… | |
… | |
39 | package AnyEvent::HTTP; |
39 | package AnyEvent::HTTP; |
40 | |
40 | |
41 | use strict; |
41 | use strict; |
42 | no warnings; |
42 | no warnings; |
43 | |
43 | |
44 | use Carp; |
44 | use Errno (); |
45 | |
45 | |
46 | use AnyEvent (); |
46 | use AnyEvent 4.8 (); |
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 = '1.41'; |
54 | |
54 | |
55 | our @EXPORT = qw(http_get http_post http_head http_request); |
55 | our @EXPORT = qw(http_get http_post http_head http_request); |
56 | |
56 | |
57 | our $USERAGENT = "Mozilla/5.0 (compatible; AnyEvent::HTTP/$VERSION; +http://software.schmorp.de/pkg/AnyEvent)"; |
57 | our $USERAGENT = "Mozilla/5.0 (compatible; U; AnyEvent-HTTP/$VERSION; +http://software.schmorp.de/pkg/AnyEvent)"; |
58 | our $MAX_RECURSE = 10; |
58 | our $MAX_RECURSE = 10; |
59 | our $MAX_PERSISTENT = 8; |
59 | our $MAX_PERSISTENT = 8; |
60 | our $PERSISTENT_TIMEOUT = 2; |
60 | our $PERSISTENT_TIMEOUT = 2; |
61 | our $TIMEOUT = 300; |
61 | our $TIMEOUT = 300; |
62 | |
62 | |
63 | # changing these is evil |
63 | # changing these is evil |
64 | our $MAX_PERSISTENT_PER_HOST = 2; |
64 | our $MAX_PERSISTENT_PER_HOST = 0; |
65 | our $MAX_PER_HOST = 4; |
65 | our $MAX_PER_HOST = 4; |
66 | |
66 | |
67 | our $PROXY; |
67 | our $PROXY; |
68 | our $ACTIVE = 0; |
68 | our $ACTIVE = 0; |
69 | |
69 | |
… | |
… | |
94 | When called in void context, nothing is returned. In other contexts, |
94 | When called in void context, nothing is returned. In other contexts, |
95 | C<http_request> returns a "cancellation guard" - you have to keep the |
95 | C<http_request> returns a "cancellation guard" - you have to keep the |
96 | object at least alive until the callback get called. If the object gets |
96 | object at least alive until the callback get called. If the object gets |
97 | destroyed before the callbakc is called, the request will be cancelled. |
97 | destroyed before the callbakc is called, the request will be cancelled. |
98 | |
98 | |
99 | The callback will be called with the response data as first argument |
99 | The callback will be called with the response body data as first argument |
100 | (or C<undef> if it wasn't available due to errors), and a hash-ref with |
100 | (or C<undef> if an error occured), and a hash-ref with response headers as |
101 | response headers as second argument. |
101 | second argument. |
102 | |
102 | |
103 | All the headers in that hash are lowercased. In addition to the response |
103 | All the headers in that hash are lowercased. In addition to the response |
104 | headers, the "pseudo-headers" C<HTTPVersion>, C<Status> and C<Reason> |
104 | headers, the "pseudo-headers" C<HTTPVersion>, C<Status> and C<Reason> |
105 | contain the three parts of the HTTP Status-Line of the same name. The |
105 | contain the three parts of the HTTP Status-Line of the same name. The |
106 | pseudo-header C<URL> contains the original URL (which can differ from the |
106 | pseudo-header C<URL> contains the original URL (which can differ from the |
… | |
… | |
108 | |
108 | |
109 | If the server sends a header multiple times, then their contents will be |
109 | 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. |
110 | joined together with a comma (C<,>), as per the HTTP spec. |
111 | |
111 | |
112 | If an internal error occurs, such as not being able to resolve a hostname, |
112 | If an internal error occurs, such as not being able to resolve a hostname, |
113 | then C<$data> will be C<undef>, C<< $headers->{Status} >> will be C<599> |
113 | then C<$data> will be C<undef>, C<< $headers->{Status} >> will be C<59x> |
114 | and the C<Reason> pseudo-header will contain an error message. |
114 | (usually C<599>) and the C<Reason> pseudo-header will contain an error |
|
|
115 | message. |
115 | |
116 | |
116 | A typical callback might look like this: |
117 | A typical callback might look like this: |
117 | |
118 | |
118 | sub { |
119 | sub { |
119 | my ($body, $hdr) = @_; |
120 | my ($body, $hdr) = @_; |
… | |
… | |
165 | based on the original netscape specification. |
166 | based on the original netscape specification. |
166 | |
167 | |
167 | The C<$hash_ref> must be an (initially empty) hash reference which will |
168 | The C<$hash_ref> must be an (initially empty) hash reference which will |
168 | get updated automatically. It is possible to save the cookie_jar to |
169 | get updated automatically. It is possible to save the cookie_jar to |
169 | persistent storage with something like JSON or Storable, but this is not |
170 | persistent storage with something like JSON or Storable, but this is not |
170 | recommended, as expire times are currently being ignored. |
171 | recommended, as expiry times are currently being ignored. |
171 | |
172 | |
172 | Note that this cookie implementation is not of very high quality, nor |
173 | Note that this cookie implementation is not of very high quality, nor |
173 | meant to be complete. If you want complete cookie management you have to |
174 | meant to be complete. If you want complete cookie management you have to |
174 | do that on your own. C<cookie_jar> is meant as a quick fix to get some |
175 | do that on your own. C<cookie_jar> is meant as a quick fix to get some |
175 | cookie-using sites working. Cookies are a privacy disaster, do not use |
176 | cookie-using sites working. Cookies are a privacy disaster, do not use |
176 | them unless required to. |
177 | them unless required to. |
|
|
178 | |
|
|
179 | =item tls_ctx => $scheme | $tls_ctx |
|
|
180 | |
|
|
181 | Specifies the AnyEvent::TLS context to be used for https connections. This |
|
|
182 | parameter follows the same rules as the C<tls_ctx> parameter to |
|
|
183 | L<AnyEvent::Handle>, but additionally, the two strings C<low> or |
|
|
184 | C<high> can be specified, which give you a predefined low-security (no |
|
|
185 | verification, highest compatibility) and high-security (CA and common-name |
|
|
186 | verification) TLS context. |
|
|
187 | |
|
|
188 | The default for this option is C<low>, which could be interpreted as "give |
|
|
189 | me the page, no matter what". |
|
|
190 | |
|
|
191 | =item on_header => $callback->($headers) |
|
|
192 | |
|
|
193 | 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 |
|
|
195 | locally-generated errors). |
|
|
196 | |
|
|
197 | It has to return either true (in which case AnyEvent::HTTP will continue), |
|
|
198 | or false, in which case AnyEvent::HTTP will cancel the download (and call |
|
|
199 | the finish callback with an error code of C<598>). |
|
|
200 | |
|
|
201 | This callback is useful, among other things, to quickly reject unwanted |
|
|
202 | content, which, if it is supposed to be rare, can be faster than first |
|
|
203 | doing a C<HEAD> request. |
|
|
204 | |
|
|
205 | Example: cancel the request unless the content-type is "text/html". |
|
|
206 | |
|
|
207 | on_header => sub { |
|
|
208 | $_[0]{"content-type"} =~ /^text\/html\s*(?:;|$)/ |
|
|
209 | }, |
|
|
210 | |
|
|
211 | =item on_body => $callback->($partial_body, $headers) |
|
|
212 | |
|
|
213 | When specified, all body data will be passed to this callback instead of |
|
|
214 | to the completion callback. The completion callback will get the empty |
|
|
215 | string instead of the body data. |
|
|
216 | |
|
|
217 | It has to return either true (in which case AnyEvent::HTTP will continue), |
|
|
218 | or false, in which case AnyEvent::HTTP will cancel the download (and call |
|
|
219 | the completion callback with an error code of C<598>). |
|
|
220 | |
|
|
221 | 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 |
|
|
223 | be extracted, or when the body should be processed incrementally. |
|
|
224 | |
|
|
225 | It is usually preferred over doing your own body handling via |
|
|
226 | C<want_body_handle>, but in case of streaming APIs, where HTTP is |
|
|
227 | only used to create a connection, C<want_body_handle> is the better |
|
|
228 | alternative, as it allows you to install your own event handler, reducing |
|
|
229 | resource usage. |
|
|
230 | |
|
|
231 | =item want_body_handle => $enable |
|
|
232 | |
|
|
233 | When enabled (default is disabled), the behaviour of AnyEvent::HTTP |
|
|
234 | changes considerably: after parsing the headers, and instead of |
|
|
235 | downloading the body (if any), the completion callback will be |
|
|
236 | called. Instead of the C<$body> argument containing the body data, the |
|
|
237 | callback will receive the L<AnyEvent::Handle> object associated with the |
|
|
238 | connection. In error cases, C<undef> will be passed. When there is no body |
|
|
239 | (e.g. status C<304>), the empty string will be passed. |
|
|
240 | |
|
|
241 | The handle object might or might not be in TLS mode, might be connected to |
|
|
242 | a proxy, be a persistent connection etc., and configured in unspecified |
|
|
243 | ways. The user is responsible for this handle (it will not be used by this |
|
|
244 | module anymore). |
|
|
245 | |
|
|
246 | This is useful with some push-type services, where, after the initial |
|
|
247 | headers, an interactive protocol is used (typical example would be the |
|
|
248 | push-style twitter API which starts a JSON/XML stream). |
|
|
249 | |
|
|
250 | If you think you need this, first have a look at C<on_body>, to see if |
|
|
251 | that doesn't solve your problem in a better way. |
177 | |
252 | |
178 | =back |
253 | =back |
179 | |
254 | |
180 | Example: make a simple HTTP GET request for http://www.nethype.de/ |
255 | Example: make a simple HTTP GET request for http://www.nethype.de/ |
181 | |
256 | |
… | |
… | |
237 | push @{ $CO_SLOT{$_[0]}[1] }, $_[1]; |
312 | push @{ $CO_SLOT{$_[0]}[1] }, $_[1]; |
238 | |
313 | |
239 | _slot_schedule $_[0]; |
314 | _slot_schedule $_[0]; |
240 | } |
315 | } |
241 | |
316 | |
242 | our $qr_nl = qr<\015?\012>; |
317 | our $qr_nl = qr{\015?\012}; |
243 | our $qr_nlnl = qr<\015?\012\015?\012>; |
318 | our $qr_nlnl = qr{(?<![^\012])\015?\012}; |
|
|
319 | |
|
|
320 | our $TLS_CTX_LOW = { cache => 1, sslv2 => 1 }; |
|
|
321 | our $TLS_CTX_HIGH = { cache => 1, verify => 1, verify_peername => "https" }; |
244 | |
322 | |
245 | sub http_request($$@) { |
323 | sub http_request($$@) { |
246 | my $cb = pop; |
324 | my $cb = pop; |
247 | my ($method, $url, %arg) = @_; |
325 | my ($method, $url, %arg) = @_; |
248 | |
326 | |
249 | my %hdr; |
327 | my %hdr; |
|
|
328 | |
|
|
329 | $arg{tls_ctx} = $TLS_CTX_LOW if $arg{tls_ctx} eq "low" || !exists $arg{tls_ctx}; |
|
|
330 | $arg{tls_ctx} = $TLS_CTX_HIGH if $arg{tls_ctx} eq "high"; |
250 | |
331 | |
251 | $method = uc $method; |
332 | $method = uc $method; |
252 | |
333 | |
253 | if (my $hdr = $arg{headers}) { |
334 | if (my $hdr = $arg{headers}) { |
254 | while (my ($k, $v) = each %$hdr) { |
335 | while (my ($k, $v) = each %$hdr) { |
… | |
… | |
256 | } |
337 | } |
257 | } |
338 | } |
258 | |
339 | |
259 | my $recurse = exists $arg{recurse} ? delete $arg{recurse} : $MAX_RECURSE; |
340 | my $recurse = exists $arg{recurse} ? delete $arg{recurse} : $MAX_RECURSE; |
260 | |
341 | |
261 | return $cb->(undef, { Status => 599, Reason => "recursion limit reached", URL => $url }) |
342 | return $cb->(undef, { Status => 599, Reason => "Too many redirections", URL => $url }) |
262 | if $recurse < 0; |
343 | if $recurse < 0; |
263 | |
344 | |
264 | my $proxy = $arg{proxy} || $PROXY; |
345 | my $proxy = $arg{proxy} || $PROXY; |
265 | my $timeout = $arg{timeout} || $TIMEOUT; |
346 | my $timeout = $arg{timeout} || $TIMEOUT; |
266 | |
347 | |
267 | $hdr{"user-agent"} ||= $USERAGENT; |
|
|
268 | |
|
|
269 | my ($uscheme, $uauthority, $upath, $query, $fragment) = |
348 | my ($uscheme, $uauthority, $upath, $query, $fragment) = |
270 | $url =~ m|(?:([^:/?#]+):)?(?://([^/?#]*))?([^?#]*)(?:\?([^#]*))?(?:#(.*))?|; |
349 | $url =~ m|(?:([^:/?#]+):)?(?://([^/?#]*))?([^?#]*)(?:\?([^#]*))?(?:#(.*))?|; |
271 | |
350 | |
272 | $uscheme = lc $uscheme; |
351 | $uscheme = lc $uscheme; |
273 | |
352 | |
274 | my $uport = $uscheme eq "http" ? 80 |
353 | my $uport = $uscheme eq "http" ? 80 |
275 | : $uscheme eq "https" ? 443 |
354 | : $uscheme eq "https" ? 443 |
276 | : return $cb->(undef, { Status => 599, Reason => "only http and https URL schemes supported", URL => $url }); |
355 | : return $cb->(undef, { Status => 599, Reason => "Only http and https URL schemes supported", URL => $url }); |
277 | |
|
|
278 | $hdr{referer} ||= "$uscheme://$uauthority$upath"; # leave out fragment and query string, just a heuristic |
|
|
279 | |
356 | |
280 | $uauthority =~ /^(?: .*\@ )? ([^\@:]+) (?: : (\d+) )?$/x |
357 | $uauthority =~ /^(?: .*\@ )? ([^\@:]+) (?: : (\d+) )?$/x |
281 | or return $cb->(undef, { Status => 599, Reason => "unparsable URL", URL => $url }); |
358 | or return $cb->(undef, { Status => 599, Reason => "Unparsable URL", URL => $url }); |
282 | |
359 | |
283 | my $uhost = $1; |
360 | my $uhost = $1; |
284 | $uport = $2 if defined $2; |
361 | $uport = $2 if defined $2; |
|
|
362 | |
|
|
363 | $hdr{host} = defined $2 ? "$uhost:$2" : "$uhost"; |
285 | |
364 | |
286 | $uhost =~ s/^\[(.*)\]$/$1/; |
365 | $uhost =~ s/^\[(.*)\]$/$1/; |
287 | $upath .= "?$query" if length $query; |
366 | $upath .= "?$query" if length $query; |
288 | |
367 | |
289 | $upath =~ s%^/?%/%; |
368 | $upath =~ s%^/?%/%; |
… | |
… | |
329 | $rscheme = "http" if $uscheme eq "https" && $rscheme eq "https"; |
408 | $rscheme = "http" if $uscheme eq "https" && $rscheme eq "https"; |
330 | } else { |
409 | } else { |
331 | ($rhost, $rport, $rscheme, $rpath) = ($uhost, $uport, $uscheme, $upath); |
410 | ($rhost, $rport, $rscheme, $rpath) = ($uhost, $uport, $uscheme, $upath); |
332 | } |
411 | } |
333 | |
412 | |
334 | $hdr{host} = $uhost; |
413 | $hdr{"user-agent"} ||= $USERAGENT; |
|
|
414 | $hdr{referer} ||= "$uscheme://$uauthority$upath"; # leave out fragment and query string, just a heuristic |
|
|
415 | |
335 | $hdr{"content-length"} = length $arg{body}; |
416 | $hdr{"content-length"} = length $arg{body}; |
336 | |
417 | |
337 | my %state = (connect_guard => 1); |
418 | my %state = (connect_guard => 1); |
338 | |
419 | |
339 | _get_slot $uhost, sub { |
420 | _get_slot $uhost, sub { |
… | |
… | |
341 | |
422 | |
342 | return unless $state{connect_guard}; |
423 | return unless $state{connect_guard}; |
343 | |
424 | |
344 | $state{connect_guard} = AnyEvent::Socket::tcp_connect $rhost, $rport, sub { |
425 | $state{connect_guard} = AnyEvent::Socket::tcp_connect $rhost, $rport, sub { |
345 | $state{fh} = shift |
426 | $state{fh} = shift |
|
|
427 | or do { |
|
|
428 | my $err = "$!"; |
|
|
429 | %state = (); |
346 | or return $cb->(undef, { Status => 599, Reason => "$!", URL => $url }); |
430 | return $cb->(undef, { Status => 599, Reason => $err, URL => $url }); |
|
|
431 | }; |
|
|
432 | |
347 | pop; # free memory, save a tree |
433 | pop; # free memory, save a tree |
348 | |
434 | |
349 | return unless delete $state{connect_guard}; |
435 | return unless delete $state{connect_guard}; |
350 | |
436 | |
351 | # get handle |
437 | # get handle |
352 | $state{handle} = new AnyEvent::Handle |
438 | $state{handle} = new AnyEvent::Handle |
353 | fh => $state{fh}, |
439 | fh => $state{fh}, |
354 | timeout => $timeout; |
440 | timeout => $timeout, |
|
|
441 | peername => $rhost, |
|
|
442 | tls_ctx => $arg{tls_ctx}; |
355 | |
443 | |
356 | # limit the number of persistent connections |
444 | # limit the number of persistent connections |
357 | # keepalive not yet supported |
445 | # keepalive not yet supported |
358 | if ($KA_COUNT{$_[1]} < $MAX_PERSISTENT_PER_HOST) { |
446 | if ($KA_COUNT{$_[1]} < $MAX_PERSISTENT_PER_HOST) { |
359 | ++$KA_COUNT{$_[1]}; |
447 | ++$KA_COUNT{$_[1]}; |
… | |
… | |
365 | delete $hdr{connection}; |
453 | delete $hdr{connection}; |
366 | } |
454 | } |
367 | |
455 | |
368 | # (re-)configure handle |
456 | # (re-)configure handle |
369 | $state{handle}->on_error (sub { |
457 | $state{handle}->on_error (sub { |
370 | my $errno = "$!"; |
|
|
371 | %state = (); |
458 | %state = (); |
372 | $cb->(undef, { Status => 599, Reason => $errno, URL => $url }); |
459 | $cb->(undef, { Status => 599, Reason => $_[2], URL => $url }); |
373 | }); |
460 | }); |
374 | $state{handle}->on_eof (sub { |
461 | $state{handle}->on_eof (sub { |
375 | %state = (); |
462 | %state = (); |
376 | $cb->(undef, { Status => 599, Reason => "unexpected end-of-file", URL => $url }); |
463 | $cb->(undef, { Status => 599, Reason => "Unexpected end-of-file", URL => $url }); |
377 | }); |
464 | }); |
378 | |
465 | |
379 | $state{handle}->starttls ("connect") if $rscheme eq "https"; |
466 | $state{handle}->starttls ("connect") if $rscheme eq "https"; |
380 | |
467 | |
381 | # handle actual, non-tunneled, request |
468 | # handle actual, non-tunneled, request |
… | |
… | |
393 | %hdr = (); # reduce memory usage, save a kitten |
480 | %hdr = (); # reduce memory usage, save a kitten |
394 | |
481 | |
395 | # status line |
482 | # status line |
396 | $state{handle}->push_read (line => $qr_nl, sub { |
483 | $state{handle}->push_read (line => $qr_nl, sub { |
397 | $_[1] =~ /^HTTP\/([0-9\.]+) \s+ ([0-9]{3}) (?: \s+ ([^\015\012]*) )?/ix |
484 | $_[1] =~ /^HTTP\/([0-9\.]+) \s+ ([0-9]{3}) (?: \s+ ([^\015\012]*) )?/ix |
398 | or return (%state = (), $cb->(undef, { Status => 599, Reason => "invalid server response ($_[1])", URL => $url })); |
485 | or return (%state = (), $cb->(undef, { Status => 599, Reason => "Invalid server response ($_[1])", URL => $url })); |
399 | |
486 | |
400 | my %hdr = ( # response headers |
487 | my %hdr = ( # response headers |
401 | HTTPVersion => ",$1", |
488 | HTTPVersion => ",$1", |
402 | Status => ",$2", |
489 | Status => ",$2", |
403 | Reason => ",$3", |
490 | Reason => ",$3", |
404 | URL => ",$url" |
491 | URL => ",$url" |
405 | ); |
492 | ); |
406 | |
493 | |
407 | # headers, could be optimized a bit |
494 | # headers, could be optimized a bit |
408 | $state{handle}->unshift_read (line => $qr_nlnl, sub { |
495 | $state{handle}->unshift_read (line => $qr_nlnl, sub { |
409 | for ("$_[1]\012") { |
496 | for ("$_[1]") { |
410 | y/\015//d; # weed out any \015, as they show up in the weirdest of places. |
497 | y/\015//d; # weed out any \015, as they show up in the weirdest of places. |
411 | |
498 | |
412 | # we support spaces in field names, as lotus domino |
499 | # things seen, not parsed: |
413 | # creates them (actually spaces around seperators |
500 | # p3pP="NON CUR OTPi OUR NOR UNI" |
414 | # are strictly allowed in http, they are a security issue). |
501 | |
415 | $hdr{lc $1} .= ",$2" |
502 | $hdr{lc $1} .= ",$2" |
416 | while /\G |
503 | while /\G |
417 | ([^:\000-\037]+): |
504 | ([^:\000-\037]*): |
418 | [\011\040]* |
505 | [\011\040]* |
419 | ((?: [^\012]+ | \012[\011\040] )*) |
506 | ((?: [^\012]+ | \012[\011\040] )*) |
420 | \012 |
507 | \012 |
421 | /gxc; |
508 | /gxc; |
422 | |
509 | |
423 | /\G$/ |
510 | /\G$/ |
424 | or return (%state = (), $cb->(undef, { Status => 599, Reason => "garbled response headers", URL => $url })); |
511 | or return (%state = (), $cb->(undef, { Status => 599, Reason => "Garbled response headers", URL => $url })); |
425 | } |
512 | } |
426 | |
513 | |
427 | substr $_, 0, 1, "" |
514 | substr $_, 0, 1, "" |
428 | for values %hdr; |
515 | for values %hdr; |
429 | |
516 | |
|
|
517 | # redirect handling |
|
|
518 | # microsoft and other shitheads don't give a shit for following standards, |
|
|
519 | # try to support some common forms of broken Location headers. |
|
|
520 | if ($hdr{location} !~ /^(?: $ | [^:\/?\#]+ : )/x) { |
|
|
521 | $hdr{location} =~ s/^\.\/+//; |
|
|
522 | |
|
|
523 | my $url = "$rscheme://$uhost:$uport"; |
|
|
524 | |
|
|
525 | unless ($hdr{location} =~ s/^\///) { |
|
|
526 | $url .= $upath; |
|
|
527 | $url =~ s/\/[^\/]*$//; |
|
|
528 | } |
|
|
529 | |
|
|
530 | $hdr{location} = "$url/$hdr{location}"; |
|
|
531 | } |
|
|
532 | |
|
|
533 | my $redirect; |
|
|
534 | |
|
|
535 | if ($recurse) { |
|
|
536 | if ($hdr{Status} =~ /^30[12]$/ && $method ne "POST") { |
|
|
537 | # apparently, mozilla et al. just change POST to GET here |
|
|
538 | # more research is needed before we do the same |
|
|
539 | $redirect = 1; |
|
|
540 | } elsif ($hdr{Status} == 303) { |
|
|
541 | # even http/1.1 is unclear on how to mutate the method |
|
|
542 | $method = "GET" unless $method eq "HEAD"; |
|
|
543 | $redirect = 1; |
|
|
544 | } elsif ($hdr{Status} == 307 && $method =~ /^(?:GET|HEAD)$/) { |
|
|
545 | $redirect = 1; |
|
|
546 | } |
|
|
547 | } |
|
|
548 | |
430 | my $finish = sub { |
549 | my $finish = sub { |
431 | $state{handle}->destroy; |
550 | $state{handle}->destroy if $state{handle}; |
432 | %state = (); |
551 | %state = (); |
433 | |
552 | |
434 | # set-cookie processing |
553 | # set-cookie processing |
435 | if ($arg{cookie_jar}) { |
554 | if ($arg{cookie_jar}) { |
436 | for ($hdr{"set-cookie"}) { |
555 | for ($_[1]{"set-cookie"}) { |
437 | # parse NAME=VALUE |
556 | # parse NAME=VALUE |
438 | my @kv; |
557 | my @kv; |
439 | |
558 | |
440 | while (/\G\s* ([^=;,[:space:]]+) \s*=\s* (?: "((?:[^\\"]+|\\.)*)" | ([^=;,[:space:]]*) )/gcxs) { |
559 | while (/\G\s* ([^=;,[:space:]]+) \s*=\s* (?: "((?:[^\\"]+|\\.)*)" | ([^=;,[:space:]]*) )/gcxs) { |
441 | my $name = $1; |
560 | my $name = $1; |
… | |
… | |
479 | |
598 | |
480 | redo if /\G\s*,/gc; |
599 | redo if /\G\s*,/gc; |
481 | } |
600 | } |
482 | } |
601 | } |
483 | |
602 | |
484 | # microsoft and other shitheads don't give a shit for following standards, |
603 | if ($redirect) { |
485 | # try to support some common forms of broken Location headers. |
604 | # we ignore any errors, as it is very common to receive |
486 | if ($_[1]{location} !~ /^(?: $ | [^:\/?\#]+ : )/x) { |
605 | # Content-Length != 0 but no actual body |
487 | $_[1]{location} =~ s/^\.\/+//; |
606 | # we also access %hdr, as $_[1] might be an erro |
488 | |
|
|
489 | my $url = "$rscheme://$uhost:$uport"; |
|
|
490 | |
|
|
491 | unless ($_[1]{location} =~ s/^\///) { |
|
|
492 | $url .= $upath; |
|
|
493 | $url =~ s/\/[^\/]*$//; |
|
|
494 | } |
|
|
495 | |
|
|
496 | $_[1]{location} = "$url/$_[1]{location}"; |
|
|
497 | } |
|
|
498 | |
|
|
499 | if ($_[1]{Status} =~ /^30[12]$/ && $recurse && $method ne "POST") { |
|
|
500 | # apparently, mozilla et al. just change POST to GET here |
|
|
501 | # more research is needed before we do the same |
|
|
502 | http_request ($method, $_[1]{location}, %arg, recurse => $recurse - 1, $cb); |
|
|
503 | } elsif ($_[1]{Status} == 303 && $recurse) { |
|
|
504 | # even http/1.1 is unclear on how to mutate the method |
|
|
505 | $method = "GET" unless $method eq "HEAD"; |
|
|
506 | http_request ($method => $_[1]{location}, %arg, recurse => $recurse - 1, $cb); |
607 | http_request ($method => $hdr{location}, %arg, recurse => $recurse - 1, $cb); |
507 | } elsif ($_[1]{Status} == 307 && $recurse && $method =~ /^(?:GET|HEAD)$/) { |
|
|
508 | http_request ($method => $_[1]{location}, %arg, recurse => $recurse - 1, $cb); |
|
|
509 | } else { |
608 | } else { |
510 | $cb->($_[0], $_[1]); |
609 | $cb->($_[0], $_[1]); |
511 | } |
610 | } |
512 | }; |
611 | }; |
513 | |
612 | |
514 | if ($hdr{Status} =~ /^(?:1..|204|304)$/ or $method eq "HEAD") { |
613 | my $len = $hdr{"content-length"}; |
|
|
614 | |
|
|
615 | if (!$redirect && $arg{on_header} && !$arg{on_header}(\%hdr)) { |
|
|
616 | $finish->(undef, { Status => 598, Reason => "Request cancelled by on_header", URL => $url }); |
|
|
617 | } elsif ( |
|
|
618 | $hdr{Status} =~ /^(?:1..|[23]04)$/ |
|
|
619 | or $method eq "HEAD" |
|
|
620 | or (defined $len && !$len) |
|
|
621 | ) { |
|
|
622 | # no body |
515 | $finish->(undef, \%hdr); |
623 | $finish->("", \%hdr); |
516 | } else { |
624 | } else { |
517 | if (exists $hdr{"content-length"}) { |
625 | # body handling, four different code paths |
518 | $_[0]->unshift_read (chunk => $hdr{"content-length"}, sub { |
626 | # for want_body_handle, on_body (2x), normal (2x) |
519 | # could cache persistent connection now |
627 | # we might read too much here, but it does not matter yet (no pers. connections) |
520 | if ($hdr{connection} =~ /\bkeep-alive\b/i) { |
628 | if (!$redirect && $arg{want_body_handle}) { |
521 | # but we don't, due to misdesigns, this is annoyingly complex |
629 | $_[0]->on_eof (undef); |
|
|
630 | $_[0]->on_error (undef); |
|
|
631 | $_[0]->on_read (undef); |
|
|
632 | |
|
|
633 | $finish->(delete $state{handle}, \%hdr); |
|
|
634 | |
|
|
635 | } elsif ($arg{on_body}) { |
|
|
636 | $_[0]->on_error (sub { $finish->(undef, { Status => 599, Reason => $_[2], URL => $url }) }); |
|
|
637 | if ($len) { |
|
|
638 | $_[0]->on_eof (undef); |
|
|
639 | $_[0]->on_read (sub { |
|
|
640 | $len -= length $_[0]{rbuf}; |
|
|
641 | |
|
|
642 | $arg{on_body}(delete $_[0]{rbuf}, \%hdr) |
|
|
643 | or $finish->(undef, { Status => 598, Reason => "Request cancelled by on_body", URL => $url }); |
|
|
644 | |
|
|
645 | $len > 0 |
|
|
646 | or $finish->("", \%hdr); |
522 | }; |
647 | }); |
523 | |
648 | } else { |
|
|
649 | $_[0]->on_eof (sub { |
524 | $finish->($_[1], \%hdr); |
650 | $finish->("", \%hdr); |
|
|
651 | }); |
|
|
652 | $_[0]->on_read (sub { |
|
|
653 | $arg{on_body}(delete $_[0]{rbuf}, \%hdr) |
|
|
654 | or $finish->(undef, { Status => 598, Reason => "Request cancelled by on_body", URL => $url }); |
|
|
655 | }); |
525 | }); |
656 | } |
526 | } else { |
657 | } else { |
527 | # too bad, need to read until we get an error or EOF, |
|
|
528 | # no way to detect winged data. |
|
|
529 | $_[0]->on_error (sub { |
|
|
530 | # delete ought to be more efficient, as we would have to make |
|
|
531 | # a copy otherwise as $_[0] gets destroyed. |
|
|
532 | $finish->(delete $_[0]{rbuf}, \%hdr); |
|
|
533 | }); |
|
|
534 | $_[0]->on_eof (undef); |
658 | $_[0]->on_eof (undef); |
|
|
659 | |
|
|
660 | if ($len) { |
|
|
661 | $_[0]->on_error (sub { $finish->(undef, { Status => 599, Reason => $_[2], URL => $url }) }); |
|
|
662 | $_[0]->on_read (sub { |
|
|
663 | $finish->((substr delete $_[0]{rbuf}, 0, $len, ""), \%hdr) |
|
|
664 | if $len <= length $_[0]{rbuf}; |
|
|
665 | }); |
|
|
666 | } else { |
|
|
667 | $_[0]->on_error (sub { |
|
|
668 | $! == Errno::EPIPE || !$! |
|
|
669 | ? $finish->(delete $_[0]{rbuf}, \%hdr) |
|
|
670 | : $finish->(undef, { Status => 599, Reason => $_[2], URL => $url }); |
|
|
671 | }); |
535 | $_[0]->on_read (sub { }); |
672 | $_[0]->on_read (sub { }); |
|
|
673 | } |
536 | } |
674 | } |
537 | } |
675 | } |
538 | }); |
676 | }); |
539 | }); |
677 | }); |
540 | }; |
678 | }; |
… | |
… | |
545 | |
683 | |
546 | # maybe re-use $uauthority with patched port? |
684 | # maybe re-use $uauthority with patched port? |
547 | $state{handle}->push_write ("CONNECT $uhost:$uport HTTP/1.0\015\012Host: $uhost\015\012\015\012"); |
685 | $state{handle}->push_write ("CONNECT $uhost:$uport HTTP/1.0\015\012Host: $uhost\015\012\015\012"); |
548 | $state{handle}->push_read (line => $qr_nlnl, sub { |
686 | $state{handle}->push_read (line => $qr_nlnl, sub { |
549 | $_[1] =~ /^HTTP\/([0-9\.]+) \s+ ([0-9]{3}) (?: \s+ ([^\015\012]*) )?/ix |
687 | $_[1] =~ /^HTTP\/([0-9\.]+) \s+ ([0-9]{3}) (?: \s+ ([^\015\012]*) )?/ix |
550 | or return (%state = (), $cb->(undef, { Status => 599, Reason => "invalid proxy connect response ($_[1])", URL => $url })); |
688 | or return (%state = (), $cb->(undef, { Status => 599, Reason => "Invalid proxy connect response ($_[1])", URL => $url })); |
551 | |
689 | |
552 | if ($2 == 200) { |
690 | if ($2 == 200) { |
553 | $rpath = $upath; |
691 | $rpath = $upath; |
554 | &$handle_actual_request; |
692 | &$handle_actual_request; |
555 | } else { |
693 | } else { |
… | |
… | |
601 | The default value for the C<recurse> request parameter (default: C<10>). |
739 | The default value for the C<recurse> request parameter (default: C<10>). |
602 | |
740 | |
603 | =item $AnyEvent::HTTP::USERAGENT |
741 | =item $AnyEvent::HTTP::USERAGENT |
604 | |
742 | |
605 | The default value for the C<User-Agent> header (the default is |
743 | The default value for the C<User-Agent> header (the default is |
606 | C<Mozilla/5.0 (compatible; AnyEvent::HTTP/$VERSION; +http://software.schmorp.de/pkg/AnyEvent)>). |
744 | C<Mozilla/5.0 (compatible; U; AnyEvent-HTTP/$VERSION; +http://software.schmorp.de/pkg/AnyEvent)>). |
607 | |
745 | |
608 | =item $AnyEvent::HTTP::MAX_PERSISTENT |
746 | =item $AnyEvent::HTTP::MAX_PER_HOST |
609 | |
747 | |
610 | The maximum number of persistent connections to keep open (default: 8). |
748 | The maximum number of concurrent conenctions to the same host (identified |
|
|
749 | by the hostname). If the limit is exceeded, then the additional requests |
|
|
750 | are queued until previous connections are closed. |
611 | |
751 | |
612 | Not implemented currently. |
752 | The default value for this is C<4>, and it is highly advisable to not |
613 | |
753 | increase it. |
614 | =item $AnyEvent::HTTP::PERSISTENT_TIMEOUT |
|
|
615 | |
|
|
616 | The maximum time to cache a persistent connection, in seconds (default: 2). |
|
|
617 | |
|
|
618 | Not implemented currently. |
|
|
619 | |
754 | |
620 | =item $AnyEvent::HTTP::ACTIVE |
755 | =item $AnyEvent::HTTP::ACTIVE |
621 | |
756 | |
622 | The number of active connections. This is not the number of currently |
757 | The number of active connections. This is not the number of currently |
623 | running requests, but the number of currently open and non-idle TCP |
758 | running requests, but the number of currently open and non-idle TCP |