… | |
… | |
41 | use strict; |
41 | use strict; |
42 | no warnings; |
42 | no warnings; |
43 | |
43 | |
44 | use Errno (); |
44 | use Errno (); |
45 | |
45 | |
46 | use AnyEvent 4.452 (); |
46 | use AnyEvent 5.0 (); |
47 | use AnyEvent::Util (); |
47 | use AnyEvent::Util (); |
48 | use AnyEvent::Socket (); |
48 | use AnyEvent::Socket (); |
49 | use AnyEvent::Handle (); |
49 | use AnyEvent::Handle (); |
50 | |
50 | |
51 | use base Exporter::; |
51 | use base Exporter::; |
52 | |
52 | |
53 | our $VERSION = '1.12'; |
53 | our $VERSION = '1.45'; |
54 | |
54 | |
55 | our @EXPORT = qw(http_get http_post http_head http_request); |
55 | our @EXPORT = qw(http_get http_post http_head http_request); |
56 | |
56 | |
57 | our $USERAGENT = "Mozilla/5.0 (compatible; U; AnyEvent-HTTP/$VERSION; +http://software.schmorp.de/pkg/AnyEvent)"; |
57 | our $USERAGENT = "Mozilla/5.0 (compatible; U; AnyEvent-HTTP/$VERSION; +http://software.schmorp.de/pkg/AnyEvent)"; |
58 | our $MAX_RECURSE = 10; |
58 | our $MAX_RECURSE = 10; |
… | |
… | |
99 | The callback will be called with the response body data as first argument |
99 | The callback will be called with the response body data as first argument |
100 | (or C<undef> if an error occured), and a hash-ref with response headers as |
100 | (or C<undef> if an error occured), and a hash-ref with response headers as |
101 | second argument. |
101 | second argument. |
102 | |
102 | |
103 | All the headers in that hash are lowercased. In addition to the response |
103 | All the headers in that hash are lowercased. In addition to the response |
104 | headers, the "pseudo-headers" C<HTTPVersion>, C<Status> and C<Reason> |
104 | headers, the "pseudo-headers" (uppercase to avoid clashing with possible |
|
|
105 | response headers) C<HTTPVersion>, C<Status> and C<Reason> contain the |
105 | contain the three parts of the HTTP Status-Line of the same name. The |
106 | three parts of the HTTP Status-Line of the same name. |
|
|
107 | |
106 | pseudo-header C<URL> contains the original URL (which can differ from the |
108 | The pseudo-header C<URL> contains the actual URL (which can differ from |
107 | requested URL when following redirects). |
109 | the requested URL when following redirects - for example, you might get |
|
|
110 | an error that your URL scheme is not supported even though your URL is a |
|
|
111 | valid http URL because it redirected to an ftp URL, in which case you can |
|
|
112 | look at the URL pseudo header). |
|
|
113 | |
|
|
114 | The pseudo-header C<Redirect> only exists when the request was a result |
|
|
115 | of an internal redirect. In that case it is an array reference with |
|
|
116 | the C<($data, $headers)> from the redirect response. Note that this |
|
|
117 | response could in turn be the result of a redirect itself, and C<< |
|
|
118 | $headers->{Redirect}[1]{Redirect} >> will then contain the original |
|
|
119 | response, and so on. |
108 | |
120 | |
109 | If the server sends a header multiple times, then their contents will be |
121 | If the server sends a header multiple times, then their contents will be |
110 | joined together with a comma (C<,>), as per the HTTP spec. |
122 | joined together with a comma (C<,>), as per the HTTP spec. |
111 | |
123 | |
112 | If an internal error occurs, such as not being able to resolve a hostname, |
124 | If an internal error occurs, such as not being able to resolve a hostname, |
… | |
… | |
138 | |
150 | |
139 | =item headers => hashref |
151 | =item headers => hashref |
140 | |
152 | |
141 | The request headers to use. Currently, C<http_request> may provide its |
153 | The request headers to use. Currently, C<http_request> may provide its |
142 | own C<Host:>, C<Content-Length:>, C<Connection:> and C<Cookie:> headers |
154 | own C<Host:>, C<Content-Length:>, C<Connection:> and C<Cookie:> headers |
143 | and will provide defaults for C<User-Agent:> and C<Referer:>. |
155 | and will provide defaults for C<User-Agent:> and C<Referer:> (this can be |
|
|
156 | suppressed by using C<undef> for these headers in which case they won't be |
|
|
157 | sent at all). |
144 | |
158 | |
145 | =item timeout => $seconds |
159 | =item timeout => $seconds |
146 | |
160 | |
147 | The time-out to use for various stages - each connect attempt will reset |
161 | The time-out to use for various stages - each connect attempt will reset |
148 | the timeout, as will read or write activity. Default timeout is 5 minutes. |
162 | the timeout, as will read or write activity, i.e. this is not an overall |
|
|
163 | timeout. |
|
|
164 | |
|
|
165 | Default timeout is 5 minutes. |
149 | |
166 | |
150 | =item proxy => [$host, $port[, $scheme]] or undef |
167 | =item proxy => [$host, $port[, $scheme]] or undef |
151 | |
168 | |
152 | Use the given http proxy for all requests. If not specified, then the |
169 | Use the given http proxy for all requests. If not specified, then the |
153 | default proxy (as specified by C<$ENV{http_proxy}>) is used. |
170 | default proxy (as specified by C<$ENV{http_proxy}>) is used. |
154 | |
171 | |
155 | C<$scheme> must be either missing or C<http> for HTTP, or C<https> for |
172 | C<$scheme> must be either missing, C<http> for HTTP or C<https> for |
156 | HTTPS. |
173 | HTTPS. |
157 | |
174 | |
158 | =item body => $string |
175 | =item body => $string |
159 | |
176 | |
160 | The request body, usually empty. Will be-sent as-is (future versions of |
177 | The request body, usually empty. Will be-sent as-is (future versions of |
… | |
… | |
186 | verification) TLS context. |
203 | verification) TLS context. |
187 | |
204 | |
188 | The default for this option is C<low>, which could be interpreted as "give |
205 | The default for this option is C<low>, which could be interpreted as "give |
189 | me the page, no matter what". |
206 | me the page, no matter what". |
190 | |
207 | |
|
|
208 | =item on_prepare => $callback->($fh) |
|
|
209 | |
|
|
210 | In rare cases you need to "tune" the socket before it is used to |
|
|
211 | connect (for exmaple, to bind it on a given IP address). This parameter |
|
|
212 | overrides the prepare callback passed to C<AnyEvent::Socket::tcp_connect> |
|
|
213 | and behaves exactly the same way (e.g. it has to provide a |
|
|
214 | timeout). See the description for the C<$prepare_cb> argument of |
|
|
215 | C<AnyEvent::Socket::tcp_connect> for details. |
|
|
216 | |
191 | =item on_header => $callback->($headers) |
217 | =item on_header => $callback->($headers) |
192 | |
218 | |
193 | When specified, this callback will be called with the header hash as soon |
219 | When specified, this callback will be called with the header hash as soon |
194 | as headers have been successfully received from the remote server (not on |
220 | as headers have been successfully received from the remote server (not on |
195 | locally-generated errors). |
221 | locally-generated errors). |
… | |
… | |
221 | This callback is useful when the data is too large to be held in memory |
247 | This callback is useful when the data is too large to be held in memory |
222 | (so the callback writes it to a file) or when only some information should |
248 | (so the callback writes it to a file) or when only some information should |
223 | be extracted, or when the body should be processed incrementally. |
249 | be extracted, or when the body should be processed incrementally. |
224 | |
250 | |
225 | It is usually preferred over doing your own body handling via |
251 | It is usually preferred over doing your own body handling via |
226 | C<want_body_handle>. |
252 | C<want_body_handle>, but in case of streaming APIs, where HTTP is |
|
|
253 | only used to create a connection, C<want_body_handle> is the better |
|
|
254 | alternative, as it allows you to install your own event handler, reducing |
|
|
255 | resource usage. |
227 | |
256 | |
228 | =item want_body_handle => $enable |
257 | =item want_body_handle => $enable |
229 | |
258 | |
230 | When enabled (default is disabled), the behaviour of AnyEvent::HTTP |
259 | When enabled (default is disabled), the behaviour of AnyEvent::HTTP |
231 | changes considerably: after parsing the headers, and instead of |
260 | changes considerably: after parsing the headers, and instead of |
… | |
… | |
243 | This is useful with some push-type services, where, after the initial |
272 | This is useful with some push-type services, where, after the initial |
244 | headers, an interactive protocol is used (typical example would be the |
273 | headers, an interactive protocol is used (typical example would be the |
245 | push-style twitter API which starts a JSON/XML stream). |
274 | push-style twitter API which starts a JSON/XML stream). |
246 | |
275 | |
247 | If you think you need this, first have a look at C<on_body>, to see if |
276 | If you think you need this, first have a look at C<on_body>, to see if |
248 | that doesn'T solve your problem in a better way. |
277 | that doesn't solve your problem in a better way. |
249 | |
278 | |
250 | =back |
279 | =back |
251 | |
280 | |
252 | Example: make a simple HTTP GET request for http://www.nethype.de/ |
281 | Example: make a simple HTTP GET request for http://www.nethype.de/ |
253 | |
282 | |
… | |
… | |
309 | push @{ $CO_SLOT{$_[0]}[1] }, $_[1]; |
338 | push @{ $CO_SLOT{$_[0]}[1] }, $_[1]; |
310 | |
339 | |
311 | _slot_schedule $_[0]; |
340 | _slot_schedule $_[0]; |
312 | } |
341 | } |
313 | |
342 | |
314 | our $qr_nl = qr<\015?\012>; |
343 | our $qr_nl = qr{\015?\012}; |
315 | our $qr_nlnl = qr<\015?\012\015?\012>; |
344 | our $qr_nlnl = qr{(?<![^\012])\015?\012}; |
316 | |
345 | |
317 | our $TLS_CTX_LOW = { cache => 1, sslv2 => 1 }; |
346 | our $TLS_CTX_LOW = { cache => 1, sslv2 => 1 }; |
318 | our $TLS_CTX_HIGH = { cache => 1, verify => 1, verify_peername => "https" }; |
347 | our $TLS_CTX_HIGH = { cache => 1, verify => 1, verify_peername => "https" }; |
319 | |
348 | |
320 | sub http_request($$@) { |
349 | sub http_request($$@) { |
… | |
… | |
332 | while (my ($k, $v) = each %$hdr) { |
361 | while (my ($k, $v) = each %$hdr) { |
333 | $hdr{lc $k} = $v; |
362 | $hdr{lc $k} = $v; |
334 | } |
363 | } |
335 | } |
364 | } |
336 | |
365 | |
|
|
366 | # pseudo headers for all subsequent responses |
|
|
367 | my @pseudo = (URL => $url); |
|
|
368 | push @pseudo, Redirect => delete $arg{Redirect} if exists $arg{Redirect}; |
|
|
369 | |
337 | my $recurse = exists $arg{recurse} ? delete $arg{recurse} : $MAX_RECURSE; |
370 | my $recurse = exists $arg{recurse} ? delete $arg{recurse} : $MAX_RECURSE; |
338 | |
371 | |
339 | return $cb->(undef, { Status => 599, Reason => "Too many redirections", URL => $url }) |
372 | return $cb->(undef, { Status => 599, Reason => "Too many redirections", @pseudo }) |
340 | if $recurse < 0; |
373 | if $recurse < 0; |
341 | |
374 | |
342 | my $proxy = $arg{proxy} || $PROXY; |
375 | my $proxy = $arg{proxy} || $PROXY; |
343 | my $timeout = $arg{timeout} || $TIMEOUT; |
376 | my $timeout = $arg{timeout} || $TIMEOUT; |
344 | |
377 | |
345 | my ($uscheme, $uauthority, $upath, $query, $fragment) = |
378 | my ($uscheme, $uauthority, $upath, $query, $fragment) = |
346 | $url =~ m|(?:([^:/?#]+):)?(?://([^/?#]*))?([^?#]*)(?:\?([^#]*))?(?:#(.*))?|; |
379 | $url =~ m|(?:([^:/?#]+):)?(?://([^/?#]*))?([^?#]*)(?:(\?[^#]*))?(?:#(.*))?|; |
347 | |
380 | |
348 | $uscheme = lc $uscheme; |
381 | $uscheme = lc $uscheme; |
349 | |
382 | |
350 | my $uport = $uscheme eq "http" ? 80 |
383 | my $uport = $uscheme eq "http" ? 80 |
351 | : $uscheme eq "https" ? 443 |
384 | : $uscheme eq "https" ? 443 |
352 | : return $cb->(undef, { Status => 599, Reason => "Only http and https URL schemes supported", URL => $url }); |
385 | : return $cb->(undef, { Status => 599, Reason => "Only http and https URL schemes supported", @pseudo }); |
353 | |
386 | |
354 | $uauthority =~ /^(?: .*\@ )? ([^\@:]+) (?: : (\d+) )?$/x |
387 | $uauthority =~ /^(?: .*\@ )? ([^\@:]+) (?: : (\d+) )?$/x |
355 | or return $cb->(undef, { Status => 599, Reason => "Unparsable URL", URL => $url }); |
388 | or return $cb->(undef, { Status => 599, Reason => "Unparsable URL", @pseudo }); |
356 | |
389 | |
357 | my $uhost = $1; |
390 | my $uhost = $1; |
358 | $uport = $2 if defined $2; |
391 | $uport = $2 if defined $2; |
359 | |
392 | |
360 | $hdr{host} = defined $2 ? "$uhost:$2" : "$uhost"; |
393 | $hdr{host} = defined $2 ? "$uhost:$2" : "$uhost" |
|
|
394 | unless exists $hdr{host}; |
361 | |
395 | |
362 | $uhost =~ s/^\[(.*)\]$/$1/; |
396 | $uhost =~ s/^\[(.*)\]$/$1/; |
363 | $upath .= "?$query" if length $query; |
397 | $upath .= $query if length $query; |
364 | |
398 | |
365 | $upath =~ s%^/?%/%; |
399 | $upath =~ s%^/?%/%; |
366 | |
400 | |
367 | # cookie processing |
401 | # cookie processing |
368 | if (my $jar = $arg{cookie_jar}) { |
402 | if (my $jar = $arg{cookie_jar}) { |
… | |
… | |
398 | my ($rhost, $rport, $rscheme, $rpath); # request host, port, path |
432 | my ($rhost, $rport, $rscheme, $rpath); # request host, port, path |
399 | |
433 | |
400 | if ($proxy) { |
434 | if ($proxy) { |
401 | ($rpath, $rhost, $rport, $rscheme) = ($url, @$proxy); |
435 | ($rpath, $rhost, $rport, $rscheme) = ($url, @$proxy); |
402 | |
436 | |
|
|
437 | $rscheme = "http" unless defined $rscheme; |
|
|
438 | |
403 | # don't support https requests over https-proxy transport, |
439 | # don't support https requests over https-proxy transport, |
404 | # can't be done with tls as spec'ed, unless you double-encrypt. |
440 | # can't be done with tls as spec'ed, unless you double-encrypt. |
405 | $rscheme = "http" if $uscheme eq "https" && $rscheme eq "https"; |
441 | $rscheme = "http" if $uscheme eq "https" && $rscheme eq "https"; |
406 | } else { |
442 | } else { |
407 | ($rhost, $rport, $rscheme, $rpath) = ($uhost, $uport, $uscheme, $upath); |
443 | ($rhost, $rport, $rscheme, $rpath) = ($uhost, $uport, $uscheme, $upath); |
408 | } |
444 | } |
409 | |
445 | |
410 | $hdr{"user-agent"} ||= $USERAGENT; |
446 | # leave out fragment and query string, just a heuristic |
411 | $hdr{referer} ||= "$uscheme://$uauthority$upath"; # leave out fragment and query string, just a heuristic |
447 | $hdr{referer} ||= "$uscheme://$uauthority$upath" unless exists $hdr{referer}; |
|
|
448 | $hdr{"user-agent"} ||= $USERAGENT unless exists $hdr{"user-agent"}; |
412 | |
449 | |
413 | $hdr{"content-length"} = length $arg{body}; |
450 | $hdr{"content-length"} = length $arg{body} |
|
|
451 | if length $arg{body} || $method ne "GET"; |
414 | |
452 | |
415 | my %state = (connect_guard => 1); |
453 | my %state = (connect_guard => 1); |
416 | |
454 | |
417 | _get_slot $uhost, sub { |
455 | _get_slot $uhost, sub { |
418 | $state{slot_guard} = shift; |
456 | $state{slot_guard} = shift; |
419 | |
457 | |
420 | return unless $state{connect_guard}; |
458 | return unless $state{connect_guard}; |
421 | |
459 | |
422 | $state{connect_guard} = AnyEvent::Socket::tcp_connect $rhost, $rport, sub { |
460 | $state{connect_guard} = AnyEvent::Socket::tcp_connect $rhost, $rport, sub { |
423 | $state{fh} = shift |
461 | $state{fh} = shift |
|
|
462 | or do { |
|
|
463 | my $err = "$!"; |
|
|
464 | %state = (); |
424 | or return (%state = (), $cb->(undef, { Status => 599, Reason => "$!", URL => $url })); |
465 | return $cb->(undef, { Status => 599, Reason => $err, @pseudo }); |
|
|
466 | }; |
|
|
467 | |
425 | pop; # free memory, save a tree |
468 | pop; # free memory, save a tree |
426 | |
469 | |
427 | return unless delete $state{connect_guard}; |
470 | return unless delete $state{connect_guard}; |
428 | |
471 | |
429 | # get handle |
472 | # get handle |
430 | $state{handle} = new AnyEvent::Handle |
473 | $state{handle} = new AnyEvent::Handle |
431 | fh => $state{fh}, |
474 | fh => $state{fh}, |
|
|
475 | peername => $rhost, |
|
|
476 | tls_ctx => $arg{tls_ctx}, |
|
|
477 | # these need to be reconfigured on keepalive handles |
432 | timeout => $timeout, |
478 | timeout => $timeout, |
433 | peername => $rhost, |
479 | on_error => sub { |
434 | tls_ctx => $arg{tls_ctx}; |
480 | %state = (); |
|
|
481 | $cb->(undef, { Status => 599, Reason => $_[2], @pseudo }); |
|
|
482 | }, |
|
|
483 | on_eof => sub { |
|
|
484 | %state = (); |
|
|
485 | $cb->(undef, { Status => 599, Reason => "Unexpected end-of-file", @pseudo }); |
|
|
486 | }, |
|
|
487 | ; |
435 | |
488 | |
436 | # limit the number of persistent connections |
489 | # limit the number of persistent connections |
437 | # keepalive not yet supported |
490 | # keepalive not yet supported |
438 | if ($KA_COUNT{$_[1]} < $MAX_PERSISTENT_PER_HOST) { |
491 | # if ($KA_COUNT{$_[1]} < $MAX_PERSISTENT_PER_HOST) { |
439 | ++$KA_COUNT{$_[1]}; |
492 | # ++$KA_COUNT{$_[1]}; |
440 | $state{handle}{ka_count_guard} = AnyEvent::Util::guard { |
493 | # $state{handle}{ka_count_guard} = AnyEvent::Util::guard { |
441 | --$KA_COUNT{$_[1]} |
494 | # --$KA_COUNT{$_[1]} |
442 | }; |
495 | # }; |
443 | $hdr{connection} = "keep-alive"; |
496 | # $hdr{connection} = "keep-alive"; |
444 | } else { |
497 | # } else { |
445 | delete $hdr{connection}; |
498 | delete $hdr{connection}; |
446 | } |
499 | # } |
447 | |
|
|
448 | # (re-)configure handle |
|
|
449 | $state{handle}->on_error (sub { |
|
|
450 | %state = (); |
|
|
451 | $cb->(undef, { Status => 599, Reason => $_[2], URL => $url }); |
|
|
452 | }); |
|
|
453 | $state{handle}->on_eof (sub { |
|
|
454 | %state = (); |
|
|
455 | $cb->(undef, { Status => 599, Reason => "Unexpected end-of-file", URL => $url }); |
|
|
456 | }); |
|
|
457 | |
500 | |
458 | $state{handle}->starttls ("connect") if $rscheme eq "https"; |
501 | $state{handle}->starttls ("connect") if $rscheme eq "https"; |
459 | |
502 | |
460 | # handle actual, non-tunneled, request |
503 | # handle actual, non-tunneled, request |
461 | my $handle_actual_request = sub { |
504 | my $handle_actual_request = sub { |
462 | $state{handle}->starttls ("connect") if $uscheme eq "https" && !exists $state{handle}{tls}; |
505 | $state{handle}->starttls ("connect") if $uscheme eq "https" && !exists $state{handle}{tls}; |
463 | |
506 | |
464 | # send request |
507 | # send request |
465 | $state{handle}->push_write ( |
508 | $state{handle}->push_write ( |
466 | "$method $rpath HTTP/1.0\015\012" |
509 | "$method $rpath HTTP/1.0\015\012" |
467 | . (join "", map "\u$_: $hdr{$_}\015\012", keys %hdr) |
510 | . (join "", map "\u$_: $hdr{$_}\015\012", grep defined $hdr{$_}, keys %hdr) |
468 | . "\015\012" |
511 | . "\015\012" |
469 | . (delete $arg{body}) |
512 | . (delete $arg{body}) |
470 | ); |
513 | ); |
471 | |
514 | |
|
|
515 | # return if error occured during push_write() |
|
|
516 | return unless %state; |
|
|
517 | |
472 | %hdr = (); # reduce memory usage, save a kitten |
518 | %hdr = (); # reduce memory usage, save a kitten, also make it possible to re-use |
473 | |
519 | |
474 | # status line |
520 | # status line |
475 | $state{handle}->push_read (line => $qr_nl, sub { |
521 | $state{handle}->push_read (line => $qr_nl, sub { |
476 | $_[1] =~ /^HTTP\/([0-9\.]+) \s+ ([0-9]{3}) (?: \s+ ([^\015\012]*) )?/ix |
522 | $_[1] =~ /^HTTP\/([0-9\.]+) \s+ ([0-9]{3}) (?: \s+ ([^\015\012]*) )?/ix |
477 | or return (%state = (), $cb->(undef, { Status => 599, Reason => "Invalid server response ($_[1])", URL => $url })); |
523 | or return (%state = (), $cb->(undef, { Status => 599, Reason => "Invalid server response ($_[1])", @pseudo })); |
478 | |
524 | |
479 | my %hdr = ( # response headers |
525 | push @pseudo, |
480 | HTTPVersion => ",$1", |
526 | HTTPVersion => $1, |
481 | Status => ",$2", |
527 | Status => $2, |
482 | Reason => ",$3", |
528 | Reason => $3, |
483 | URL => ",$url" |
|
|
484 | ); |
529 | ; |
485 | |
530 | |
486 | # headers, could be optimized a bit |
531 | # headers, could be optimized a bit |
487 | $state{handle}->unshift_read (line => $qr_nlnl, sub { |
532 | $state{handle}->unshift_read (line => $qr_nlnl, sub { |
488 | for ("$_[1]\012") { |
533 | for ("$_[1]") { |
489 | y/\015//d; # weed out any \015, as they show up in the weirdest of places. |
534 | y/\015//d; # weed out any \015, as they show up in the weirdest of places. |
490 | |
535 | |
491 | # things seen, not parsed: |
536 | # things seen, not parsed: |
492 | # p3pP="NON CUR OTPi OUR NOR UNI" |
537 | # p3pP="NON CUR OTPi OUR NOR UNI" |
493 | |
538 | |
… | |
… | |
498 | ((?: [^\012]+ | \012[\011\040] )*) |
543 | ((?: [^\012]+ | \012[\011\040] )*) |
499 | \012 |
544 | \012 |
500 | /gxc; |
545 | /gxc; |
501 | |
546 | |
502 | /\G$/ |
547 | /\G$/ |
503 | or return (%state = (), $cb->(undef, { Status => 599, Reason => "Garbled response headers", URL => $url })); |
548 | or return (%state = (), $cb->(undef, { Status => 599, Reason => "Garbled response headers", @pseudo })); |
504 | } |
549 | } |
505 | |
550 | |
|
|
551 | # remove the "," prefix we added to all headers above |
506 | substr $_, 0, 1, "" |
552 | substr $_, 0, 1, "" |
507 | for values %hdr; |
553 | for values %hdr; |
|
|
554 | |
|
|
555 | # patch in all pseudo headers |
|
|
556 | %hdr = (%hdr, @pseudo); |
508 | |
557 | |
509 | # redirect handling |
558 | # redirect handling |
510 | # microsoft and other shitheads don't give a shit for following standards, |
559 | # microsoft and other shitheads don't give a shit for following standards, |
511 | # try to support some common forms of broken Location headers. |
560 | # try to support some common forms of broken Location headers. |
512 | if ($hdr{location} !~ /^(?: $ | [^:\/?\#]+ : )/x) { |
561 | if ($hdr{location} !~ /^(?: $ | [^:\/?\#]+ : )/x) { |
… | |
… | |
523 | } |
572 | } |
524 | |
573 | |
525 | my $redirect; |
574 | my $redirect; |
526 | |
575 | |
527 | if ($recurse) { |
576 | if ($recurse) { |
|
|
577 | my $status = $hdr{Status}; |
|
|
578 | |
528 | if ($hdr{Status} =~ /^30[12]$/ && $method ne "POST") { |
579 | if (($status == 301 || $status == 302) && $method ne "POST") { |
529 | # apparently, mozilla et al. just change POST to GET here |
580 | # apparently, mozilla et al. just change POST to GET here |
530 | # more research is needed before we do the same |
581 | # more research is needed before we do the same |
531 | $redirect = 1; |
582 | $redirect = 1; |
532 | } elsif ($hdr{Status} == 303) { |
583 | } elsif ($status == 303) { |
533 | # even http/1.1 is unclear on how to mutate the method |
584 | # even http/1.1 is unclear on how to mutate the method |
534 | $method = "GET" unless $method eq "HEAD"; |
585 | $method = "GET" unless $method eq "HEAD"; |
535 | $redirect = 1; |
586 | $redirect = 1; |
536 | } elsif ($hdr{Status} == 307 && $method =~ /^(?:GET|HEAD)$/) { |
587 | } elsif ($status == 307 && $method =~ /^(?:GET|HEAD)$/) { |
537 | $redirect = 1; |
588 | $redirect = 1; |
538 | } |
589 | } |
539 | } |
590 | } |
540 | |
591 | |
541 | my $finish = sub { |
592 | my $finish = sub { |
… | |
… | |
590 | |
641 | |
591 | redo if /\G\s*,/gc; |
642 | redo if /\G\s*,/gc; |
592 | } |
643 | } |
593 | } |
644 | } |
594 | |
645 | |
595 | if ($redirect) { |
646 | if ($redirect && exists $hdr{location}) { |
596 | # we ignore any errors, as it is very common to receive |
647 | # we ignore any errors, as it is very common to receive |
597 | # Content-Length != 0 but no actual body |
648 | # Content-Length != 0 but no actual body |
598 | # we also access %hdr, as $_[1] might be an erro |
649 | # we also access %hdr, as $_[1] might be an erro |
599 | http_request ($method => $hdr{location}, %arg, recurse => $recurse - 1, $cb); |
650 | http_request ( |
|
|
651 | $method => $hdr{location}, |
|
|
652 | %arg, |
|
|
653 | recurse => $recurse - 1, |
|
|
654 | Redirect => \@_, |
|
|
655 | $cb); |
600 | } else { |
656 | } else { |
601 | $cb->($_[0], $_[1]); |
657 | $cb->($_[0], $_[1]); |
602 | } |
658 | } |
603 | }; |
659 | }; |
604 | |
660 | |
605 | my $len = $hdr{"content-length"}; |
661 | my $len = $hdr{"content-length"}; |
606 | |
662 | |
607 | if (!$redirect && $arg{on_header} && !$arg{on_header}(\%hdr)) { |
663 | if (!$redirect && $arg{on_header} && !$arg{on_header}(\%hdr)) { |
608 | $finish->(undef, { Status => 598, Reason => "Request cancelled by on_header", URL => $url }); |
664 | $finish->(undef, { Status => 598, Reason => "Request cancelled by on_header", @pseudo }); |
609 | } elsif ( |
665 | } elsif ( |
610 | $hdr{Status} =~ /^(?:1..|[23]04)$/ |
666 | $hdr{Status} =~ /^(?:1..|[23]04)$/ |
611 | or $method eq "HEAD" |
667 | or $method eq "HEAD" |
612 | or (defined $len && !$len) |
668 | or (defined $len && !$len) |
613 | ) { |
669 | ) { |
… | |
… | |
623 | $_[0]->on_read (undef); |
679 | $_[0]->on_read (undef); |
624 | |
680 | |
625 | $finish->(delete $state{handle}, \%hdr); |
681 | $finish->(delete $state{handle}, \%hdr); |
626 | |
682 | |
627 | } elsif ($arg{on_body}) { |
683 | } elsif ($arg{on_body}) { |
628 | $_[0]->on_error (sub { $finish->(undef, { Status => 599, Reason => $_[2], URL => $url }) }); |
684 | $_[0]->on_error (sub { $finish->(undef, { Status => 599, Reason => $_[2], @pseudo }) }); |
629 | if ($len) { |
685 | if ($len) { |
630 | $_[0]->on_eof (undef); |
686 | $_[0]->on_eof (undef); |
631 | $_[0]->on_read (sub { |
687 | $_[0]->on_read (sub { |
632 | $len -= length $_[0]{rbuf}; |
688 | $len -= length $_[0]{rbuf}; |
633 | |
689 | |
634 | $arg{on_body}(delete $_[0]{rbuf}, \%hdr) |
690 | $arg{on_body}(delete $_[0]{rbuf}, \%hdr) |
635 | or $finish->(undef, { Status => 598, Reason => "Request cancelled by on_body", URL => $url }); |
691 | or $finish->(undef, { Status => 598, Reason => "Request cancelled by on_body", @pseudo }); |
636 | |
692 | |
637 | $len > 0 |
693 | $len > 0 |
638 | or $finish->("", \%hdr); |
694 | or $finish->("", \%hdr); |
639 | }); |
695 | }); |
640 | } else { |
696 | } else { |
641 | $_[0]->on_eof (sub { |
697 | $_[0]->on_eof (sub { |
642 | $finish->("", \%hdr); |
698 | $finish->("", \%hdr); |
643 | }); |
699 | }); |
644 | $_[0]->on_read (sub { |
700 | $_[0]->on_read (sub { |
645 | $arg{on_body}(delete $_[0]{rbuf}, \%hdr) |
701 | $arg{on_body}(delete $_[0]{rbuf}, \%hdr) |
646 | or $finish->(undef, { Status => 598, Reason => "Request cancelled by on_body", URL => $url }); |
702 | or $finish->(undef, { Status => 598, Reason => "Request cancelled by on_body", @pseudo }); |
647 | }); |
703 | }); |
648 | } |
704 | } |
649 | } else { |
705 | } else { |
650 | $_[0]->on_eof (undef); |
706 | $_[0]->on_eof (undef); |
651 | |
707 | |
652 | if ($len) { |
708 | if ($len) { |
653 | $_[0]->on_error (sub { $finish->(undef, { Status => 599, Reason => $_[2], URL => $url }) }); |
709 | $_[0]->on_error (sub { $finish->(undef, { Status => 599, Reason => $_[2], @pseudo }) }); |
654 | $_[0]->on_read (sub { |
710 | $_[0]->on_read (sub { |
655 | $finish->((substr delete $_[0]{rbuf}, 0, $len, ""), \%hdr) |
711 | $finish->((substr delete $_[0]{rbuf}, 0, $len, ""), \%hdr) |
656 | if $len <= length $_[0]{rbuf}; |
712 | if $len <= length $_[0]{rbuf}; |
657 | }); |
713 | }); |
658 | } else { |
714 | } else { |
659 | $_[0]->on_error (sub { |
715 | $_[0]->on_error (sub { |
660 | $! == Errno::EPIPE || !$! |
716 | $! == Errno::EPIPE || !$! |
661 | ? $finish->(delete $_[0]{rbuf}, \%hdr) |
717 | ? $finish->(delete $_[0]{rbuf}, \%hdr) |
662 | : $finish->(undef, { Status => 599, Reason => $_[2], URL => $url }); |
718 | : $finish->(undef, { Status => 599, Reason => $_[2], @pseudo }); |
663 | }); |
719 | }); |
664 | $_[0]->on_read (sub { }); |
720 | $_[0]->on_read (sub { }); |
665 | } |
721 | } |
666 | } |
722 | } |
667 | } |
723 | } |
… | |
… | |
675 | |
731 | |
676 | # maybe re-use $uauthority with patched port? |
732 | # maybe re-use $uauthority with patched port? |
677 | $state{handle}->push_write ("CONNECT $uhost:$uport HTTP/1.0\015\012Host: $uhost\015\012\015\012"); |
733 | $state{handle}->push_write ("CONNECT $uhost:$uport HTTP/1.0\015\012Host: $uhost\015\012\015\012"); |
678 | $state{handle}->push_read (line => $qr_nlnl, sub { |
734 | $state{handle}->push_read (line => $qr_nlnl, sub { |
679 | $_[1] =~ /^HTTP\/([0-9\.]+) \s+ ([0-9]{3}) (?: \s+ ([^\015\012]*) )?/ix |
735 | $_[1] =~ /^HTTP\/([0-9\.]+) \s+ ([0-9]{3}) (?: \s+ ([^\015\012]*) )?/ix |
680 | or return (%state = (), $cb->(undef, { Status => 599, Reason => "Invalid proxy connect response ($_[1])", URL => $url })); |
736 | or return (%state = (), $cb->(undef, { Status => 599, Reason => "Invalid proxy connect response ($_[1])", @pseudo })); |
681 | |
737 | |
682 | if ($2 == 200) { |
738 | if ($2 == 200) { |
683 | $rpath = $upath; |
739 | $rpath = $upath; |
684 | &$handle_actual_request; |
740 | &$handle_actual_request; |
685 | } else { |
741 | } else { |
686 | %state = (); |
742 | %state = (); |
687 | $cb->(undef, { Status => $2, Reason => $3, URL => $url }); |
743 | $cb->(undef, { Status => $2, Reason => $3, @pseudo }); |
688 | } |
744 | } |
689 | }); |
745 | }); |
690 | } else { |
746 | } else { |
691 | &$handle_actual_request; |
747 | &$handle_actual_request; |
692 | } |
748 | } |
693 | |
749 | |
694 | }, sub { |
750 | }, $arg{on_prepare} || sub { $timeout }; |
695 | $timeout |
|
|
696 | }; |
|
|
697 | }; |
751 | }; |
698 | |
752 | |
699 | defined wantarray && AnyEvent::Util::guard { %state = () } |
753 | defined wantarray && AnyEvent::Util::guard { %state = () } |
700 | } |
754 | } |
701 | |
755 | |
… | |
… | |
715 | &http_request |
769 | &http_request |
716 | } |
770 | } |
717 | |
771 | |
718 | =back |
772 | =back |
719 | |
773 | |
|
|
774 | =head2 DNS CACHING |
|
|
775 | |
|
|
776 | AnyEvent::HTTP uses the AnyEvent::Socket::tcp_connect function for |
|
|
777 | the actual connection, which in turn uses AnyEvent::DNS to resolve |
|
|
778 | hostnames. The latter is a simple stub resolver and does no caching |
|
|
779 | on its own. If you want DNS caching, you currently have to provide |
|
|
780 | your own default resolver (by storing a suitable resolver object in |
|
|
781 | C<$AnyEvent::DNS::RESOLVER>). |
|
|
782 | |
720 | =head2 GLOBAL FUNCTIONS AND VARIABLES |
783 | =head2 GLOBAL FUNCTIONS AND VARIABLES |
721 | |
784 | |
722 | =over 4 |
785 | =over 4 |
723 | |
786 | |
724 | =item AnyEvent::HTTP::set_proxy "proxy-url" |
787 | =item AnyEvent::HTTP::set_proxy "proxy-url" |
725 | |
788 | |
726 | Sets the default proxy server to use. The proxy-url must begin with a |
789 | Sets the default proxy server to use. The proxy-url must begin with a |
727 | string of the form C<http://host:port> (optionally C<https:...>). |
790 | string of the form C<http://host:port> (optionally C<https:...>), croaks |
|
|
791 | otherwise. |
|
|
792 | |
|
|
793 | To clear an already-set proxy, use C<undef>. |
728 | |
794 | |
729 | =item $AnyEvent::HTTP::MAX_RECURSE |
795 | =item $AnyEvent::HTTP::MAX_RECURSE |
730 | |
796 | |
731 | The default value for the C<recurse> request parameter (default: C<10>). |
797 | The default value for the C<recurse> request parameter (default: C<10>). |
732 | |
798 | |
… | |
… | |
735 | The default value for the C<User-Agent> header (the default is |
801 | The default value for the C<User-Agent> header (the default is |
736 | C<Mozilla/5.0 (compatible; U; AnyEvent-HTTP/$VERSION; +http://software.schmorp.de/pkg/AnyEvent)>). |
802 | C<Mozilla/5.0 (compatible; U; AnyEvent-HTTP/$VERSION; +http://software.schmorp.de/pkg/AnyEvent)>). |
737 | |
803 | |
738 | =item $AnyEvent::HTTP::MAX_PER_HOST |
804 | =item $AnyEvent::HTTP::MAX_PER_HOST |
739 | |
805 | |
740 | The maximum number of concurrent conenctions to the same host (identified |
806 | The maximum number of concurrent connections to the same host (identified |
741 | by the hostname). If the limit is exceeded, then the additional requests |
807 | by the hostname). If the limit is exceeded, then the additional requests |
742 | are queued until previous connections are closed. |
808 | are queued until previous connections are closed. |
743 | |
809 | |
744 | The default value for this is C<4>, and it is highly advisable to not |
810 | The default value for this is C<4>, and it is highly advisable to not |
745 | increase it. |
811 | increase it. |
… | |
… | |
753 | =back |
819 | =back |
754 | |
820 | |
755 | =cut |
821 | =cut |
756 | |
822 | |
757 | sub set_proxy($) { |
823 | sub set_proxy($) { |
|
|
824 | if (length $_[0]) { |
758 | $PROXY = [$2, $3 || 3128, $1] if $_[0] =~ m%^(https?):// ([^:/]+) (?: : (\d*) )?%ix; |
825 | $_[0] =~ m%^(https?):// ([^:/]+) (?: : (\d*) )?%ix |
|
|
826 | or Carp::croak "$_[0]: invalid proxy URL"; |
|
|
827 | $PROXY = [$2, $3 || 3128, $1] |
|
|
828 | } else { |
|
|
829 | undef $PROXY; |
|
|
830 | } |
759 | } |
831 | } |
760 | |
832 | |
761 | # initialise proxy from environment |
833 | # initialise proxy from environment |
|
|
834 | eval { |
762 | set_proxy $ENV{http_proxy}; |
835 | set_proxy $ENV{http_proxy}; |
|
|
836 | }; |
763 | |
837 | |
764 | =head1 SEE ALSO |
838 | =head1 SEE ALSO |
765 | |
839 | |
766 | L<AnyEvent>. |
840 | L<AnyEvent>. |
767 | |
841 | |