… | |
… | |
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 (); |
|
|
49 | use AnyEvent::Handle (); |
48 | use AnyEvent::Handle (); |
50 | |
49 | |
51 | use base Exporter::; |
50 | use base Exporter::; |
52 | |
51 | |
53 | our $VERSION = '1.12'; |
52 | our $VERSION = '1.5'; |
54 | |
53 | |
55 | our @EXPORT = qw(http_get http_post http_head http_request); |
54 | our @EXPORT = qw(http_get http_post http_head http_request); |
56 | |
55 | |
57 | our $USERAGENT = "Mozilla/5.0 (compatible; U; AnyEvent-HTTP/$VERSION; +http://software.schmorp.de/pkg/AnyEvent)"; |
56 | our $USERAGENT = "Mozilla/5.0 (compatible; U; AnyEvent-HTTP/$VERSION; +http://software.schmorp.de/pkg/AnyEvent)"; |
58 | our $MAX_RECURSE = 10; |
57 | our $MAX_RECURSE = 10; |
59 | our $MAX_PERSISTENT = 8; |
58 | our $MAX_PERSISTENT = 8; |
60 | our $PERSISTENT_TIMEOUT = 2; |
59 | our $PERSISTENT_TIMEOUT = 2; |
61 | our $TIMEOUT = 300; |
60 | our $TIMEOUT = 300; |
62 | |
61 | |
63 | # changing these is evil |
62 | # changing these is evil |
64 | our $MAX_PERSISTENT_PER_HOST = 2; |
63 | our $MAX_PERSISTENT_PER_HOST = 0; |
65 | our $MAX_PER_HOST = 4; |
64 | our $MAX_PER_HOST = 4; |
66 | |
65 | |
67 | our $PROXY; |
66 | our $PROXY; |
68 | our $ACTIVE = 0; |
67 | our $ACTIVE = 0; |
69 | |
68 | |
… | |
… | |
92 | must be an absolute http or https URL. |
91 | must be an absolute http or https URL. |
93 | |
92 | |
94 | When called in void context, nothing is returned. In other contexts, |
93 | When called in void context, nothing is returned. In other contexts, |
95 | C<http_request> returns a "cancellation guard" - you have to keep the |
94 | 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 |
95 | 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. |
96 | destroyed before the callback is called, the request will be cancelled. |
98 | |
97 | |
99 | The callback will be called with the response body data as first argument |
98 | 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 |
99 | (or C<undef> if an error occured), and a hash-ref with response headers |
101 | second argument. |
100 | (and trailers) as second argument. |
102 | |
101 | |
103 | All the headers in that hash are lowercased. In addition to the response |
102 | 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> |
103 | headers, the "pseudo-headers" (uppercase to avoid clashing with possible |
|
|
104 | 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 |
105 | three parts of the HTTP Status-Line of the same name. If an error occurs |
|
|
106 | during the body phase of a request, then the original C<Status> and |
|
|
107 | C<Reason> values from the header are available as C<OrigStatus> and |
|
|
108 | C<OrigReason>. |
|
|
109 | |
106 | pseudo-header C<URL> contains the original URL (which can differ from the |
110 | The pseudo-header C<URL> contains the actual URL (which can differ from |
107 | requested URL when following redirects). |
111 | the requested URL when following redirects - for example, you might get |
|
|
112 | an error that your URL scheme is not supported even though your URL is a |
|
|
113 | valid http URL because it redirected to an ftp URL, in which case you can |
|
|
114 | look at the URL pseudo header). |
|
|
115 | |
|
|
116 | The pseudo-header C<Redirect> only exists when the request was a result |
|
|
117 | of an internal redirect. In that case it is an array reference with |
|
|
118 | the C<($data, $headers)> from the redirect response. Note that this |
|
|
119 | response could in turn be the result of a redirect itself, and C<< |
|
|
120 | $headers->{Redirect}[1]{Redirect} >> will then contain the original |
|
|
121 | response, and so on. |
108 | |
122 | |
109 | If the server sends a header multiple times, then their contents will be |
123 | 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. |
124 | joined together with a comma (C<,>), as per the HTTP spec. |
111 | |
125 | |
112 | If an internal error occurs, such as not being able to resolve a hostname, |
126 | If an internal error occurs, such as not being able to resolve a hostname, |
… | |
… | |
136 | Whether to recurse requests or not, e.g. on redirects, authentication |
150 | Whether to recurse requests or not, e.g. on redirects, authentication |
137 | retries and so on, and how often to do so. |
151 | retries and so on, and how often to do so. |
138 | |
152 | |
139 | =item headers => hashref |
153 | =item headers => hashref |
140 | |
154 | |
141 | The request headers to use. Currently, C<http_request> may provide its |
155 | The request headers to use. Currently, C<http_request> may provide its own |
142 | own C<Host:>, C<Content-Length:>, C<Connection:> and C<Cookie:> headers |
156 | C<Host:>, C<Content-Length:>, C<Connection:> and C<Cookie:> headers and |
143 | and will provide defaults for C<User-Agent:> and C<Referer:>. |
157 | will provide defaults at least for C<TE:>, C<Referer:> and C<User-Agent:> |
|
|
158 | (this can be suppressed by using C<undef> for these headers in which case |
|
|
159 | they won't be sent at all). |
144 | |
160 | |
145 | =item timeout => $seconds |
161 | =item timeout => $seconds |
146 | |
162 | |
147 | The time-out to use for various stages - each connect attempt will reset |
163 | The time-out to use for various stages - each connect attempt will reset |
148 | the timeout, as will read or write activity. Default timeout is 5 minutes. |
164 | the timeout, as will read or write activity, i.e. this is not an overall |
|
|
165 | timeout. |
|
|
166 | |
|
|
167 | Default timeout is 5 minutes. |
149 | |
168 | |
150 | =item proxy => [$host, $port[, $scheme]] or undef |
169 | =item proxy => [$host, $port[, $scheme]] or undef |
151 | |
170 | |
152 | Use the given http proxy for all requests. If not specified, then the |
171 | Use the given http proxy for all requests. If not specified, then the |
153 | default proxy (as specified by C<$ENV{http_proxy}>) is used. |
172 | default proxy (as specified by C<$ENV{http_proxy}>) is used. |
154 | |
173 | |
155 | C<$scheme> must be either missing or C<http> for HTTP, or C<https> for |
174 | C<$scheme> must be either missing, C<http> for HTTP or C<https> for |
156 | HTTPS. |
175 | HTTPS. |
157 | |
176 | |
158 | =item body => $string |
177 | =item body => $string |
159 | |
178 | |
160 | The request body, usually empty. Will be-sent as-is (future versions of |
179 | The request body, usually empty. Will be sent as-is (future versions of |
161 | this module might offer more options). |
180 | this module might offer more options). |
162 | |
181 | |
163 | =item cookie_jar => $hash_ref |
182 | =item cookie_jar => $hash_ref |
164 | |
183 | |
165 | Passing this parameter enables (simplified) cookie-processing, loosely |
184 | Passing this parameter enables (simplified) cookie-processing, loosely |
166 | based on the original netscape specification. |
185 | based on the original netscape specification. |
167 | |
186 | |
168 | The C<$hash_ref> must be an (initially empty) hash reference which will |
187 | The C<$hash_ref> must be an (initially empty) hash reference which will |
169 | get updated automatically. It is possible to save the cookie_jar to |
188 | get updated automatically. It is possible to save the cookie jar to |
170 | persistent storage with something like JSON or Storable, but this is not |
189 | persistent storage with something like JSON or Storable, but this is not |
171 | recommended, as expiry times are currently being ignored. |
190 | recommended, as session-only cookies might survive longer than expected. |
172 | |
191 | |
173 | Note that this cookie implementation is not of very high quality, nor |
192 | Note that this cookie implementation is not meant to be complete. If |
174 | meant to be complete. If you want complete cookie management you have to |
193 | you want complete cookie management you have to do that on your |
175 | do that on your own. C<cookie_jar> is meant as a quick fix to get some |
194 | own. C<cookie_jar> is meant as a quick fix to get some cookie-using sites |
176 | cookie-using sites working. Cookies are a privacy disaster, do not use |
195 | working. Cookies are a privacy disaster, do not use them unless required |
177 | them unless required to. |
196 | to. |
|
|
197 | |
|
|
198 | 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 |
|
|
200 | left untouched. |
178 | |
201 | |
179 | =item tls_ctx => $scheme | $tls_ctx |
202 | =item tls_ctx => $scheme | $tls_ctx |
180 | |
203 | |
181 | Specifies the AnyEvent::TLS context to be used for https connections. This |
204 | 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 |
205 | parameter follows the same rules as the C<tls_ctx> parameter to |
… | |
… | |
186 | verification) TLS context. |
209 | verification) TLS context. |
187 | |
210 | |
188 | The default for this option is C<low>, which could be interpreted as "give |
211 | The default for this option is C<low>, which could be interpreted as "give |
189 | me the page, no matter what". |
212 | me the page, no matter what". |
190 | |
213 | |
|
|
214 | =item on_prepare => $callback->($fh) |
|
|
215 | |
|
|
216 | In rare cases you need to "tune" the socket before it is used to |
|
|
217 | connect (for exmaple, to bind it on a given IP address). This parameter |
|
|
218 | overrides the prepare callback passed to C<AnyEvent::Socket::tcp_connect> |
|
|
219 | and behaves exactly the same way (e.g. it has to provide a |
|
|
220 | timeout). See the description for the C<$prepare_cb> argument of |
|
|
221 | C<AnyEvent::Socket::tcp_connect> for details. |
|
|
222 | |
|
|
223 | =item tcp_connect => $callback->($host, $service, $connect_cb, $prepare_cb) |
|
|
224 | |
|
|
225 | In even rarer cases you want total control over how AnyEvent::HTTP |
|
|
226 | establishes connections. Normally it uses L<AnyEvent::Socket::tcp_connect> |
|
|
227 | to do this, but you can provide your own C<tcp_connect> function - |
|
|
228 | obviously, it has to follow the same calling conventions, except that it |
|
|
229 | may always return a connection guard object. |
|
|
230 | |
|
|
231 | There are probably lots of weird uses for this function, starting from |
|
|
232 | tracing the hosts C<http_request> actually tries to connect, to (inexact |
|
|
233 | but fast) host => IP address caching or even socks protocol support. |
|
|
234 | |
191 | =item on_header => $callback->($headers) |
235 | =item on_header => $callback->($headers) |
192 | |
236 | |
193 | When specified, this callback will be called with the header hash as soon |
237 | 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 |
238 | as headers have been successfully received from the remote server (not on |
195 | locally-generated errors). |
239 | locally-generated errors). |
… | |
… | |
200 | |
244 | |
201 | This callback is useful, among other things, to quickly reject unwanted |
245 | 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 |
246 | content, which, if it is supposed to be rare, can be faster than first |
203 | doing a C<HEAD> request. |
247 | doing a C<HEAD> request. |
204 | |
248 | |
|
|
249 | The downside is that cancelling the request makes it impossible to re-use |
|
|
250 | the connection. Also, the C<on_header> callback will not receive any |
|
|
251 | trailer (headers sent after the response body). |
|
|
252 | |
205 | Example: cancel the request unless the content-type is "text/html". |
253 | Example: cancel the request unless the content-type is "text/html". |
206 | |
254 | |
207 | on_header => sub { |
255 | on_header => sub { |
208 | $_[0]{"content-type"} =~ /^text\/html\s*(?:;|$)/ |
256 | $_[0]{"content-type"} =~ /^text\/html\s*(?:;|$)/ |
209 | }, |
257 | }, |
… | |
… | |
216 | |
264 | |
217 | It has to return either true (in which case AnyEvent::HTTP will continue), |
265 | 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 |
266 | or false, in which case AnyEvent::HTTP will cancel the download (and call |
219 | the completion callback with an error code of C<598>). |
267 | the completion callback with an error code of C<598>). |
220 | |
268 | |
|
|
269 | The downside to cancelling the request is that it makes it impossible to |
|
|
270 | re-use the connection. |
|
|
271 | |
221 | This callback is useful when the data is too large to be held in memory |
272 | 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 |
273 | (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. |
274 | be extracted, or when the body should be processed incrementally. |
224 | |
275 | |
225 | It is usually preferred over doing your own body handling via |
276 | It is usually preferred over doing your own body handling via |
226 | C<want_body_handle>. |
277 | C<want_body_handle>, but in case of streaming APIs, where HTTP is |
|
|
278 | only used to create a connection, C<want_body_handle> is the better |
|
|
279 | alternative, as it allows you to install your own event handler, reducing |
|
|
280 | resource usage. |
227 | |
281 | |
228 | =item want_body_handle => $enable |
282 | =item want_body_handle => $enable |
229 | |
283 | |
230 | When enabled (default is disabled), the behaviour of AnyEvent::HTTP |
284 | When enabled (default is disabled), the behaviour of AnyEvent::HTTP |
231 | changes considerably: after parsing the headers, and instead of |
285 | changes considerably: after parsing the headers, and instead of |
… | |
… | |
243 | This is useful with some push-type services, where, after the initial |
297 | This is useful with some push-type services, where, after the initial |
244 | headers, an interactive protocol is used (typical example would be the |
298 | headers, an interactive protocol is used (typical example would be the |
245 | push-style twitter API which starts a JSON/XML stream). |
299 | push-style twitter API which starts a JSON/XML stream). |
246 | |
300 | |
247 | If you think you need this, first have a look at C<on_body>, to see if |
301 | If you think you need this, first have a look at C<on_body>, to see if |
248 | that doesn'T solve your problem in a better way. |
302 | that doesn't solve your problem in a better way. |
249 | |
303 | |
250 | =back |
304 | =back |
251 | |
305 | |
252 | Example: make a simple HTTP GET request for http://www.nethype.de/ |
306 | Example: do a simple HTTP GET request for http://www.nethype.de/ and print |
|
|
307 | the response body. |
253 | |
308 | |
254 | http_request GET => "http://www.nethype.de/", sub { |
309 | http_request GET => "http://www.nethype.de/", sub { |
255 | my ($body, $hdr) = @_; |
310 | my ($body, $hdr) = @_; |
256 | print "$body\n"; |
311 | print "$body\n"; |
257 | }; |
312 | }; |
258 | |
313 | |
259 | Example: make a HTTP HEAD request on https://www.google.com/, use a |
314 | Example: do a HTTP HEAD request on https://www.google.com/, use a |
260 | timeout of 30 seconds. |
315 | timeout of 30 seconds. |
261 | |
316 | |
262 | http_request |
317 | http_request |
263 | GET => "https://www.google.com", |
318 | GET => "https://www.google.com", |
264 | timeout => 30, |
319 | timeout => 30, |
… | |
… | |
267 | use Data::Dumper; |
322 | use Data::Dumper; |
268 | print Dumper $hdr; |
323 | print Dumper $hdr; |
269 | } |
324 | } |
270 | ; |
325 | ; |
271 | |
326 | |
272 | Example: make another simple HTTP GET request, but immediately try to |
327 | Example: do another simple HTTP GET request, but immediately try to |
273 | cancel it. |
328 | cancel it. |
274 | |
329 | |
275 | my $request = http_request GET => "http://www.nethype.de/", sub { |
330 | my $request = http_request GET => "http://www.nethype.de/", sub { |
276 | my ($body, $hdr) = @_; |
331 | my ($body, $hdr) = @_; |
277 | print "$body\n"; |
332 | print "$body\n"; |
… | |
… | |
309 | push @{ $CO_SLOT{$_[0]}[1] }, $_[1]; |
364 | push @{ $CO_SLOT{$_[0]}[1] }, $_[1]; |
310 | |
365 | |
311 | _slot_schedule $_[0]; |
366 | _slot_schedule $_[0]; |
312 | } |
367 | } |
313 | |
368 | |
314 | our $qr_nl = qr<\015?\012>; |
369 | sub cookie_jar_extract($$$$) { |
315 | our $qr_nlnl = qr<\015?\012\015?\012>; |
370 | my ($jar, $uscheme, $uhost, $upath) = @_; |
|
|
371 | |
|
|
372 | %$jar = () if $jar->{version} != 1; |
|
|
373 | |
|
|
374 | my @cookies; |
|
|
375 | |
|
|
376 | while (my ($chost, $paths) = each %$jar) { |
|
|
377 | next unless ref $paths; |
|
|
378 | |
|
|
379 | if ($chost =~ /^\./) { |
|
|
380 | next unless $chost eq substr $uhost, -length $chost; |
|
|
381 | } elsif ($chost =~ /\./) { |
|
|
382 | next unless $chost eq $uhost; |
|
|
383 | } else { |
|
|
384 | next; |
|
|
385 | } |
|
|
386 | |
|
|
387 | while (my ($cpath, $cookies) = each %$paths) { |
|
|
388 | next unless $cpath eq substr $upath, 0, length $cpath; |
|
|
389 | |
|
|
390 | while (my ($cookie, $kv) = each %$cookies) { |
|
|
391 | next if $uscheme ne "https" && exists $kv->{secure}; |
|
|
392 | |
|
|
393 | if (exists $kv->{expires}) { |
|
|
394 | if (AE::now > parse_date ($kv->{expires})) { |
|
|
395 | delete $cookies->{$cookie}; |
|
|
396 | next; |
|
|
397 | } |
|
|
398 | } |
|
|
399 | |
|
|
400 | my $value = $kv->{value}; |
|
|
401 | |
|
|
402 | if ($value =~ /[=;,[:space:]]/) { |
|
|
403 | $value =~ s/([\\"])/\\$1/g; |
|
|
404 | $value = "\"$value\""; |
|
|
405 | } |
|
|
406 | |
|
|
407 | push @cookies, "$cookie=$value"; |
|
|
408 | } |
|
|
409 | } |
|
|
410 | } |
|
|
411 | |
|
|
412 | \@cookies |
|
|
413 | } |
|
|
414 | |
|
|
415 | # continue to parse $_ for headers and place them into the arg |
|
|
416 | sub parse_hdr() { |
|
|
417 | my %hdr; |
|
|
418 | |
|
|
419 | # things seen, not parsed: |
|
|
420 | # p3pP="NON CUR OTPi OUR NOR UNI" |
|
|
421 | |
|
|
422 | $hdr{lc $1} .= ",$2" |
|
|
423 | while /\G |
|
|
424 | ([^:\000-\037]*): |
|
|
425 | [\011\040]* |
|
|
426 | ((?: [^\012]+ | \012[\011\040] )*) |
|
|
427 | \012 |
|
|
428 | /gxc; |
|
|
429 | |
|
|
430 | /\G$/ |
|
|
431 | or return; |
|
|
432 | |
|
|
433 | # remove the "," prefix we added to all headers above |
|
|
434 | substr $_, 0, 1, "" |
|
|
435 | for values %hdr; |
|
|
436 | |
|
|
437 | \%hdr |
|
|
438 | } |
|
|
439 | |
|
|
440 | our $qr_nlnl = qr{(?<![^\012])\015?\012}; |
316 | |
441 | |
317 | our $TLS_CTX_LOW = { cache => 1, sslv2 => 1 }; |
442 | our $TLS_CTX_LOW = { cache => 1, sslv2 => 1 }; |
318 | our $TLS_CTX_HIGH = { cache => 1, verify => 1, verify_peername => "https" }; |
443 | our $TLS_CTX_HIGH = { cache => 1, verify => 1, verify_peername => "https" }; |
319 | |
444 | |
320 | sub http_request($$@) { |
445 | sub http_request($$@) { |
… | |
… | |
332 | while (my ($k, $v) = each %$hdr) { |
457 | while (my ($k, $v) = each %$hdr) { |
333 | $hdr{lc $k} = $v; |
458 | $hdr{lc $k} = $v; |
334 | } |
459 | } |
335 | } |
460 | } |
336 | |
461 | |
|
|
462 | # pseudo headers for all subsequent responses |
|
|
463 | my @pseudo = (URL => $url); |
|
|
464 | push @pseudo, Redirect => delete $arg{Redirect} if exists $arg{Redirect}; |
|
|
465 | |
337 | my $recurse = exists $arg{recurse} ? delete $arg{recurse} : $MAX_RECURSE; |
466 | my $recurse = exists $arg{recurse} ? delete $arg{recurse} : $MAX_RECURSE; |
338 | |
467 | |
339 | return $cb->(undef, { Status => 599, Reason => "Too many redirections", URL => $url }) |
468 | return $cb->(undef, { @pseudo, Status => 599, Reason => "Too many redirections" }) |
340 | if $recurse < 0; |
469 | if $recurse < 0; |
341 | |
470 | |
342 | my $proxy = $arg{proxy} || $PROXY; |
471 | my $proxy = $arg{proxy} || $PROXY; |
343 | my $timeout = $arg{timeout} || $TIMEOUT; |
472 | my $timeout = $arg{timeout} || $TIMEOUT; |
344 | |
473 | |
345 | my ($uscheme, $uauthority, $upath, $query, $fragment) = |
474 | my ($uscheme, $uauthority, $upath, $query, $fragment) = |
346 | $url =~ m|(?:([^:/?#]+):)?(?://([^/?#]*))?([^?#]*)(?:\?([^#]*))?(?:#(.*))?|; |
475 | $url =~ m|(?:([^:/?#]+):)?(?://([^/?#]*))?([^?#]*)(?:(\?[^#]*))?(?:#(.*))?|; |
347 | |
476 | |
348 | $uscheme = lc $uscheme; |
477 | $uscheme = lc $uscheme; |
349 | |
478 | |
350 | my $uport = $uscheme eq "http" ? 80 |
479 | my $uport = $uscheme eq "http" ? 80 |
351 | : $uscheme eq "https" ? 443 |
480 | : $uscheme eq "https" ? 443 |
352 | : return $cb->(undef, { Status => 599, Reason => "Only http and https URL schemes supported", URL => $url }); |
481 | : return $cb->(undef, { @pseudo, Status => 599, Reason => "Only http and https URL schemes supported" }); |
353 | |
482 | |
354 | $uauthority =~ /^(?: .*\@ )? ([^\@:]+) (?: : (\d+) )?$/x |
483 | $uauthority =~ /^(?: .*\@ )? ([^\@:]+) (?: : (\d+) )?$/x |
355 | or return $cb->(undef, { Status => 599, Reason => "Unparsable URL", URL => $url }); |
484 | or return $cb->(undef, { @pseudo, Status => 599, Reason => "Unparsable URL" }); |
356 | |
485 | |
357 | my $uhost = $1; |
486 | my $uhost = $1; |
358 | $uport = $2 if defined $2; |
487 | $uport = $2 if defined $2; |
359 | |
488 | |
|
|
489 | $hdr{host} = defined $2 ? "$uhost:$2" : "$uhost" |
|
|
490 | unless exists $hdr{host}; |
|
|
491 | |
360 | $uhost =~ s/^\[(.*)\]$/$1/; |
492 | $uhost =~ s/^\[(.*)\]$/$1/; |
361 | $upath .= "?$query" if length $query; |
493 | $upath .= $query if length $query; |
362 | |
494 | |
363 | $upath =~ s%^/?%/%; |
495 | $upath =~ s%^/?%/%; |
364 | |
496 | |
365 | # cookie processing |
497 | # cookie processing |
366 | if (my $jar = $arg{cookie_jar}) { |
498 | if (my $jar = $arg{cookie_jar}) { |
367 | %$jar = () if $jar->{version} != 1; |
499 | my $cookies = cookie_jar_extract $jar, $uscheme, $uhost, $upath; |
368 | |
500 | |
369 | my @cookie; |
|
|
370 | |
|
|
371 | while (my ($chost, $v) = each %$jar) { |
|
|
372 | if ($chost =~ /^\./) { |
|
|
373 | next unless $chost eq substr $uhost, -length $chost; |
|
|
374 | } elsif ($chost =~ /\./) { |
|
|
375 | next unless $chost eq $uhost; |
|
|
376 | } else { |
|
|
377 | next; |
|
|
378 | } |
|
|
379 | |
|
|
380 | while (my ($cpath, $v) = each %$v) { |
|
|
381 | next unless $cpath eq substr $upath, 0, length $cpath; |
|
|
382 | |
|
|
383 | while (my ($k, $v) = each %$v) { |
|
|
384 | next if $uscheme ne "https" && exists $v->{secure}; |
|
|
385 | my $value = $v->{value}; |
|
|
386 | $value =~ s/([\\"])/\\$1/g; |
|
|
387 | push @cookie, "$k=\"$value\""; |
|
|
388 | } |
|
|
389 | } |
|
|
390 | } |
|
|
391 | |
|
|
392 | $hdr{cookie} = join "; ", @cookie |
501 | $hdr{cookie} = join "; ", @$cookies |
393 | if @cookie; |
502 | if @$cookies; |
394 | } |
503 | } |
395 | |
504 | |
396 | my ($rhost, $rport, $rscheme, $rpath); # request host, port, path |
505 | my ($rhost, $rport, $rscheme, $rpath); # request host, port, path |
397 | |
506 | |
398 | if ($proxy) { |
507 | if ($proxy) { |
399 | ($rpath, $rhost, $rport, $rscheme) = ($url, @$proxy); |
508 | ($rpath, $rhost, $rport, $rscheme) = ($url, @$proxy); |
|
|
509 | |
|
|
510 | $rscheme = "http" unless defined $rscheme; |
400 | |
511 | |
401 | # don't support https requests over https-proxy transport, |
512 | # don't support https requests over https-proxy transport, |
402 | # can't be done with tls as spec'ed, unless you double-encrypt. |
513 | # can't be done with tls as spec'ed, unless you double-encrypt. |
403 | $rscheme = "http" if $uscheme eq "https" && $rscheme eq "https"; |
514 | $rscheme = "http" if $uscheme eq "https" && $rscheme eq "https"; |
404 | } else { |
515 | } else { |
405 | ($rhost, $rport, $rscheme, $rpath) = ($uhost, $uport, $uscheme, $upath); |
516 | ($rhost, $rport, $rscheme, $rpath) = ($uhost, $uport, $uscheme, $upath); |
406 | } |
517 | } |
407 | |
518 | |
408 | $hdr{"user-agent"} ||= $USERAGENT; |
519 | # leave out fragment and query string, just a heuristic |
409 | $hdr{referer} ||= "$uscheme://$uauthority$upath"; # leave out fragment and query string, just a heuristic |
520 | $hdr{referer} = "$uscheme://$uauthority$upath" unless exists $hdr{referer}; |
|
|
521 | $hdr{"user-agent"} = $USERAGENT unless exists $hdr{"user-agent"}; |
410 | |
522 | |
411 | $hdr{host} = "$uhost:$uport"; |
|
|
412 | $hdr{"content-length"} = length $arg{body}; |
523 | $hdr{"content-length"} = length $arg{body} |
|
|
524 | if length $arg{body} || $method ne "GET"; |
|
|
525 | |
|
|
526 | $hdr{connection} = "close TE"; #1.1 |
|
|
527 | $hdr{te} = "trailers" unless exists $hdr{te}; #1.1 |
413 | |
528 | |
414 | my %state = (connect_guard => 1); |
529 | my %state = (connect_guard => 1); |
415 | |
530 | |
416 | _get_slot $uhost, sub { |
531 | _get_slot $uhost, sub { |
417 | $state{slot_guard} = shift; |
532 | $state{slot_guard} = shift; |
418 | |
533 | |
419 | return unless $state{connect_guard}; |
534 | return unless $state{connect_guard}; |
420 | |
535 | |
421 | $state{connect_guard} = AnyEvent::Socket::tcp_connect $rhost, $rport, sub { |
536 | my $connect_cb = sub { |
422 | $state{fh} = shift |
537 | $state{fh} = shift |
423 | or return (%state = (), $cb->(undef, { Status => 599, Reason => "$!", URL => $url })); |
538 | or do { |
|
|
539 | my $err = "$!"; |
|
|
540 | %state = (); |
|
|
541 | return $cb->(undef, { @pseudo, Status => 599, Reason => $err }); |
|
|
542 | }; |
|
|
543 | |
424 | pop; # free memory, save a tree |
544 | pop; # free memory, save a tree |
425 | |
545 | |
426 | return unless delete $state{connect_guard}; |
546 | return unless delete $state{connect_guard}; |
427 | |
547 | |
428 | # get handle |
548 | # get handle |
429 | $state{handle} = new AnyEvent::Handle |
549 | $state{handle} = new AnyEvent::Handle |
430 | fh => $state{fh}, |
550 | fh => $state{fh}, |
|
|
551 | peername => $rhost, |
|
|
552 | tls_ctx => $arg{tls_ctx}, |
|
|
553 | # these need to be reconfigured on keepalive handles |
431 | timeout => $timeout, |
554 | timeout => $timeout, |
432 | peername => $rhost, |
555 | on_error => sub { |
433 | tls_ctx => $arg{tls_ctx}; |
556 | %state = (); |
|
|
557 | $cb->(undef, { @pseudo, Status => 599, Reason => $_[2] }); |
|
|
558 | }, |
|
|
559 | on_eof => sub { |
|
|
560 | %state = (); |
|
|
561 | $cb->(undef, { @pseudo, Status => 599, Reason => "Unexpected end-of-file" }); |
|
|
562 | }, |
|
|
563 | ; |
434 | |
564 | |
435 | # limit the number of persistent connections |
565 | # limit the number of persistent connections |
436 | # keepalive not yet supported |
566 | # keepalive not yet supported |
437 | if ($KA_COUNT{$_[1]} < $MAX_PERSISTENT_PER_HOST) { |
567 | # if ($KA_COUNT{$_[1]} < $MAX_PERSISTENT_PER_HOST) { |
438 | ++$KA_COUNT{$_[1]}; |
568 | # ++$KA_COUNT{$_[1]}; |
439 | $state{handle}{ka_count_guard} = AnyEvent::Util::guard { |
569 | # $state{handle}{ka_count_guard} = AnyEvent::Util::guard { |
440 | --$KA_COUNT{$_[1]} |
570 | # --$KA_COUNT{$_[1]} |
441 | }; |
571 | # }; |
442 | $hdr{connection} = "keep-alive"; |
572 | # $hdr{connection} = "keep-alive"; |
443 | } else { |
|
|
444 | delete $hdr{connection}; |
|
|
445 | } |
573 | # } |
446 | |
|
|
447 | # (re-)configure handle |
|
|
448 | $state{handle}->on_error (sub { |
|
|
449 | %state = (); |
|
|
450 | $cb->(undef, { Status => 599, Reason => $_[2], URL => $url }); |
|
|
451 | }); |
|
|
452 | $state{handle}->on_eof (sub { |
|
|
453 | %state = (); |
|
|
454 | $cb->(undef, { Status => 599, Reason => "Unexpected end-of-file", URL => $url }); |
|
|
455 | }); |
|
|
456 | |
574 | |
457 | $state{handle}->starttls ("connect") if $rscheme eq "https"; |
575 | $state{handle}->starttls ("connect") if $rscheme eq "https"; |
458 | |
576 | |
459 | # handle actual, non-tunneled, request |
577 | # handle actual, non-tunneled, request |
460 | my $handle_actual_request = sub { |
578 | my $handle_actual_request = sub { |
461 | $state{handle}->starttls ("connect") if $uscheme eq "https" && !exists $state{handle}{tls}; |
579 | $state{handle}->starttls ("connect") if $uscheme eq "https" && !exists $state{handle}{tls}; |
462 | |
580 | |
463 | # send request |
581 | # send request |
464 | $state{handle}->push_write ( |
582 | $state{handle}->push_write ( |
465 | "$method $rpath HTTP/1.0\015\012" |
583 | "$method $rpath HTTP/1.1\015\012" |
466 | . (join "", map "\u$_: $hdr{$_}\015\012", keys %hdr) |
584 | . (join "", map "\u$_: $hdr{$_}\015\012", grep defined $hdr{$_}, keys %hdr) |
467 | . "\015\012" |
585 | . "\015\012" |
468 | . (delete $arg{body}) |
586 | . (delete $arg{body}) |
469 | ); |
587 | ); |
470 | |
588 | |
|
|
589 | # return if error occured during push_write() |
|
|
590 | return unless %state; |
|
|
591 | |
471 | %hdr = (); # reduce memory usage, save a kitten |
592 | %hdr = (); # reduce memory usage, save a kitten, also make it possible to re-use |
472 | |
593 | |
473 | # status line |
594 | # status line and headers |
474 | $state{handle}->push_read (line => $qr_nl, sub { |
595 | $state{read_response} = sub { |
475 | $_[1] =~ /^HTTP\/([0-9\.]+) \s+ ([0-9]{3}) (?: \s+ ([^\015\012]*) )?/ix |
|
|
476 | or return (%state = (), $cb->(undef, { Status => 599, Reason => "Invalid server response ($_[1])", URL => $url })); |
|
|
477 | |
|
|
478 | my %hdr = ( # response headers |
|
|
479 | HTTPVersion => ",$1", |
|
|
480 | Status => ",$2", |
|
|
481 | Reason => ",$3", |
|
|
482 | URL => ",$url" |
|
|
483 | ); |
|
|
484 | |
|
|
485 | # headers, could be optimized a bit |
|
|
486 | $state{handle}->unshift_read (line => $qr_nlnl, sub { |
|
|
487 | for ("$_[1]\012") { |
596 | for ("$_[1]") { |
488 | y/\015//d; # weed out any \015, as they show up in the weirdest of places. |
597 | y/\015//d; # weed out any \015, as they show up in the weirdest of places. |
489 | |
598 | |
490 | # things seen, not parsed: |
599 | /^HTTP\/([0-9\.]+) \s+ ([0-9]{3}) (?: \s+ ([^\012]*) )? \012/igxc |
491 | # p3pP="NON CUR OTPi OUR NOR UNI" |
600 | or return (%state = (), $cb->(undef, { @pseudo, Status => 599, Reason => "Invalid server response" })); |
492 | |
601 | |
493 | $hdr{lc $1} .= ",$2" |
602 | # 100 Continue handling |
494 | while /\G |
603 | # should not happen as we don't send expect: 100-continue, |
495 | ([^:\000-\037]+): |
604 | # but we handle it just in case. |
496 | [\011\040]* |
605 | # since we send the request body regardless, if we get an error |
497 | ((?: [^\012]+ | \012[\011\040] )*) |
606 | # we are out of-sync, which we currently do NOT handle correctly. |
498 | \012 |
607 | return $state{handle}->push_read (line => $qr_nlnl, $state{read_response}) |
499 | /gxc; |
608 | if $2 eq 100; |
500 | |
609 | |
|
|
610 | push @pseudo, |
|
|
611 | HTTPVersion => $1, |
|
|
612 | Status => $2, |
|
|
613 | Reason => $3, |
501 | /\G$/ |
614 | ; |
|
|
615 | |
|
|
616 | my $hdr = parse_hdr |
502 | or return (%state = (), $cb->(undef, { Status => 599, Reason => "Garbled response headers", URL => $url })); |
617 | or return (%state = (), $cb->(undef, { @pseudo, Status => 599, Reason => "Garbled response headers" })); |
|
|
618 | |
|
|
619 | %hdr = (%$hdr, @pseudo); |
|
|
620 | } |
|
|
621 | |
|
|
622 | # redirect handling |
|
|
623 | # microsoft and other shitheads don't give a shit for following standards, |
|
|
624 | # try to support some common forms of broken Location headers. |
|
|
625 | if ($hdr{location} !~ /^(?: $ | [^:\/?\#]+ : )/x) { |
|
|
626 | $hdr{location} =~ s/^\.\/+//; |
|
|
627 | |
|
|
628 | my $url = "$rscheme://$uhost:$uport"; |
|
|
629 | |
|
|
630 | unless ($hdr{location} =~ s/^\///) { |
|
|
631 | $url .= $upath; |
|
|
632 | $url =~ s/\/[^\/]*$//; |
503 | } |
633 | } |
504 | |
634 | |
505 | substr $_, 0, 1, "" |
|
|
506 | for values %hdr; |
|
|
507 | |
|
|
508 | # redirect handling |
|
|
509 | # microsoft and other shitheads don't give a shit for following standards, |
|
|
510 | # try to support some common forms of broken Location headers. |
|
|
511 | if ($hdr{location} !~ /^(?: $ | [^:\/?\#]+ : )/x) { |
|
|
512 | $hdr{location} =~ s/^\.\/+//; |
|
|
513 | |
|
|
514 | my $url = "$rscheme://$uhost:$uport"; |
|
|
515 | |
|
|
516 | unless ($hdr{location} =~ s/^\///) { |
|
|
517 | $url .= $upath; |
|
|
518 | $url =~ s/\/[^\/]*$//; |
|
|
519 | } |
|
|
520 | |
|
|
521 | $hdr{location} = "$url/$hdr{location}"; |
635 | $hdr{location} = "$url/$hdr{location}"; |
|
|
636 | } |
|
|
637 | |
|
|
638 | my $redirect; |
|
|
639 | |
|
|
640 | if ($recurse) { |
|
|
641 | my $status = $hdr{Status}; |
|
|
642 | |
|
|
643 | # industry standard is to redirect POST as GET for |
|
|
644 | # 301, 302 and 303, in contrast to http/1.0 and 1.1. |
|
|
645 | # also, the UA should ask the user for 301 and 307 and POST, |
|
|
646 | # industry standard seems to be to simply follow. |
|
|
647 | # we go with the industry standard. |
|
|
648 | if ($status == 301 or $status == 302 or $status == 303) { |
|
|
649 | # HTTP/1.1 is unclear on how to mutate the method |
|
|
650 | $method = "GET" unless $method eq "HEAD"; |
|
|
651 | $redirect = 1; |
|
|
652 | } elsif ($status == 307) { |
|
|
653 | $redirect = 1; |
522 | } |
654 | } |
|
|
655 | } |
523 | |
656 | |
|
|
657 | my $finish = sub { # ($data, $err_status, $err_reason[, $keepalive]) |
524 | my $redirect; |
658 | my $keepalive = pop; |
525 | |
659 | |
|
|
660 | $state{handle}->destroy if $state{handle}; |
|
|
661 | %state = (); |
|
|
662 | |
|
|
663 | if (defined $_[1]) { |
|
|
664 | $hdr{OrigStatus} = $hdr{Status}; $hdr{Status} = $_[1]; |
|
|
665 | $hdr{OrigReason} = $hdr{Reason}; $hdr{Reason} = $_[2]; |
|
|
666 | } |
|
|
667 | |
|
|
668 | # set-cookie processing |
526 | if ($recurse) { |
669 | if ($arg{cookie_jar}) { |
527 | if ($hdr{Status} =~ /^30[12]$/ && $method ne "POST") { |
670 | for ($hdr{"set-cookie"}) { |
528 | # apparently, mozilla et al. just change POST to GET here |
671 | # parse NAME=VALUE |
529 | # more research is needed before we do the same |
672 | my @kv; |
|
|
673 | |
|
|
674 | while ( |
|
|
675 | m{ |
|
|
676 | \G\s* |
|
|
677 | (?: |
|
|
678 | expires \s*=\s* ([A-Z][a-z][a-z],\ [^,;]+) |
|
|
679 | | ([^=;,[:space:]]+) \s*=\s* (?: "((?:[^\\"]+|\\.)*)" | ([^=;,[:space:]]*) ) |
|
|
680 | ) |
|
|
681 | }gcxsi |
|
|
682 | ) { |
|
|
683 | my $name = $2; |
|
|
684 | my $value = $4; |
|
|
685 | |
|
|
686 | unless (defined $name) { |
|
|
687 | # expires |
|
|
688 | $name = "expires"; |
|
|
689 | $value = $1; |
|
|
690 | } elsif (!defined $value) { |
|
|
691 | # quoted |
|
|
692 | $value = $3; |
|
|
693 | $value =~ s/\\(.)/$1/gs; |
|
|
694 | } |
|
|
695 | |
|
|
696 | push @kv, lc $name, $value; |
|
|
697 | |
|
|
698 | last unless /\G\s*;/gc; |
|
|
699 | } |
|
|
700 | |
|
|
701 | last unless @kv; |
|
|
702 | |
|
|
703 | my $name = shift @kv; |
|
|
704 | my %kv = (value => shift @kv, @kv); |
|
|
705 | |
|
|
706 | $kv{expires} ||= format_date (AE::now + $kv{"max-age"}) |
|
|
707 | if exists $kv{"max-age"}; |
|
|
708 | |
|
|
709 | my $cdom; |
|
|
710 | my $cpath = (delete $kv{path}) || "/"; |
|
|
711 | |
|
|
712 | if (exists $kv{domain}) { |
|
|
713 | $cdom = delete $kv{domain}; |
|
|
714 | |
|
|
715 | $cdom =~ s/^\.?/./; # make sure it starts with a "." |
|
|
716 | |
|
|
717 | next if $cdom =~ /\.$/; |
|
|
718 | |
|
|
719 | # this is not rfc-like and not netscape-like. go figure. |
|
|
720 | my $ndots = $cdom =~ y/.//; |
|
|
721 | next if $ndots < ($cdom =~ /\.[^.][^.]\.[^.][^.]$/ ? 3 : 2); |
|
|
722 | } else { |
|
|
723 | $cdom = $uhost; |
|
|
724 | } |
|
|
725 | |
530 | $redirect = 1; |
726 | # store it |
531 | } elsif ($hdr{Status} == 303) { |
727 | $arg{cookie_jar}{version} = 1; |
532 | # even http/1.1 is unclear on how to mutate the method |
728 | $arg{cookie_jar}{$cdom}{$cpath}{$name} = \%kv; |
533 | $method = "GET" unless $method eq "HEAD"; |
729 | |
534 | $redirect = 1; |
730 | redo if /\G\s*,/gc; |
535 | } elsif ($hdr{Status} == 307 && $method =~ /^(?:GET|HEAD)$/) { |
|
|
536 | $redirect = 1; |
|
|
537 | } |
731 | } |
538 | } |
732 | } |
539 | |
733 | |
540 | my $finish = sub { |
734 | if ($redirect && exists $hdr{location}) { |
541 | $state{handle}->destroy if $state{handle}; |
|
|
542 | %state = (); |
|
|
543 | |
|
|
544 | # set-cookie processing |
|
|
545 | if ($arg{cookie_jar}) { |
|
|
546 | for ($_[1]{"set-cookie"}) { |
|
|
547 | # parse NAME=VALUE |
|
|
548 | my @kv; |
|
|
549 | |
|
|
550 | while (/\G\s* ([^=;,[:space:]]+) \s*=\s* (?: "((?:[^\\"]+|\\.)*)" | ([^=;,[:space:]]*) )/gcxs) { |
|
|
551 | my $name = $1; |
|
|
552 | my $value = $3; |
|
|
553 | |
|
|
554 | unless ($value) { |
|
|
555 | $value = $2; |
|
|
556 | $value =~ s/\\(.)/$1/gs; |
|
|
557 | } |
|
|
558 | |
|
|
559 | push @kv, $name => $value; |
|
|
560 | |
|
|
561 | last unless /\G\s*;/gc; |
|
|
562 | } |
|
|
563 | |
|
|
564 | last unless @kv; |
|
|
565 | |
|
|
566 | my $name = shift @kv; |
|
|
567 | my %kv = (value => shift @kv, @kv); |
|
|
568 | |
|
|
569 | my $cdom; |
|
|
570 | my $cpath = (delete $kv{path}) || "/"; |
|
|
571 | |
|
|
572 | if (exists $kv{domain}) { |
|
|
573 | $cdom = delete $kv{domain}; |
|
|
574 | |
|
|
575 | $cdom =~ s/^\.?/./; # make sure it starts with a "." |
|
|
576 | |
|
|
577 | next if $cdom =~ /\.$/; |
|
|
578 | |
|
|
579 | # this is not rfc-like and not netscape-like. go figure. |
|
|
580 | my $ndots = $cdom =~ y/.//; |
|
|
581 | next if $ndots < ($cdom =~ /\.[^.][^.]\.[^.][^.]$/ ? 3 : 2); |
|
|
582 | } else { |
|
|
583 | $cdom = $uhost; |
|
|
584 | } |
|
|
585 | |
|
|
586 | # store it |
|
|
587 | $arg{cookie_jar}{version} = 1; |
|
|
588 | $arg{cookie_jar}{$cdom}{$cpath}{$name} = \%kv; |
|
|
589 | |
|
|
590 | redo if /\G\s*,/gc; |
|
|
591 | } |
|
|
592 | } |
|
|
593 | |
|
|
594 | if ($redirect) { |
|
|
595 | # we ignore any errors, as it is very common to receive |
735 | # we ignore any errors, as it is very common to receive |
596 | # Content-Length != 0 but no actual body |
736 | # Content-Length != 0 but no actual body |
597 | # we also access %hdr, as $_[1] might be an erro |
737 | # we also access %hdr, as $_[1] might be an erro |
598 | http_request ($method => $hdr{location}, %arg, recurse => $recurse - 1, $cb); |
738 | http_request ( |
599 | } else { |
739 | $method => $hdr{location}, |
600 | $cb->($_[0], $_[1]); |
|
|
601 | } |
740 | %arg, |
|
|
741 | recurse => $recurse - 1, |
|
|
742 | Redirect => [$_[0], \%hdr], |
602 | }; |
743 | $cb); |
603 | |
|
|
604 | my $len = $hdr{"content-length"}; |
|
|
605 | |
|
|
606 | if (!$redirect && $arg{on_header} && !$arg{on_header}(\%hdr)) { |
|
|
607 | $finish->(undef, { Status => 598, Reason => "Request cancelled by on_header", URL => $url }); |
|
|
608 | } elsif ( |
|
|
609 | $hdr{Status} =~ /^(?:1..|[23]04)$/ |
|
|
610 | or $method eq "HEAD" |
|
|
611 | or (defined $len && !$len) |
|
|
612 | ) { |
|
|
613 | # no body |
|
|
614 | $finish->("", \%hdr); |
|
|
615 | } else { |
744 | } else { |
|
|
745 | $cb->($_[0], \%hdr); |
|
|
746 | } |
|
|
747 | }; |
|
|
748 | |
|
|
749 | my $len = $hdr{"content-length"}; |
|
|
750 | |
|
|
751 | if (!$redirect && $arg{on_header} && !$arg{on_header}(\%hdr)) { |
|
|
752 | $finish->(undef, 598 => "Request cancelled by on_header"); |
|
|
753 | } elsif ( |
|
|
754 | $hdr{Status} =~ /^(?:1..|204|205|304)$/ |
|
|
755 | or $method eq "HEAD" |
|
|
756 | or (defined $len && !$len) |
|
|
757 | ) { |
|
|
758 | # no body |
|
|
759 | $finish->("", undef, undef, 1); |
|
|
760 | } else { |
616 | # body handling, four different code paths |
761 | # body handling, many different code paths |
617 | # for want_body_handle, on_body (2x), normal (2x) |
762 | # - no body expected |
618 | # we might read too much here, but it does not matter yet (no pers. connections) |
763 | # - want_body_handle |
|
|
764 | # - te chunked |
|
|
765 | # - 2x length known (with or without on_body) |
|
|
766 | # - 2x length not known (with or without on_body) |
619 | if (!$redirect && $arg{want_body_handle}) { |
767 | if (!$redirect && $arg{want_body_handle}) { |
620 | $_[0]->on_eof (undef); |
768 | $_[0]->on_eof (undef); |
621 | $_[0]->on_error (undef); |
769 | $_[0]->on_error (undef); |
622 | $_[0]->on_read (undef); |
770 | $_[0]->on_read (undef); |
623 | |
771 | |
624 | $finish->(delete $state{handle}, \%hdr); |
772 | $finish->(delete $state{handle}); |
625 | |
773 | |
626 | } elsif ($arg{on_body}) { |
774 | } elsif ($hdr{"transfer-encoding"} =~ /\bchunked\b/i) { |
|
|
775 | my $cl = 0; |
|
|
776 | my $body = undef; |
|
|
777 | my $on_body = $arg{on_body} || sub { $body .= shift; 1 }; |
|
|
778 | |
627 | $_[0]->on_error (sub { $finish->(undef, { Status => 599, Reason => $_[2], URL => $url }) }); |
779 | $_[0]->on_error (sub { $finish->(undef, 599 => $_[2]) }); |
|
|
780 | |
|
|
781 | my $read_chunk; $read_chunk = sub { |
|
|
782 | $_[1] =~ /^([0-9a-fA-F]+)/ |
|
|
783 | or $finish->(undef, 599 => "Garbled chunked transfer encoding"); |
|
|
784 | |
|
|
785 | my $len = hex $1; |
|
|
786 | |
628 | if ($len) { |
787 | if ($len) { |
629 | $_[0]->on_eof (undef); |
788 | $cl += $len; |
|
|
789 | |
|
|
790 | $_[0]->push_read (chunk => $len, sub { |
|
|
791 | $on_body->($_[1], \%hdr) |
|
|
792 | or return $finish->(undef, 598 => "Request cancelled by on_body"); |
|
|
793 | |
630 | $_[0]->on_read (sub { |
794 | $_[0]->push_read (line => sub { |
631 | $len -= length $_[0]{rbuf}; |
795 | length $_[1] |
632 | |
796 | and return $finish->(undef, 599 => "Garbled chunked transfer encoding"); |
633 | $arg{on_body}(delete $_[0]{rbuf}, \%hdr) |
797 | $_[0]->push_read (line => $read_chunk); |
634 | or $finish->(undef, { Status => 598, Reason => "Request cancelled by on_body", URL => $url }); |
|
|
635 | |
|
|
636 | $len > 0 |
798 | }); |
637 | or $finish->("", \%hdr); |
|
|
638 | }); |
799 | }); |
639 | } else { |
800 | } else { |
640 | $_[0]->on_eof (sub { |
801 | $hdr{"content-length"} ||= $cl; |
641 | $finish->("", \%hdr); |
802 | |
|
|
803 | $_[0]->push_read (line => $qr_nlnl, sub { |
|
|
804 | if (length $_[1]) { |
|
|
805 | for ("$_[1]") { |
|
|
806 | y/\015//d; # weed out any \015, as they show up in the weirdest of places. |
|
|
807 | |
|
|
808 | my $hdr = parse_hdr |
|
|
809 | or return $finish->(undef, 599 => "Garbled response trailers"); |
|
|
810 | |
|
|
811 | %hdr = (%hdr, %$hdr); |
|
|
812 | } |
642 | }); |
813 | } |
643 | $_[0]->on_read (sub { |
814 | |
644 | $arg{on_body}(delete $_[0]{rbuf}, \%hdr) |
815 | $finish->($body, undef, undef, 1); |
645 | or $finish->(undef, { Status => 598, Reason => "Request cancelled by on_body", URL => $url }); |
|
|
646 | }); |
816 | }); |
647 | } |
817 | } |
|
|
818 | }; |
|
|
819 | |
|
|
820 | $_[0]->push_read (line => $read_chunk); |
|
|
821 | |
|
|
822 | } elsif ($arg{on_body}) { |
|
|
823 | $_[0]->on_error (sub { $finish->(undef, 599 => $_[2]) }); |
|
|
824 | |
|
|
825 | if ($len) { |
|
|
826 | $_[0]->on_read (sub { |
|
|
827 | $len -= length $_[0]{rbuf}; |
|
|
828 | |
|
|
829 | $arg{on_body}(delete $_[0]{rbuf}, \%hdr) |
|
|
830 | or return $finish->(undef, 598 => "Request cancelled by on_body"); |
|
|
831 | |
|
|
832 | $len > 0 |
|
|
833 | or $finish->("", undef, undef, 1); |
|
|
834 | }); |
648 | } else { |
835 | } else { |
649 | $_[0]->on_eof (undef); |
836 | $_[0]->on_eof (sub { |
650 | |
837 | $finish->(""); |
651 | if ($len) { |
838 | }); |
652 | $_[0]->on_error (sub { $finish->(undef, { Status => 599, Reason => $_[2], URL => $url }) }); |
|
|
653 | $_[0]->on_read (sub { |
839 | $_[0]->on_read (sub { |
|
|
840 | $arg{on_body}(delete $_[0]{rbuf}, \%hdr) |
|
|
841 | or $finish->(undef, 598 => "Request cancelled by on_body"); |
|
|
842 | }); |
|
|
843 | } |
|
|
844 | } else { |
|
|
845 | $_[0]->on_eof (undef); |
|
|
846 | |
|
|
847 | if ($len) { |
|
|
848 | $_[0]->on_error (sub { $finish->(undef, 599 => $_[2]) }); |
|
|
849 | $_[0]->on_read (sub { |
654 | $finish->((substr delete $_[0]{rbuf}, 0, $len, ""), \%hdr) |
850 | $finish->((substr delete $_[0]{rbuf}, 0, $len, ""), undef, undef, 1) |
655 | if $len <= length $_[0]{rbuf}; |
851 | if $len <= length $_[0]{rbuf}; |
656 | }); |
852 | }); |
657 | } else { |
853 | } else { |
658 | $_[0]->on_error (sub { |
854 | $_[0]->on_error (sub { |
659 | $! == Errno::EPIPE |
855 | ($! == Errno::EPIPE || !$!) |
660 | ? $finish->(delete $_[0]{rbuf}, \%hdr) |
856 | ? $finish->(delete $_[0]{rbuf}) |
661 | : $finish->(undef, { Status => 599, Reason => $_[2], URL => $url }); |
857 | : $finish->(undef, 599 => $_[2]); |
662 | }); |
858 | }); |
663 | $_[0]->on_read (sub { }); |
859 | $_[0]->on_read (sub { }); |
664 | } |
|
|
665 | } |
860 | } |
666 | } |
861 | } |
667 | }); |
862 | } |
668 | }); |
863 | }; |
|
|
864 | |
|
|
865 | $state{handle}->push_read (line => $qr_nlnl, $state{read_response}); |
669 | }; |
866 | }; |
670 | |
867 | |
671 | # now handle proxy-CONNECT method |
868 | # now handle proxy-CONNECT method |
672 | if ($proxy && $uscheme eq "https") { |
869 | if ($proxy && $uscheme eq "https") { |
673 | # oh dear, we have to wrap it into a connect request |
870 | # oh dear, we have to wrap it into a connect request |
674 | |
871 | |
675 | # maybe re-use $uauthority with patched port? |
872 | # maybe re-use $uauthority with patched port? |
676 | $state{handle}->push_write ("CONNECT $uhost:$uport HTTP/1.0\015\012Host: $uhost\015\012\015\012"); |
873 | $state{handle}->push_write ("CONNECT $uhost:$uport HTTP/1.0\015\012Host: $uhost\015\012\015\012"); |
677 | $state{handle}->push_read (line => $qr_nlnl, sub { |
874 | $state{handle}->push_read (line => $qr_nlnl, sub { |
678 | $_[1] =~ /^HTTP\/([0-9\.]+) \s+ ([0-9]{3}) (?: \s+ ([^\015\012]*) )?/ix |
875 | $_[1] =~ /^HTTP\/([0-9\.]+) \s+ ([0-9]{3}) (?: \s+ ([^\015\012]*) )?/ix |
679 | or return (%state = (), $cb->(undef, { Status => 599, Reason => "Invalid proxy connect response ($_[1])", URL => $url })); |
876 | or return (%state = (), $cb->(undef, { @pseudo, Status => 599, Reason => "Invalid proxy connect response ($_[1])" })); |
680 | |
877 | |
681 | if ($2 == 200) { |
878 | if ($2 == 200) { |
682 | $rpath = $upath; |
879 | $rpath = $upath; |
683 | &$handle_actual_request; |
880 | &$handle_actual_request; |
684 | } else { |
881 | } else { |
685 | %state = (); |
882 | %state = (); |
686 | $cb->(undef, { Status => $2, Reason => $3, URL => $url }); |
883 | $cb->(undef, { @pseudo, Status => $2, Reason => $3 }); |
687 | } |
884 | } |
688 | }); |
885 | }); |
689 | } else { |
886 | } else { |
690 | &$handle_actual_request; |
887 | &$handle_actual_request; |
691 | } |
888 | } |
692 | |
|
|
693 | }, sub { |
|
|
694 | $timeout |
|
|
695 | }; |
889 | }; |
|
|
890 | |
|
|
891 | my $tcp_connect = $arg{tcp_connect} |
|
|
892 | || do { require AnyEvent::Socket; \&AnyEvent::Socket::tcp_connect }; |
|
|
893 | |
|
|
894 | $state{connect_guard} = $tcp_connect->($rhost, $rport, $connect_cb, $arg{on_prepare} || sub { $timeout }); |
|
|
895 | |
696 | }; |
896 | }; |
697 | |
897 | |
698 | defined wantarray && AnyEvent::Util::guard { %state = () } |
898 | defined wantarray && AnyEvent::Util::guard { %state = () } |
699 | } |
899 | } |
700 | |
900 | |
… | |
… | |
714 | &http_request |
914 | &http_request |
715 | } |
915 | } |
716 | |
916 | |
717 | =back |
917 | =back |
718 | |
918 | |
|
|
919 | =head2 DNS CACHING |
|
|
920 | |
|
|
921 | AnyEvent::HTTP uses the AnyEvent::Socket::tcp_connect function for |
|
|
922 | the actual connection, which in turn uses AnyEvent::DNS to resolve |
|
|
923 | hostnames. The latter is a simple stub resolver and does no caching |
|
|
924 | on its own. If you want DNS caching, you currently have to provide |
|
|
925 | your own default resolver (by storing a suitable resolver object in |
|
|
926 | C<$AnyEvent::DNS::RESOLVER>). |
|
|
927 | |
719 | =head2 GLOBAL FUNCTIONS AND VARIABLES |
928 | =head2 GLOBAL FUNCTIONS AND VARIABLES |
720 | |
929 | |
721 | =over 4 |
930 | =over 4 |
722 | |
931 | |
723 | =item AnyEvent::HTTP::set_proxy "proxy-url" |
932 | =item AnyEvent::HTTP::set_proxy "proxy-url" |
724 | |
933 | |
725 | Sets the default proxy server to use. The proxy-url must begin with a |
934 | Sets the default proxy server to use. The proxy-url must begin with a |
726 | string of the form C<http://host:port> (optionally C<https:...>). |
935 | string of the form C<http://host:port> (optionally C<https:...>), croaks |
|
|
936 | otherwise. |
|
|
937 | |
|
|
938 | To clear an already-set proxy, use C<undef>. |
|
|
939 | |
|
|
940 | =item $date = AnyEvent::HTTP::format_date $timestamp |
|
|
941 | |
|
|
942 | Takes a POSIX timestamp (seconds since the epoch) and formats it as a HTTP |
|
|
943 | Date (RFC 2616). |
|
|
944 | |
|
|
945 | =item $timestamp = AnyEvent::HTTP::parse_date $date |
|
|
946 | |
|
|
947 | Takes a HTTP Date (RFC 2616) or a Cookie date (netscape cookie spec) and |
|
|
948 | returns the corresponding POSIX timestamp, or C<undef> if the date cannot |
|
|
949 | be parsed. |
727 | |
950 | |
728 | =item $AnyEvent::HTTP::MAX_RECURSE |
951 | =item $AnyEvent::HTTP::MAX_RECURSE |
729 | |
952 | |
730 | The default value for the C<recurse> request parameter (default: C<10>). |
953 | The default value for the C<recurse> request parameter (default: C<10>). |
731 | |
954 | |
732 | =item $AnyEvent::HTTP::USERAGENT |
955 | =item $AnyEvent::HTTP::USERAGENT |
733 | |
956 | |
734 | The default value for the C<User-Agent> header (the default is |
957 | The default value for the C<User-Agent> header (the default is |
735 | C<Mozilla/5.0 (compatible; U; AnyEvent-HTTP/$VERSION; +http://software.schmorp.de/pkg/AnyEvent)>). |
958 | C<Mozilla/5.0 (compatible; U; AnyEvent-HTTP/$VERSION; +http://software.schmorp.de/pkg/AnyEvent)>). |
736 | |
959 | |
737 | =item $AnyEvent::HTTP::MAX_PERSISTENT |
960 | =item $AnyEvent::HTTP::MAX_PER_HOST |
738 | |
961 | |
739 | The maximum number of persistent connections to keep open (default: 8). |
962 | The maximum number of concurrent connections to the same host (identified |
|
|
963 | by the hostname). If the limit is exceeded, then the additional requests |
|
|
964 | are queued until previous connections are closed. |
740 | |
965 | |
741 | Not implemented currently. |
966 | The default value for this is C<4>, and it is highly advisable to not |
742 | |
967 | increase it. |
743 | =item $AnyEvent::HTTP::PERSISTENT_TIMEOUT |
|
|
744 | |
|
|
745 | The maximum time to cache a persistent connection, in seconds (default: 2). |
|
|
746 | |
|
|
747 | Not implemented currently. |
|
|
748 | |
968 | |
749 | =item $AnyEvent::HTTP::ACTIVE |
969 | =item $AnyEvent::HTTP::ACTIVE |
750 | |
970 | |
751 | The number of active connections. This is not the number of currently |
971 | The number of active connections. This is not the number of currently |
752 | running requests, but the number of currently open and non-idle TCP |
972 | running requests, but the number of currently open and non-idle TCP |
… | |
… | |
754 | |
974 | |
755 | =back |
975 | =back |
756 | |
976 | |
757 | =cut |
977 | =cut |
758 | |
978 | |
|
|
979 | our @month = qw(Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec); |
|
|
980 | our @weekday = qw(Sun Mon Tue Wed Thu Fri Sat); |
|
|
981 | |
|
|
982 | sub format_date($) { |
|
|
983 | my ($time) = @_; |
|
|
984 | |
|
|
985 | # RFC 822/1123 format |
|
|
986 | my ($S, $M, $H, $mday, $mon, $year, $wday, $yday, undef) = gmtime $time; |
|
|
987 | |
|
|
988 | sprintf "%s, %02d %s %04d %02d:%02d:%02d GMT", |
|
|
989 | $weekday[$wday], $mday, $month[$mon], $year + 1900, |
|
|
990 | $H, $M, $S; |
|
|
991 | } |
|
|
992 | |
|
|
993 | sub parse_date($) { |
|
|
994 | my ($date) = @_; |
|
|
995 | |
|
|
996 | my ($d, $m, $y, $H, $M, $S); |
|
|
997 | |
|
|
998 | if ($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$/) { |
|
|
999 | # RFC 822/1123, required by RFC 2616 (with " ") |
|
|
1000 | # cookie dates (with "-") |
|
|
1001 | |
|
|
1002 | ($d, $m, $y, $H, $M, $S) = ($1, $2, $3, $4, $5, $6); |
|
|
1003 | |
|
|
1004 | } elsif ($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$/) { |
|
|
1005 | # RFC 850 |
|
|
1006 | ($d, $m, $y, $H, $M, $S) = ($1, $2, $3 < 69 ? $3 + 2000 : $3 + 1900, $4, $5, $6); |
|
|
1007 | |
|
|
1008 | } elsif ($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])$/) { |
|
|
1009 | # ISO C's asctime |
|
|
1010 | ($d, $m, $y, $H, $M, $S) = ($2, $1, $6, $3, $4, $5); |
|
|
1011 | } |
|
|
1012 | # other formats fail in the loop below |
|
|
1013 | |
|
|
1014 | for (0..11) { |
|
|
1015 | if ($m eq $month[$_]) { |
|
|
1016 | require Time::Local; |
|
|
1017 | return Time::Local::timegm ($S, $M, $H, $d, $_, $y); |
|
|
1018 | } |
|
|
1019 | } |
|
|
1020 | |
|
|
1021 | undef |
|
|
1022 | } |
|
|
1023 | |
759 | sub set_proxy($) { |
1024 | sub set_proxy($) { |
|
|
1025 | if (length $_[0]) { |
760 | $PROXY = [$2, $3 || 3128, $1] if $_[0] =~ m%^(https?):// ([^:/]+) (?: : (\d*) )?%ix; |
1026 | $_[0] =~ m%^(https?):// ([^:/]+) (?: : (\d*) )?%ix |
|
|
1027 | or Carp::croak "$_[0]: invalid proxy URL"; |
|
|
1028 | $PROXY = [$2, $3 || 3128, $1] |
|
|
1029 | } else { |
|
|
1030 | undef $PROXY; |
|
|
1031 | } |
761 | } |
1032 | } |
762 | |
1033 | |
763 | # initialise proxy from environment |
1034 | # initialise proxy from environment |
|
|
1035 | eval { |
764 | set_proxy $ENV{http_proxy}; |
1036 | set_proxy $ENV{http_proxy}; |
|
|
1037 | }; |
|
|
1038 | |
|
|
1039 | =head2 SOCKS PROXIES |
|
|
1040 | |
|
|
1041 | Socks proxies are not directly supported by AnyEvent::HTTP. You can |
|
|
1042 | compile your perl to support socks, or use an external program such as |
|
|
1043 | F<socksify> (dante) or F<tsocks> to make your program use a socks proxy |
|
|
1044 | transparently. |
|
|
1045 | |
|
|
1046 | Alternatively, for AnyEvent::HTTP only, you can use your own |
|
|
1047 | C<tcp_connect> function that does the proxy handshake - here is an example |
|
|
1048 | that works with socks4a proxies: |
|
|
1049 | |
|
|
1050 | use Errno; |
|
|
1051 | use AnyEvent::Util; |
|
|
1052 | use AnyEvent::Socket; |
|
|
1053 | use AnyEvent::Handle; |
|
|
1054 | |
|
|
1055 | # host, port and username of/for your socks4a proxy |
|
|
1056 | my $socks_host = "10.0.0.23"; |
|
|
1057 | my $socks_port = 9050; |
|
|
1058 | my $socks_user = ""; |
|
|
1059 | |
|
|
1060 | sub socks4a_connect { |
|
|
1061 | my ($host, $port, $connect_cb, $prepare_cb) = @_; |
|
|
1062 | |
|
|
1063 | my $hdl = new AnyEvent::Handle |
|
|
1064 | connect => [$socks_host, $socks_port], |
|
|
1065 | on_prepare => sub { $prepare_cb->($_[0]{fh}) }, |
|
|
1066 | on_error => sub { $connect_cb->() }, |
|
|
1067 | ; |
|
|
1068 | |
|
|
1069 | $hdl->push_write (pack "CCnNZ*Z*", 4, 1, $port, 1, $socks_user, $host); |
|
|
1070 | |
|
|
1071 | $hdl->push_read (chunk => 8, sub { |
|
|
1072 | my ($hdl, $chunk) = @_; |
|
|
1073 | my ($status, $port, $ipn) = unpack "xCna4", $chunk; |
|
|
1074 | |
|
|
1075 | if ($status == 0x5a) { |
|
|
1076 | $connect_cb->($hdl->{fh}, (format_address $ipn) . ":$port"); |
|
|
1077 | } else { |
|
|
1078 | $! = Errno::ENXIO; $connect_cb->(); |
|
|
1079 | } |
|
|
1080 | }); |
|
|
1081 | |
|
|
1082 | $hdl |
|
|
1083 | } |
|
|
1084 | |
|
|
1085 | Use C<socks4a_connect> instead of C<tcp_connect> when doing C<http_request>s, |
|
|
1086 | possibly after switching off other proxy types: |
|
|
1087 | |
|
|
1088 | AnyEvent::HTTP::set_proxy undef; # usually you do not want other proxies |
|
|
1089 | |
|
|
1090 | http_get 'http://www.google.com', tcp_connect => \&socks4a_connect, sub { |
|
|
1091 | my ($data, $headers) = @_; |
|
|
1092 | ... |
|
|
1093 | }; |
765 | |
1094 | |
766 | =head1 SEE ALSO |
1095 | =head1 SEE ALSO |
767 | |
1096 | |
768 | L<AnyEvent>. |
1097 | L<AnyEvent>. |
769 | |
1098 | |