | or $finish->(undef, 598 => "Request cancelled by on_body"); |
|
|
873 | }); |
|
|
874 | } |
|
|
875 | } else { |
|
|
876 | $_[0]->on_eof (undef); |
|
|
877 | |
|
|
878 | if (defined $len) { |
|
|
879 | $_[0]->on_read (sub { |
|
|
880 | $finish->((substr delete $_[0]{rbuf}, 0, $len, ""), undef, undef, 1) |
|
|
881 | if $len <= length $_[0]{rbuf}; |
|
|
882 | }); |
|
|
883 | } else { |
|
|
884 | $_[0]->on_error (sub { |
|
|
885 | ($! == Errno::EPIPE || !$!) |
|
|
886 | ? $finish->(delete $_[0]{rbuf}) |
|
|
887 | : $finish->(undef, $ae_error => $_[2]); |
|
|
888 | }); |
|
|
889 | $_[0]->on_read (sub { }); |
|
|
890 | } |
|
|
891 | } |
|
|
892 | }; |
|
|
893 | |
|
|
894 | $state{handle}->push_read (line => $qr_nlnl, $state{read_response}); |
|
|
895 | }; |
|
|
896 | |
|
|
897 | my $connect_cb = sub { |
|
|
898 | $state{fh} = shift |
|
|
899 | or do { |
|
|
900 | my $err = "$!"; |
|
|
901 | %state = (); |
|
|
902 | return $cb->(undef, { @pseudo, Status => $ae_error, Reason => $err }); |
|
|
903 | }; |
|
|
904 | |
| 531 | return unless delete $state{connect_guard}; |
905 | return unless delete $state{connect_guard}; |
| 532 | |
906 | |
| 533 | # get handle |
907 | # get handle |
| 534 | $state{handle} = new AnyEvent::Handle |
908 | $state{handle} = new AnyEvent::Handle |
| 535 | fh => $state{fh}, |
909 | fh => $state{fh}, |
| 536 | peername => $rhost, |
910 | peername => $rhost, |
| 537 | tls_ctx => $arg{tls_ctx}, |
911 | tls_ctx => $arg{tls_ctx}, |
| 538 | # these need to be reconfigured on keepalive handles |
912 | # these need to be reconfigured on keepalive handles |
| 539 | timeout => $timeout, |
913 | timeout => $timeout, |
| 540 | on_error => sub { |
914 | on_error => sub { |
| 541 | %state = (); |
915 | %state = (); |
| 542 | $cb->(undef, { @pseudo, Status => 599, Reason => $_[2] }); |
916 | $cb->(undef, { @pseudo, Status => $ae_error, Reason => $_[2] }); |
| 543 | }, |
917 | }, |
| 544 | on_eof => sub { |
918 | on_eof => sub { |
| 545 | %state = (); |
919 | %state = (); |
| 546 | $cb->(undef, { @pseudo, Status => 599, Reason => "Unexpected end-of-file" }); |
920 | $cb->(undef, { @pseudo, Status => $ae_error, Reason => "Unexpected end-of-file" }); |
| 547 | }, |
921 | }, |
| 548 | ; |
922 | ; |
| 549 | |
923 | |
| 550 | # limit the number of persistent connections |
924 | # limit the number of persistent connections |
| 551 | # keepalive not yet supported |
925 | # keepalive not yet supported |
| 552 | # if ($KA_COUNT{$_[1]} < $MAX_PERSISTENT_PER_HOST) { |
926 | # if ($KA_COUNT{$_[1]} < $MAX_PERSISTENT_PER_HOST) { |
| 553 | # ++$KA_COUNT{$_[1]}; |
927 | # ++$KA_COUNT{$_[1]}; |
| 554 | # $state{handle}{ka_count_guard} = AnyEvent::Util::guard { |
928 | # $state{handle}{ka_count_guard} = AnyEvent::Util::guard { |
| 555 | # --$KA_COUNT{$_[1]} |
929 | # --$KA_COUNT{$_[1]} |
| 556 | # }; |
930 | # }; |
| 557 | # $hdr{connection} = "keep-alive"; |
931 | # $hdr{connection} = "keep-alive"; |
| 558 | # } |
932 | # } |
| 559 | |
933 | |
| 560 | $state{handle}->starttls ("connect") if $rscheme eq "https"; |
934 | $state{handle}->starttls ("connect") if $rscheme eq "https"; |
| 561 | |
935 | |
| 562 | # handle actual, non-tunneled, request |
|
|
| 563 | my $handle_actual_request = sub { |
|
|
| 564 | $state{handle}->starttls ("connect") if $uscheme eq "https" && !exists $state{handle}{tls}; |
|
|
| 565 | |
|
|
| 566 | # send request |
|
|
| 567 | $state{handle}->push_write ( |
|
|
| 568 | "$method $rpath HTTP/1.1\015\012" |
|
|
| 569 | . (join "", map "\u$_: $hdr{$_}\015\012", grep defined $hdr{$_}, keys %hdr) |
|
|
| 570 | . "\015\012" |
|
|
| 571 | . (delete $arg{body}) |
|
|
| 572 | ); |
|
|
| 573 | |
|
|
| 574 | # return if error occured during push_write() |
|
|
| 575 | return unless %state; |
|
|
| 576 | |
|
|
| 577 | %hdr = (); # reduce memory usage, save a kitten, also make it possible to re-use |
|
|
| 578 | |
|
|
| 579 | # status line and headers |
|
|
| 580 | $state{read_response} = sub { |
|
|
| 581 | for ("$_[1]") { |
|
|
| 582 | y/\015//d; # weed out any \015, as they show up in the weirdest of places. |
|
|
| 583 | |
|
|
| 584 | /^HTTP\/([0-9\.]+) \s+ ([0-9]{3}) (?: \s+ ([^\012]*) )? \012/igxc |
|
|
| 585 | or return (%state = (), $cb->(undef, { @pseudo, Status => 599, Reason => "Invalid server response" })); |
|
|
| 586 | |
|
|
| 587 | # 100 Continue handling |
|
|
| 588 | # should not happen as we don't send expect: 100-continue, |
|
|
| 589 | # but we handle it just in case. |
|
|
| 590 | # since we send the request body regardless, if we get an error |
|
|
| 591 | # we are out of-sync, which we currently do NOT handle correctly. |
|
|
| 592 | return $state{handle}->push_read (line => $qr_nlnl, $state{read_response}) |
|
|
| 593 | if $2 eq 100; |
|
|
| 594 | |
|
|
| 595 | push @pseudo, |
|
|
| 596 | HTTPVersion => $1, |
|
|
| 597 | Status => $2, |
|
|
| 598 | Reason => $3, |
|
|
| 599 | ; |
|
|
| 600 | |
|
|
| 601 | my $hdr = parse_hdr |
|
|
| 602 | or return (%state = (), $cb->(undef, { @pseudo, Status => 599, Reason => "Garbled response headers" })); |
|
|
| 603 | |
|
|
| 604 | %hdr = (%$hdr, @pseudo); |
|
|
| 605 | } |
|
|
| 606 | |
|
|
| 607 | # redirect handling |
|
|
| 608 | # microsoft and other shitheads don't give a shit for following standards, |
|
|
| 609 | # try to support some common forms of broken Location headers. |
|
|
| 610 | if ($hdr{location} !~ /^(?: $ | [^:\/?\#]+ : )/x) { |
|
|
| 611 | $hdr{location} =~ s/^\.\/+//; |
|
|
| 612 | |
|
|
| 613 | my $url = "$rscheme://$uhost:$uport"; |
|
|
| 614 | |
|
|
| 615 | unless ($hdr{location} =~ s/^\///) { |
|
|
| 616 | $url .= $upath; |
|
|
| 617 | $url =~ s/\/[^\/]*$//; |
|
|
| 618 | } |
|
|
| 619 | |
|
|
| 620 | $hdr{location} = "$url/$hdr{location}"; |
|
|
| 621 | } |
|
|
| 622 | |
|
|
| 623 | my $redirect; |
|
|
| 624 | |
|
|
| 625 | if ($recurse) { |
|
|
| 626 | my $status = $hdr{Status}; |
|
|
| 627 | |
|
|
| 628 | # industry standard is to redirect POST as GET for |
|
|
| 629 | # 301, 302 and 303, in contrast to http/1.0 and 1.1. |
|
|
| 630 | # also, the UA should ask the user for 301 and 307 and POST, |
|
|
| 631 | # industry standard seems to be to simply follow. |
|
|
| 632 | # we go with the industry standard. |
|
|
| 633 | if ($status == 301 or $status == 302 or $status == 303) { |
|
|
| 634 | # HTTP/1.1 is unclear on how to mutate the method |
|
|
| 635 | $method = "GET" unless $method eq "HEAD"; |
|
|
| 636 | $redirect = 1; |
|
|
| 637 | } elsif ($status == 307) { |
|
|
| 638 | $redirect = 1; |
|
|
| 639 | } |
|
|
| 640 | } |
|
|
| 641 | |
|
|
| 642 | my $finish = sub { # ($data, $err_status, $err_reason[, $keepalive]) |
|
|
| 643 | my $keepalive = pop; |
|
|
| 644 | |
|
|
| 645 | $state{handle}->destroy if $state{handle}; |
|
|
| 646 | %state = (); |
|
|
| 647 | |
|
|
| 648 | if (defined $_[1]) { |
|
|
| 649 | $hdr{OrigStatus} = $hdr{Status}; $hdr{Status} = $_[1]; |
|
|
| 650 | $hdr{OrigReason} = $hdr{Reason}; $hdr{Reason} = $_[2]; |
|
|
| 651 | } |
|
|
| 652 | |
|
|
| 653 | # set-cookie processing |
|
|
| 654 | if ($arg{cookie_jar}) { |
|
|
| 655 | for ($hdr{"set-cookie"}) { |
|
|
| 656 | # parse NAME=VALUE |
|
|
| 657 | my @kv; |
|
|
| 658 | |
|
|
| 659 | while ( |
|
|
| 660 | m{ |
|
|
| 661 | \G\s* |
|
|
| 662 | (?: |
|
|
| 663 | expires \s*=\s* ([A-Z][a-z][a-z],\ [^,;]+) |
|
|
| 664 | | ([^=;,[:space:]]+) \s*=\s* (?: "((?:[^\\"]+|\\.)*)" | ([^=;,[:space:]]*) ) |
|
|
| 665 | ) |
|
|
| 666 | }gcxsi |
|
|
| 667 | ) { |
|
|
| 668 | my $name = $2; |
|
|
| 669 | my $value = $4; |
|
|
| 670 | |
|
|
| 671 | unless (defined $name) { |
|
|
| 672 | # expires |
|
|
| 673 | $name = "expires"; |
|
|
| 674 | $value = $1; |
|
|
| 675 | } elsif (!defined $value) { |
|
|
| 676 | # quoted |
|
|
| 677 | $value = $3; |
|
|
| 678 | $value =~ s/\\(.)/$1/gs; |
|
|
| 679 | } |
|
|
| 680 | |
|
|
| 681 | push @kv, lc $name, $value; |
|
|
| 682 | |
|
|
| 683 | last unless /\G\s*;/gc; |
|
|
| 684 | } |
|
|
| 685 | |
|
|
| 686 | last unless @kv; |
|
|
| 687 | |
|
|
| 688 | my $name = shift @kv; |
|
|
| 689 | my %kv = (value => shift @kv, @kv); |
|
|
| 690 | |
|
|
| 691 | $kv{expires} ||= format_date (AE::now + $kv{"max-age"}) |
|
|
| 692 | if exists $kv{"max-age"}; |
|
|
| 693 | |
|
|
| 694 | my $cdom; |
|
|
| 695 | my $cpath = (delete $kv{path}) || "/"; |
|
|
| 696 | |
|
|
| 697 | if (exists $kv{domain}) { |
|
|
| 698 | $cdom = delete $kv{domain}; |
|
|
| 699 | |
|
|
| 700 | $cdom =~ s/^\.?/./; # make sure it starts with a "." |
|
|
| 701 | |
|
|
| 702 | next if $cdom =~ /\.$/; |
|
|
| 703 | |
|
|
| 704 | # this is not rfc-like and not netscape-like. go figure. |
|
|
| 705 | my $ndots = $cdom =~ y/.//; |
|
|
| 706 | next if $ndots < ($cdom =~ /\.[^.][^.]\.[^.][^.]$/ ? 3 : 2); |
|
|
| 707 | } else { |
|
|
| 708 | $cdom = $uhost; |
|
|
| 709 | } |
|
|
| 710 | |
|
|
| 711 | # store it |
|
|
| 712 | $arg{cookie_jar}{version} = 1; |
|
|
| 713 | $arg{cookie_jar}{$cdom}{$cpath}{$name} = \%kv; |
|
|
| 714 | |
|
|
| 715 | redo if /\G\s*,/gc; |
|
|
| 716 | } |
|
|
| 717 | } |
|
|
| 718 | |
|
|
| 719 | if ($redirect && exists $hdr{location}) { |
|
|
| 720 | # we ignore any errors, as it is very common to receive |
|
|
| 721 | # Content-Length != 0 but no actual body |
|
|
| 722 | # we also access %hdr, as $_[1] might be an erro |
|
|
| 723 | http_request ( |
|
|
| 724 | $method => $hdr{location}, |
|
|
| 725 | %arg, |
|
|
| 726 | recurse => $recurse - 1, |
|
|
| 727 | Redirect => [$_[0], \%hdr], |
|
|
| 728 | $cb); |
|
|
| 729 | } else { |
|
|
| 730 | $cb->($_[0], \%hdr); |
|
|
| 731 | } |
|
|
| 732 | }; |
|
|
| 733 | |
|
|
| 734 | my $len = $hdr{"content-length"}; |
|
|
| 735 | |
|
|
| 736 | if (!$redirect && $arg{on_header} && !$arg{on_header}(\%hdr)) { |
|
|
| 737 | $finish->(undef, 598 => "Request cancelled by on_header"); |
|
|
| 738 | } elsif ( |
|
|
| 739 | $hdr{Status} =~ /^(?:1..|204|205|304)$/ |
|
|
| 740 | or $method eq "HEAD" |
|
|
| 741 | or (defined $len && !$len) |
|
|
| 742 | ) { |
|
|
| 743 | # no body |
|
|
| 744 | $finish->("", undef, undef, 1); |
|
|
| 745 | } else { |
|
|
| 746 | # body handling, many different code paths |
|
|
| 747 | # - no body expected |
|
|
| 748 | # - want_body_handle |
|
|
| 749 | # - te chunked |
|
|
| 750 | # - 2x length known (with or without on_body) |
|
|
| 751 | # - 2x length not known (with or without on_body) |
|
|
| 752 | if (!$redirect && $arg{want_body_handle}) { |
|
|
| 753 | $_[0]->on_eof (undef); |
|
|
| 754 | $_[0]->on_error (undef); |
|
|
| 755 | $_[0]->on_read (undef); |
|
|
| 756 | |
|
|
| 757 | $finish->(delete $state{handle}); |
|
|
| 758 | |
|
|
| 759 | } elsif ($hdr{"transfer-encoding"} =~ /\bchunked\b/i) { |
|
|
| 760 | my $cl = 0; |
|
|
| 761 | my $body = undef; |
|
|
| 762 | my $on_body = $arg{on_body} || sub { $body .= shift; 1 }; |
|
|
| 763 | |
|
|
| 764 | $_[0]->on_error (sub { $finish->(undef, 599 => $_[2]) }); |
|
|
| 765 | |
|
|
| 766 | my $read_chunk; $read_chunk = sub { |
|
|
| 767 | $_[1] =~ /^([0-9a-fA-F]+)/ |
|
|
| 768 | or $finish->(undef, 599 => "Garbled chunked transfer encoding"); |
|
|
| 769 | |
|
|
| 770 | my $len = hex $1; |
|
|
| 771 | |
|
|
| 772 | if ($len) { |
|
|
| 773 | $cl += $len; |
|
|
| 774 | |
|
|
| 775 | $_[0]->push_read (chunk => $len, sub { |
|
|
| 776 | $on_body->($_[1], \%hdr) |
|
|
| 777 | or return $finish->(undef, 598 => "Request cancelled by on_body"); |
|
|
| 778 | |
|
|
| 779 | $_[0]->push_read (line => sub { |
|
|
| 780 | length $_[1] |
|
|
| 781 | and return $finish->(undef, 599 => "Garbled chunked transfer encoding"); |
|
|
| 782 | $_[0]->push_read (line => $read_chunk); |
|
|
| 783 | }); |
|
|
| 784 | }); |
|
|
| 785 | } else { |
|
|
| 786 | $hdr{"content-length"} ||= $cl; |
|
|
| 787 | |
|
|
| 788 | $_[0]->push_read (line => $qr_nlnl, sub { |
|
|
| 789 | if (length $_[1]) { |
|
|
| 790 | for ("$_[1]") { |
|
|
| 791 | y/\015//d; # weed out any \015, as they show up in the weirdest of places. |
|
|
| 792 | |
|
|
| 793 | my $hdr = parse_hdr |
|
|
| 794 | or return $finish->(undef, 599 => "Garbled response trailers"); |
|
|
| 795 | |
|
|
| 796 | %hdr = (%hdr, %$hdr); |
|
|
| 797 | } |
|
|
| 798 | } |
|
|
| 799 | |
|
|
| 800 | $finish->($body, undef, undef, 1); |
|
|
| 801 | }); |
|
|
| 802 | } |
|
|
| 803 | }; |
|
|
| 804 | |
|
|
| 805 | $_[0]->push_read (line => $read_chunk); |
|
|
| 806 | |
|
|
| 807 | } elsif ($arg{on_body}) { |
|
|
| 808 | $_[0]->on_error (sub { $finish->(undef, 599 => $_[2]) }); |
|
|
| 809 | |
|
|
| 810 | if ($len) { |
|
|
| 811 | $_[0]->on_read (sub { |
|
|
| 812 | $len -= length $_[0]{rbuf}; |
|
|
| 813 | |
|
|
| 814 | $arg{on_body}(delete $_[0]{rbuf}, \%hdr) |
|
|
| 815 | or return $finish->(undef, 598 => "Request cancelled by on_body"); |
|
|
| 816 | |
|
|
| 817 | $len > 0 |
|
|
| 818 | or $finish->("", undef, undef, 1); |
|
|
| 819 | }); |
|
|
| 820 | } else { |
|
|
| 821 | $_[0]->on_eof (sub { |
|
|
| 822 | $finish->(""); |
|
|
| 823 | }); |
|
|
| 824 | $_[0]->on_read (sub { |
|
|
| 825 | $arg{on_body}(delete $_[0]{rbuf}, \%hdr) |
|
|
| 826 | or $finish->(undef, 598 => "Request cancelled by on_body"); |
|
|
| 827 | }); |
|
|
| 828 | } |
|
|
| 829 | } else { |
|
|
| 830 | $_[0]->on_eof (undef); |
|
|
| 831 | |
|
|
| 832 | if ($len) { |
|
|
| 833 | $_[0]->on_error (sub { $finish->(undef, 599 => $_[2]) }); |
|
|
| 834 | $_[0]->on_read (sub { |
|
|
| 835 | $finish->((substr delete $_[0]{rbuf}, 0, $len, ""), undef, undef, 1) |
|
|
| 836 | if $len <= length $_[0]{rbuf}; |
|
|
| 837 | }); |
|
|
| 838 | } else { |
|
|
| 839 | $_[0]->on_error (sub { |
|
|
| 840 | ($! == Errno::EPIPE || !$!) |
|
|
| 841 | ? $finish->(delete $_[0]{rbuf}) |
|
|
| 842 | : $finish->(undef, 599 => $_[2]); |
|
|
| 843 | }); |
|
|
| 844 | $_[0]->on_read (sub { }); |
|
|
| 845 | } |
|
|
| 846 | } |
|
|
| 847 | } |
|
|
| 848 | }; |
|
|
| 849 | |
|
|
| 850 | $state{handle}->push_read (line => $qr_nlnl, $state{read_response}); |
|
|
| 851 | }; |
|
|
| 852 | |
|
|
| 853 | # now handle proxy-CONNECT method |
936 | # now handle proxy-CONNECT method |
| 854 | if ($proxy && $uscheme eq "https") { |
937 | if ($proxy && $uscheme eq "https") { |
| 855 | # oh dear, we have to wrap it into a connect request |
938 | # oh dear, we have to wrap it into a connect request |
| 856 | |
939 | |
| 857 | # maybe re-use $uauthority with patched port? |
940 | # maybe re-use $uauthority with patched port? |
| 858 | $state{handle}->push_write ("CONNECT $uhost:$uport HTTP/1.0\015\012Host: $uhost\015\012\015\012"); |
941 | $state{handle}->push_write ("CONNECT $uhost:$uport HTTP/1.0\015\012\015\012"); |
| 859 | $state{handle}->push_read (line => $qr_nlnl, sub { |
942 | $state{handle}->push_read (line => $qr_nlnl, sub { |
| 860 | $_[1] =~ /^HTTP\/([0-9\.]+) \s+ ([0-9]{3}) (?: \s+ ([^\015\012]*) )?/ix |
943 | $_[1] =~ /^HTTP\/([0-9\.]+) \s+ ([0-9]{3}) (?: \s+ ([^\015\012]*) )?/ix |
| 861 | or return (%state = (), $cb->(undef, { @pseudo, Status => 599, Reason => "Invalid proxy connect response ($_[1])" })); |
944 | or return (%state = (), $cb->(undef, { @pseudo, Status => 599, Reason => "Invalid proxy connect response ($_[1])" })); |
| 862 | |
945 | |
| 863 | if ($2 == 200) { |
946 | if ($2 == 200) { |
| 864 | $rpath = $upath; |
947 | $rpath = $upath; |
| 865 | &$handle_actual_request; |
948 | $handle_actual_request->(); |
| 866 | } else { |
949 | } else { |
| 867 | %state = (); |
950 | %state = (); |
| 868 | $cb->(undef, { @pseudo, Status => $2, Reason => $3 }); |
951 | $cb->(undef, { @pseudo, Status => $2, Reason => $3 }); |
| 869 | } |
|
|
| 870 | }); |
952 | } |
| 871 | } else { |
|
|
| 872 | &$handle_actual_request; |
|
|
| 873 | } |
953 | }); |
|
|
954 | } else { |
|
|
955 | $handle_actual_request->(); |
| 874 | }; |
956 | } |
|
|
957 | }; |
|
|
958 | |
|
|
959 | _get_slot $uhost, sub { |
|
|
960 | $state{slot_guard} = shift; |
|
|
961 | |
|
|
962 | return unless $state{connect_guard}; |
| 875 | |
963 | |
| 876 | my $tcp_connect = $arg{tcp_connect} |
964 | my $tcp_connect = $arg{tcp_connect} |
| 877 | || do { require AnyEvent::Socket; \&AnyEvent::Socket::tcp_connect }; |
965 | || do { require AnyEvent::Socket; \&AnyEvent::Socket::tcp_connect }; |
| 878 | |
966 | |
| 879 | $state{connect_guard} = $tcp_connect->($rhost, $rport, $connect_cb, $arg{on_prepare} || sub { $timeout }); |
967 | $state{connect_guard} = $tcp_connect->($rhost, $rport, $connect_cb, $arg{on_prepare} || sub { $timeout }); |
| 880 | |
|
|
| 881 | }; |
968 | }; |
| 882 | |
969 | |
| 883 | defined wantarray && AnyEvent::Util::guard { %state = () } |
970 | defined wantarray && AnyEvent::Util::guard { %state = () } |
| 884 | } |
971 | } |
| 885 | |
972 | |
| … | |
… | |
| 920 | string of the form C<http://host:port> (optionally C<https:...>), croaks |
1007 | string of the form C<http://host:port> (optionally C<https:...>), croaks |
| 921 | otherwise. |
1008 | otherwise. |
| 922 | |
1009 | |
| 923 | To clear an already-set proxy, use C<undef>. |
1010 | To clear an already-set proxy, use C<undef>. |
| 924 | |
1011 | |
|
|
1012 | =item AnyEvent::HTTP::cookie_jar_expire $jar[, $session_end] |
|
|
1013 | |
|
|
1014 | Remove all cookies from the cookie jar that have been expired. If |
|
|
1015 | C<$session_end> is given and true, then additionally remove all session |
|
|
1016 | cookies. |
|
|
1017 | |
|
|
1018 | You should call this function (with a true C<$session_end>) before you |
|
|
1019 | save cookies to disk, and you should call this function after loading them |
|
|
1020 | again. If you have a long-running program you can additonally call this |
|
|
1021 | function from time to time. |
|
|
1022 | |
|
|
1023 | A cookie jar is initially an empty hash-reference that is managed by this |
|
|
1024 | module. It's format is subject to change, but currently it is like this: |
|
|
1025 | |
|
|
1026 | The key C<version> has to contain C<1>, otherwise the hash gets |
|
|
1027 | emptied. All other keys are hostnames or IP addresses pointing to |
|
|
1028 | hash-references. The key for these inner hash references is the |
|
|
1029 | server path for which this cookie is meant, and the values are again |
|
|
1030 | hash-references. The keys of those hash-references is the cookie name, and |
|
|
1031 | the value, you guessed it, is another hash-reference, this time with the |
|
|
1032 | key-value pairs from the cookie, except for C<expires> and C<max-age>, |
|
|
1033 | which have been replaced by a C<_expires> key that contains the cookie |
|
|
1034 | expiry timestamp. |
|
|
1035 | |
|
|
1036 | Here is an example of a cookie jar with a single cookie, so you have a |
|
|
1037 | chance of understanding the above paragraph: |
|
|
1038 | |
|
|
1039 | { |
|
|
1040 | version => 1, |
|
|
1041 | "10.0.0.1" => { |
|
|
1042 | "/" => { |
|
|
1043 | "mythweb_id" => { |
|
|
1044 | _expires => 1293917923, |
|
|
1045 | value => "ooRung9dThee3ooyXooM1Ohm", |
|
|
1046 | }, |
|
|
1047 | }, |
|
|
1048 | }, |
|
|
1049 | } |
|
|
1050 | |
| 925 | =item $date = AnyEvent::HTTP::format_date $timestamp |
1051 | =item $date = AnyEvent::HTTP::format_date $timestamp |
| 926 | |
1052 | |
| 927 | Takes a POSIX timestamp (seconds since the epoch) and formats it as a HTTP |
1053 | Takes a POSIX timestamp (seconds since the epoch) and formats it as a HTTP |
| 928 | Date (RFC 2616). |
1054 | Date (RFC 2616). |
| 929 | |
1055 | |
| 930 | =item $timestamp = AnyEvent::HTTP::parse_date $date |
1056 | =item $timestamp = AnyEvent::HTTP::parse_date $date |
| 931 | |
1057 | |
| 932 | Takes a HTTP Date (RFC 2616) or a Cookie date (netscape cookie spec) and |
1058 | Takes a HTTP Date (RFC 2616) or a Cookie date (netscape cookie spec) or a |
| 933 | returns the corresponding POSIX timestamp, or C<undef> if the date cannot |
1059 | bunch of minor variations of those, and returns the corresponding POSIX |
| 934 | be parsed. |
1060 | timestamp, or C<undef> if the date cannot be parsed. |
| 935 | |
1061 | |
| 936 | =item $AnyEvent::HTTP::MAX_RECURSE |
1062 | =item $AnyEvent::HTTP::MAX_RECURSE |
| 937 | |
1063 | |
| 938 | The default value for the C<recurse> request parameter (default: C<10>). |
1064 | The default value for the C<recurse> request parameter (default: C<10>). |
| 939 | |
1065 | |
| … | |
… | |
| 978 | sub parse_date($) { |
1104 | sub parse_date($) { |
| 979 | my ($date) = @_; |
1105 | my ($date) = @_; |
| 980 | |
1106 | |
| 981 | my ($d, $m, $y, $H, $M, $S); |
1107 | my ($d, $m, $y, $H, $M, $S); |
| 982 | |
1108 | |
| 983 | if ($date =~ /^[A-Z][a-z][a-z], ([0-9][0-9])[\- ]([A-Z][a-z][a-z])[\- ]([0-9][0-9][0-9][0-9]) ([0-9][0-9]):([0-9][0-9]):([0-9][0-9]) GMT$/) { |
1109 | if ($date =~ /^[A-Z][a-z][a-z]+, ([0-9][0-9]?)[\- ]([A-Z][a-z][a-z])[\- ]([0-9][0-9][0-9][0-9]) ([0-9][0-9]?):([0-9][0-9]?):([0-9][0-9]?) GMT$/) { |
| 984 | # RFC 822/1123, required by RFC 2616 (with " ") |
1110 | # RFC 822/1123, required by RFC 2616 (with " ") |
| 985 | # cookie dates (with "-") |
1111 | # cookie dates (with "-") |
| 986 | |
1112 | |
| 987 | ($d, $m, $y, $H, $M, $S) = ($1, $2, $3, $4, $5, $6); |
1113 | ($d, $m, $y, $H, $M, $S) = ($1, $2, $3, $4, $5, $6); |
| 988 | |
1114 | |
| 989 | } elsif ($date =~ /^[A-Z][a-z]+, ([0-9][0-9])-([A-Z][a-z][a-z])-([0-9][0-9]) ([0-9][0-9]):([0-9][0-9]):([0-9][0-9]) GMT$/) { |
1115 | } elsif ($date =~ /^[A-Z][a-z][a-z]+, ([0-9][0-9]?)-([A-Z][a-z][a-z])-([0-9][0-9]) ([0-9][0-9]?):([0-9][0-9]?):([0-9][0-9]?) GMT$/) { |
| 990 | # RFC 850 |
1116 | # RFC 850 |
| 991 | ($d, $m, $y, $H, $M, $S) = ($1, $2, $3 < 69 ? $3 + 2000 : $3 + 1900, $4, $5, $6); |
1117 | ($d, $m, $y, $H, $M, $S) = ($1, $2, $3 < 69 ? $3 + 2000 : $3 + 1900, $4, $5, $6); |
| 992 | |
1118 | |
| 993 | } elsif ($date =~ /^[A-Z][a-z][a-z] ([A-Z][a-z][a-z]) ([0-9 ][0-9]) ([0-9][0-9]):([0-9][0-9]):([0-9][0-9]) ([0-9][0-9][0-9][0-9])$/) { |
1119 | } elsif ($date =~ /^[A-Z][a-z][a-z]+ ([A-Z][a-z][a-z]) ([0-9 ]?[0-9]) ([0-9][0-9]?):([0-9][0-9]?):([0-9][0-9]?) ([0-9][0-9][0-9][0-9])$/) { |
| 994 | # ISO C's asctime |
1120 | # ISO C's asctime |
| 995 | ($d, $m, $y, $H, $M, $S) = ($2, $1, $6, $3, $4, $5); |
1121 | ($d, $m, $y, $H, $M, $S) = ($2, $1, $6, $3, $4, $5); |
| 996 | } |
1122 | } |
| 997 | # other formats fail in the loop below |
1123 | # other formats fail in the loop below |
| 998 | |
1124 | |
