… | |
… | |
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 = '1.01'; |
53 | our $VERSION = '1.05'; |
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 $USERAGENT = "Mozilla/5.0 (compatible; AnyEvent::HTTP/$VERSION; +http://software.schmorp.de/pkg/AnyEvent)"; |
58 | our $MAX_RECURSE = 10; |
58 | our $MAX_RECURSE = 10; |
… | |
… | |
71 | my %CO_SLOT; # number of open connections, and wait queue, per host |
71 | my %CO_SLOT; # number of open connections, and wait queue, per host |
72 | |
72 | |
73 | =item http_get $url, key => value..., $cb->($data, $headers) |
73 | =item http_get $url, key => value..., $cb->($data, $headers) |
74 | |
74 | |
75 | Executes an HTTP-GET request. See the http_request function for details on |
75 | Executes an HTTP-GET request. See the http_request function for details on |
76 | additional parameters. |
76 | additional parameters and the return value. |
77 | |
77 | |
78 | =item http_head $url, key => value..., $cb->($data, $headers) |
78 | =item http_head $url, key => value..., $cb->($data, $headers) |
79 | |
79 | |
80 | Executes an HTTP-HEAD request. See the http_request function for details on |
80 | Executes an HTTP-HEAD request. See the http_request function for details |
81 | additional parameters. |
81 | on additional parameters and the return value. |
82 | |
82 | |
83 | =item http_post $url, $body, key => value..., $cb->($data, $headers) |
83 | =item http_post $url, $body, key => value..., $cb->($data, $headers) |
84 | |
84 | |
85 | Executes an HTTP-POST request with a request body of C<$bod>. See the |
85 | Executes an HTTP-POST request with a request body of C<$body>. See the |
86 | http_request function for details on additional parameters. |
86 | http_request function for details on additional parameters and the return |
|
|
87 | value. |
87 | |
88 | |
88 | =item http_request $method => $url, key => value..., $cb->($data, $headers) |
89 | =item http_request $method => $url, key => value..., $cb->($data, $headers) |
89 | |
90 | |
90 | Executes a HTTP request of type C<$method> (e.g. C<GET>, C<POST>). The URL |
91 | Executes a HTTP request of type C<$method> (e.g. C<GET>, C<POST>). The URL |
91 | must be an absolute http or https URL. |
92 | must be an absolute http or https URL. |
|
|
93 | |
|
|
94 | When called in void context, nothing is returned. In other contexts, |
|
|
95 | C<http_request> returns a "cancellation guard" - you have to keep the |
|
|
96 | object at least alive until the callback get called. If the object gets |
|
|
97 | destroyed before the callbakc is called, the request will be cancelled. |
92 | |
98 | |
93 | The callback will be called with the response data as first argument |
99 | The callback will be called with the response data as first argument |
94 | (or C<undef> if it wasn't available due to errors), and a hash-ref with |
100 | (or C<undef> if it wasn't available due to errors), and a hash-ref with |
95 | response headers as second argument. |
101 | response headers as second argument. |
96 | |
102 | |
97 | All the headers in that hash are lowercased. In addition to the response |
103 | All the headers in that hash are lowercased. In addition to the response |
98 | headers, the three "pseudo-headers" C<HTTPVersion>, C<Status> and |
104 | headers, the "pseudo-headers" C<HTTPVersion>, C<Status> and C<Reason> |
99 | C<Reason> contain the three parts of the HTTP Status-Line of the same |
105 | contain the three parts of the HTTP Status-Line of the same name. The |
|
|
106 | pseudo-header C<URL> contains the original URL (which can differ from the |
|
|
107 | requested URL when following redirects). |
|
|
108 | |
100 | name. If the server sends a header multiple lines, then their contents |
109 | If the server sends a header multiple lines, then their contents will be |
101 | will be joined together with C<\x00>. |
110 | joined together with C<\x00>. |
102 | |
111 | |
103 | If an internal error occurs, such as not being able to resolve a hostname, |
112 | If an internal error occurs, such as not being able to resolve a hostname, |
104 | then C<$data> will be C<undef>, C<< $headers->{Status} >> will be C<599> |
113 | then C<$data> will be C<undef>, C<< $headers->{Status} >> will be C<599> |
105 | and the C<Reason> pseudo-header will contain an error message. |
114 | and the C<Reason> pseudo-header will contain an error message. |
106 | |
115 | |
… | |
… | |
185 | my ($body, $hdr) = @_; |
194 | my ($body, $hdr) = @_; |
186 | use Data::Dumper; |
195 | use Data::Dumper; |
187 | print Dumper $hdr; |
196 | print Dumper $hdr; |
188 | } |
197 | } |
189 | ; |
198 | ; |
|
|
199 | |
|
|
200 | Example: make another simple HTTP GET request, but immediately try to |
|
|
201 | cancel it. |
|
|
202 | |
|
|
203 | my $request = http_request GET => "http://www.nethype.de/", sub { |
|
|
204 | my ($body, $hdr) = @_; |
|
|
205 | print "$body\n"; |
|
|
206 | }; |
|
|
207 | |
|
|
208 | undef $request; |
190 | |
209 | |
191 | =cut |
210 | =cut |
192 | |
211 | |
193 | sub _slot_schedule; |
212 | sub _slot_schedule; |
194 | sub _slot_schedule($) { |
213 | sub _slot_schedule($) { |
… | |
… | |
232 | while (my ($k, $v) = each %$hdr) { |
251 | while (my ($k, $v) = each %$hdr) { |
233 | $hdr{lc $k} = $v; |
252 | $hdr{lc $k} = $v; |
234 | } |
253 | } |
235 | } |
254 | } |
236 | |
255 | |
237 | my $recurse = exists $arg{recurse} ? $arg{recurse} : $MAX_RECURSE; |
256 | my $recurse = exists $arg{recurse} ? delete $arg{recurse} : $MAX_RECURSE; |
238 | |
257 | |
239 | return $cb->(undef, { Status => 599, Reason => "recursion limit reached" }) |
258 | return $cb->(undef, { Status => 599, Reason => "recursion limit reached", URL => $url }) |
240 | if $recurse < 0; |
259 | if $recurse < 0; |
241 | |
260 | |
242 | my $proxy = $arg{proxy} || $PROXY; |
261 | my $proxy = $arg{proxy} || $PROXY; |
243 | my $timeout = $arg{timeout} || $TIMEOUT; |
262 | my $timeout = $arg{timeout} || $TIMEOUT; |
244 | |
263 | |
… | |
… | |
249 | |
268 | |
250 | $scheme = lc $scheme; |
269 | $scheme = lc $scheme; |
251 | |
270 | |
252 | my $uport = $scheme eq "http" ? 80 |
271 | my $uport = $scheme eq "http" ? 80 |
253 | : $scheme eq "https" ? 443 |
272 | : $scheme eq "https" ? 443 |
254 | : return $cb->(undef, { Status => 599, Reason => "only http and https URL schemes supported" }); |
273 | : return $cb->(undef, { Status => 599, Reason => "only http and https URL schemes supported", URL => $url }); |
255 | |
274 | |
256 | $hdr{referer} ||= "$scheme://$authority$upath"; # leave out fragment and query string, just a heuristic |
275 | $hdr{referer} ||= "$scheme://$authority$upath"; # leave out fragment and query string, just a heuristic |
257 | |
276 | |
258 | $authority =~ /^(?: .*\@ )? ([^\@:]+) (?: : (\d+) )?$/x |
277 | $authority =~ /^(?: .*\@ )? ([^\@:]+) (?: : (\d+) )?$/x |
259 | or return $cb->(undef, { Status => 599, Reason => "unparsable URL" }); |
278 | or return $cb->(undef, { Status => 599, Reason => "unparsable URL", URL => $url }); |
260 | |
279 | |
261 | my $uhost = $1; |
280 | my $uhost = $1; |
262 | $uport = $2 if defined $2; |
281 | $uport = $2 if defined $2; |
263 | |
282 | |
264 | $uhost =~ s/^\[(.*)\]$/$1/; |
283 | $uhost =~ s/^\[(.*)\]$/$1/; |
… | |
… | |
271 | %$jar = () if $jar->{version} < 1; |
290 | %$jar = () if $jar->{version} < 1; |
272 | |
291 | |
273 | my @cookie; |
292 | my @cookie; |
274 | |
293 | |
275 | while (my ($chost, $v) = each %$jar) { |
294 | while (my ($chost, $v) = each %$jar) { |
|
|
295 | if ($chost =~ /^\./) { |
276 | next unless $chost eq substr $uhost, -length $chost; |
296 | next unless $chost eq substr $uhost, -length $chost; |
277 | next unless $chost =~ /^\./; |
297 | } elsif ($chost =~ /\./) { |
|
|
298 | next unless $chost eq $uhost; |
|
|
299 | } else { |
|
|
300 | next; |
|
|
301 | } |
278 | |
302 | |
279 | while (my ($cpath, $v) = each %$v) { |
303 | while (my ($cpath, $v) = each %$v) { |
280 | next unless $cpath eq substr $upath, 0, length $cpath; |
304 | next unless $cpath eq substr $upath, 0, length $cpath; |
281 | |
305 | |
282 | while (my ($k, $v) = each %$v) { |
306 | while (my ($k, $v) = each %$v) { |
… | |
… | |
309 | |
333 | |
310 | return unless $state{connect_guard}; |
334 | return unless $state{connect_guard}; |
311 | |
335 | |
312 | $state{connect_guard} = AnyEvent::Socket::tcp_connect $rhost, $rport, sub { |
336 | $state{connect_guard} = AnyEvent::Socket::tcp_connect $rhost, $rport, sub { |
313 | $state{fh} = shift |
337 | $state{fh} = shift |
314 | or return $cb->(undef, { Status => 599, Reason => "$!" }); |
338 | or return $cb->(undef, { Status => 599, Reason => "$!", URL => $url }); |
315 | |
339 | |
316 | delete $state{connect_guard}; # reduce memory usage, save a tree |
340 | delete $state{connect_guard}; # reduce memory usage, save a tree |
317 | |
341 | |
318 | # get handle |
342 | # get handle |
319 | $state{handle} = new AnyEvent::Handle |
343 | $state{handle} = new AnyEvent::Handle |
… | |
… | |
333 | # (re-)configure handle |
357 | # (re-)configure handle |
334 | $state{handle}->timeout ($timeout); |
358 | $state{handle}->timeout ($timeout); |
335 | $state{handle}->on_error (sub { |
359 | $state{handle}->on_error (sub { |
336 | my $errno = "$!"; |
360 | my $errno = "$!"; |
337 | %state = (); |
361 | %state = (); |
338 | $cb->(undef, { Status => 599, Reason => $errno }); |
362 | $cb->(undef, { Status => 599, Reason => $errno, URL => $url }); |
339 | }); |
363 | }); |
340 | $state{handle}->on_eof (sub { |
364 | $state{handle}->on_eof (sub { |
341 | %state = (); |
365 | %state = (); |
342 | $cb->(undef, { Status => 599, Reason => "unexpected end-of-file" }); |
366 | $cb->(undef, { Status => 599, Reason => "unexpected end-of-file", URL => $url }); |
343 | }); |
367 | }); |
344 | |
368 | |
345 | # send request |
369 | # send request |
346 | $state{handle}->push_write ( |
370 | $state{handle}->push_write ( |
347 | "$method $rpath HTTP/1.0\015\012" |
371 | "$method $rpath HTTP/1.0\015\012" |
… | |
… | |
352 | |
376 | |
353 | %hdr = (); # reduce memory usage, save a kitten |
377 | %hdr = (); # reduce memory usage, save a kitten |
354 | |
378 | |
355 | # status line |
379 | # status line |
356 | $state{handle}->push_read (line => qr/\015?\012/, sub { |
380 | $state{handle}->push_read (line => qr/\015?\012/, sub { |
357 | $_[1] =~ /^HTTP\/([0-9\.]+) \s+ ([0-9]{3}) \s+ ([^\015\012]+)/ix |
381 | $_[1] =~ /^HTTP\/([0-9\.]+) \s+ ([0-9]{3}) (?: \s+ ([^\015\012]*) )?/ix |
358 | or return (%state = (), $cb->(undef, { Status => 599, Reason => "invalid server response ($_[1])" })); |
382 | or return (%state = (), $cb->(undef, { Status => 599, Reason => "invalid server response ($_[1])", URL => $url })); |
359 | |
383 | |
360 | my %hdr = ( # response headers |
384 | my %hdr = ( # response headers |
361 | HTTPVersion => "\x00$1", |
385 | HTTPVersion => "\x00$1", |
362 | Status => "\x00$2", |
386 | Status => "\x00$2", |
363 | Reason => "\x00$3", |
387 | Reason => "\x00$3", |
|
|
388 | URL => "\x00$url" |
364 | ); |
389 | ); |
365 | |
390 | |
366 | # headers, could be optimized a bit |
391 | # headers, could be optimized a bit |
367 | $state{handle}->unshift_read (line => qr/\015?\012\015?\012/, sub { |
392 | $state{handle}->unshift_read (line => qr/\015?\012\015?\012/, sub { |
368 | for ("$_[1]\012") { |
393 | for ("$_[1]\012") { |
369 | # we support spaces in field names, as lotus domino |
394 | # we support spaces in field names, as lotus domino |
370 | # creates them. |
395 | # creates them (actually spaces around seperators |
|
|
396 | # are strictly allowed in http, they are a security issue). |
371 | $hdr{lc $1} .= "\x00$2" |
397 | $hdr{lc $1} .= "\x00$2" |
372 | while /\G |
398 | while /\G |
373 | ([^:\000-\037]+): |
399 | ([^:\000-\037]+): |
374 | [\011\040]* |
400 | [\011\040]* |
375 | ((?: [^\015\012]+ | \015?\012[\011\040] )*) |
401 | ((?: [^\015\012]+ | \015?\012[\011\040] )*) |
376 | \015?\012 |
402 | \015?\012 |
377 | /gxc; |
403 | /gxc; |
378 | |
404 | |
379 | /\G$/ |
405 | /\G$/ |
380 | or return (%state = (), $cb->(undef, { Status => 599, Reason => "garbled response headers" })); |
406 | or return (%state = (), $cb->(undef, { Status => 599, Reason => "garbled response headers", URL => $url })); |
381 | } |
407 | } |
382 | |
408 | |
383 | substr $_, 0, 1, "" |
409 | substr $_, 0, 1, "" |
384 | for values %hdr; |
410 | for values %hdr; |
385 | |
411 | |
386 | my $finish = sub { |
412 | my $finish = sub { |
|
|
413 | # TODO: use destroy method, when/if available |
|
|
414 | #$state{handle}->destroy; |
|
|
415 | $state{handle}->on_eof (undef); |
|
|
416 | $state{handle}->on_error (undef); |
387 | %state = (); |
417 | %state = (); |
388 | |
418 | |
389 | # set-cookie processing |
419 | # set-cookie processing |
390 | if ($arg{cookie_jar} && exists $hdr{"set-cookie"}) { |
420 | if ($arg{cookie_jar} && exists $hdr{"set-cookie"}) { |
391 | for (split /\x00/, $hdr{"set-cookie"}) { |
421 | for (split /\x00/, $hdr{"set-cookie"}) { |
392 | my ($cookie, @arg) = split /;\s*/; |
422 | my ($cookie, @arg) = split /;\s*/; |
393 | my ($name, $value) = split /=/, $cookie, 2; |
423 | my ($name, $value) = split /=/, $cookie, 2; |
394 | my %kv = (value => $value, map { split /=/, $_, 2 } @arg); |
424 | my %kv = (value => $value, map { split /=/, $_, 2 } @arg); |
395 | |
425 | |
396 | my $cdom = (delete $kv{domain}) || $uhost; |
426 | my $cdom; |
397 | my $cpath = (delete $kv{path}) || "/"; |
427 | my $cpath = (delete $kv{path}) || "/"; |
|
|
428 | |
|
|
429 | if (exists $kv{domain}) { |
|
|
430 | $cdom = delete $kv{domain}; |
398 | |
431 | |
399 | $cdom =~ s/^.?/./; # make sure it starts with a "." |
432 | $cdom =~ s/^\.?/./; # make sure it starts with a "." |
400 | |
433 | |
401 | next if $cdom =~ /\.$/; |
434 | next if $cdom =~ /\.$/; |
402 | |
435 | |
403 | # this is not rfc-like and not netscape-like. go figure. |
436 | # this is not rfc-like and not netscape-like. go figure. |
404 | my $ndots = $cdom =~ y/.//; |
437 | my $ndots = $cdom =~ y/.//; |
405 | next if $ndots < ($cdom =~ /\.[^.][^.]\.[^.][^.]$/ ? 3 : 2); |
438 | next if $ndots < ($cdom =~ /\.[^.][^.]\.[^.][^.]$/ ? 3 : 2); |
|
|
439 | } else { |
|
|
440 | $cdom = $uhost; |
|
|
441 | } |
406 | |
442 | |
407 | # store it |
443 | # store it |
408 | $arg{cookie_jar}{version} = 1; |
444 | $arg{cookie_jar}{version} = 1; |
409 | $arg{cookie_jar}{$cdom}{$cpath}{$name} = \%kv; |
445 | $arg{cookie_jar}{$cdom}{$cpath}{$name} = \%kv; |
410 | } |
446 | } |
411 | } |
447 | } |
412 | |
448 | |
413 | if ($_[1]{Status} =~ /^30[12]$/ && $recurse) { |
|
|
414 | # microsoft and other assholes don't give a shit for following standards, |
449 | # microsoft and other shitheads don't give a shit for following standards, |
415 | # try to support a common form of broken Location header. |
450 | # try to support some common forms of broken Location headers. |
|
|
451 | if ($_[1]{location} !~ /^(?: $ | [^:\/?\#]+ : )/x) { |
|
|
452 | $_[1]{location} =~ s/^\.\/+//; |
|
|
453 | |
416 | $_[1]{location} =~ s%^/%$scheme://$uhost:$uport/%; |
454 | my $url = "$scheme://$uhost:$uport"; |
417 | |
455 | |
|
|
456 | unless ($_[1]{location} =~ s/^\///) { |
|
|
457 | $url .= $upath; |
|
|
458 | $url =~ s/\/[^\/]*$//; |
|
|
459 | } |
|
|
460 | |
|
|
461 | $_[1]{location} = "$url/$_[1]{location}"; |
|
|
462 | } |
|
|
463 | |
|
|
464 | if ($_[1]{Status} =~ /^30[12]$/ && $recurse && $method ne "POST") { |
|
|
465 | # apparently, mozilla et al. just change POST to GET here |
|
|
466 | # more research is needed before we do the same |
418 | http_request ($method, $_[1]{location}, %arg, recurse => $recurse - 1, $cb); |
467 | http_request ($method, $_[1]{location}, %arg, recurse => $recurse - 1, $cb); |
|
|
468 | } elsif ($_[1]{Status} == 303 && $recurse) { |
|
|
469 | # even http/1.1 is unclear on how to mutate the method |
|
|
470 | $method = "GET" unless $method eq "HEAD"; |
|
|
471 | http_request ($method => $_[1]{location}, %arg, recurse => $recurse - 1, $cb); |
|
|
472 | } elsif ($_[1]{Status} == 307 && $recurse && $method =~ /^(?:GET|HEAD)$/) { |
|
|
473 | http_request ($method => $_[1]{location}, %arg, recurse => $recurse - 1, $cb); |
419 | } else { |
474 | } else { |
420 | $cb->($_[0], $_[1]); |
475 | $cb->($_[0], $_[1]); |
421 | } |
476 | } |
422 | }; |
477 | }; |
423 | |
478 | |
… | |
… | |
462 | unshift @_, "HEAD"; |
517 | unshift @_, "HEAD"; |
463 | &http_request |
518 | &http_request |
464 | } |
519 | } |
465 | |
520 | |
466 | sub http_post($$@) { |
521 | sub http_post($$@) { |
|
|
522 | my $url = shift; |
467 | unshift @_, "POST", "body"; |
523 | unshift @_, "POST", $url, "body"; |
468 | &http_request |
524 | &http_request |
469 | } |
525 | } |
470 | |
526 | |
471 | =back |
527 | =back |
472 | |
528 | |
… | |
… | |
521 | |
577 | |
522 | L<AnyEvent>. |
578 | L<AnyEvent>. |
523 | |
579 | |
524 | =head1 AUTHOR |
580 | =head1 AUTHOR |
525 | |
581 | |
526 | Marc Lehmann <schmorp@schmorp.de> |
582 | Marc Lehmann <schmorp@schmorp.de> |
527 | http://home.schmorp.de/ |
583 | http://home.schmorp.de/ |
528 | |
584 | |
529 | =cut |
585 | =cut |
530 | |
586 | |
531 | 1 |
587 | 1 |
532 | |
588 | |