1 |
root |
1.1 |
=head1 NAME |
2 |
|
|
|
3 |
|
|
AnyEvent::HTTP - simple but non-blocking HTTP/HTTPS client |
4 |
|
|
|
5 |
|
|
=head1 SYNOPSIS |
6 |
|
|
|
7 |
|
|
use AnyEvent::HTTP; |
8 |
|
|
|
9 |
root |
1.17 |
http_get "http://www.nethype.de/", sub { print $_[1] }; |
10 |
|
|
|
11 |
|
|
# ... do something else here |
12 |
|
|
|
13 |
root |
1.1 |
=head1 DESCRIPTION |
14 |
|
|
|
15 |
|
|
This module is an L<AnyEvent> user, you need to make sure that you use and |
16 |
|
|
run a supported event loop. |
17 |
|
|
|
18 |
root |
1.11 |
This module implements a simple, stateless and non-blocking HTTP |
19 |
|
|
client. It supports GET, POST and other request methods, cookies and more, |
20 |
|
|
all on a very low level. It can follow redirects supports proxies and |
21 |
|
|
automatically limits the number of connections to the values specified in |
22 |
|
|
the RFC. |
23 |
|
|
|
24 |
|
|
It should generally be a "good client" that is enough for most HTTP |
25 |
|
|
tasks. Simple tasks should be simple, but complex tasks should still be |
26 |
|
|
possible as the user retains control over request and response headers. |
27 |
|
|
|
28 |
|
|
The caller is responsible for authentication management, cookies (if |
29 |
|
|
the simplistic implementation in this module doesn't suffice), referer |
30 |
|
|
and other high-level protocol details for which this module offers only |
31 |
|
|
limited support. |
32 |
|
|
|
33 |
root |
1.1 |
=head2 METHODS |
34 |
|
|
|
35 |
|
|
=over 4 |
36 |
|
|
|
37 |
|
|
=cut |
38 |
|
|
|
39 |
|
|
package AnyEvent::HTTP; |
40 |
|
|
|
41 |
|
|
use strict; |
42 |
|
|
no warnings; |
43 |
|
|
|
44 |
|
|
use Carp; |
45 |
|
|
|
46 |
|
|
use AnyEvent (); |
47 |
|
|
use AnyEvent::Util (); |
48 |
|
|
use AnyEvent::Socket (); |
49 |
|
|
use AnyEvent::Handle (); |
50 |
|
|
|
51 |
|
|
use base Exporter::; |
52 |
|
|
|
53 |
root |
1.37 |
our $VERSION = '1.11'; |
54 |
root |
1.1 |
|
55 |
root |
1.17 |
our @EXPORT = qw(http_get http_post http_head http_request); |
56 |
root |
1.1 |
|
57 |
|
|
our $USERAGENT = "Mozilla/5.0 (compatible; AnyEvent::HTTP/$VERSION; +http://software.schmorp.de/pkg/AnyEvent)"; |
58 |
root |
1.3 |
our $MAX_RECURSE = 10; |
59 |
root |
1.2 |
our $MAX_PERSISTENT = 8; |
60 |
|
|
our $PERSISTENT_TIMEOUT = 2; |
61 |
|
|
our $TIMEOUT = 300; |
62 |
root |
1.1 |
|
63 |
|
|
# changing these is evil |
64 |
|
|
our $MAX_PERSISTENT_PER_HOST = 2; |
65 |
root |
1.11 |
our $MAX_PER_HOST = 4; |
66 |
root |
1.1 |
|
67 |
root |
1.2 |
our $PROXY; |
68 |
root |
1.14 |
our $ACTIVE = 0; |
69 |
root |
1.2 |
|
70 |
root |
1.1 |
my %KA_COUNT; # number of open keep-alive connections per host |
71 |
root |
1.11 |
my %CO_SLOT; # number of open connections, and wait queue, per host |
72 |
root |
1.1 |
|
73 |
|
|
=item http_get $url, key => value..., $cb->($data, $headers) |
74 |
|
|
|
75 |
|
|
Executes an HTTP-GET request. See the http_request function for details on |
76 |
root |
1.29 |
additional parameters and the return value. |
77 |
root |
1.1 |
|
78 |
root |
1.5 |
=item http_head $url, key => value..., $cb->($data, $headers) |
79 |
|
|
|
80 |
root |
1.29 |
Executes an HTTP-HEAD request. See the http_request function for details |
81 |
|
|
on additional parameters and the return value. |
82 |
root |
1.5 |
|
83 |
|
|
=item http_post $url, $body, key => value..., $cb->($data, $headers) |
84 |
root |
1.3 |
|
85 |
root |
1.26 |
Executes an HTTP-POST request with a request body of C<$body>. See the |
86 |
root |
1.29 |
http_request function for details on additional parameters and the return |
87 |
|
|
value. |
88 |
root |
1.3 |
|
89 |
root |
1.1 |
=item http_request $method => $url, key => value..., $cb->($data, $headers) |
90 |
|
|
|
91 |
|
|
Executes a HTTP request of type C<$method> (e.g. C<GET>, C<POST>). The URL |
92 |
|
|
must be an absolute http or https URL. |
93 |
|
|
|
94 |
root |
1.29 |
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. |
98 |
|
|
|
99 |
root |
1.2 |
The callback will be called with the response data as first argument |
100 |
|
|
(or C<undef> if it wasn't available due to errors), and a hash-ref with |
101 |
|
|
response headers as second argument. |
102 |
|
|
|
103 |
root |
1.7 |
All the headers in that hash are lowercased. In addition to the response |
104 |
root |
1.20 |
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 |
106 |
|
|
pseudo-header C<URL> contains the original URL (which can differ from the |
107 |
|
|
requested URL when following redirects). |
108 |
|
|
|
109 |
root |
1.32 |
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. |
111 |
root |
1.2 |
|
112 |
|
|
If an internal error occurs, such as not being able to resolve a hostname, |
113 |
|
|
then C<$data> will be C<undef>, C<< $headers->{Status} >> will be C<599> |
114 |
|
|
and the C<Reason> pseudo-header will contain an error message. |
115 |
|
|
|
116 |
root |
1.6 |
A typical callback might look like this: |
117 |
|
|
|
118 |
|
|
sub { |
119 |
|
|
my ($body, $hdr) = @_; |
120 |
|
|
|
121 |
|
|
if ($hdr->{Status} =~ /^2/) { |
122 |
|
|
... everything should be ok |
123 |
|
|
} else { |
124 |
|
|
print "error, $hdr->{Status} $hdr->{Reason}\n"; |
125 |
|
|
} |
126 |
|
|
} |
127 |
|
|
|
128 |
root |
1.1 |
Additional parameters are key-value pairs, and are fully optional. They |
129 |
|
|
include: |
130 |
|
|
|
131 |
|
|
=over 4 |
132 |
|
|
|
133 |
root |
1.3 |
=item recurse => $count (default: $MAX_RECURSE) |
134 |
root |
1.1 |
|
135 |
|
|
Whether to recurse requests or not, e.g. on redirects, authentication |
136 |
root |
1.3 |
retries and so on, and how often to do so. |
137 |
root |
1.1 |
|
138 |
|
|
=item headers => hashref |
139 |
|
|
|
140 |
root |
1.12 |
The request headers to use. Currently, C<http_request> may provide its |
141 |
|
|
own C<Host:>, C<Content-Length:>, C<Connection:> and C<Cookie:> headers |
142 |
|
|
and will provide defaults for C<User-Agent:> and C<Referer:>. |
143 |
root |
1.1 |
|
144 |
|
|
=item timeout => $seconds |
145 |
|
|
|
146 |
|
|
The time-out to use for various stages - each connect attempt will reset |
147 |
root |
1.2 |
the timeout, as will read or write activity. Default timeout is 5 minutes. |
148 |
|
|
|
149 |
|
|
=item proxy => [$host, $port[, $scheme]] or undef |
150 |
|
|
|
151 |
|
|
Use the given http proxy for all requests. If not specified, then the |
152 |
|
|
default proxy (as specified by C<$ENV{http_proxy}>) is used. |
153 |
|
|
|
154 |
|
|
C<$scheme> must be either missing or C<http> for HTTP, or C<https> for |
155 |
|
|
HTTPS. |
156 |
root |
1.1 |
|
157 |
root |
1.3 |
=item body => $string |
158 |
|
|
|
159 |
|
|
The request body, usually empty. Will be-sent as-is (future versions of |
160 |
|
|
this module might offer more options). |
161 |
|
|
|
162 |
root |
1.10 |
=item cookie_jar => $hash_ref |
163 |
|
|
|
164 |
|
|
Passing this parameter enables (simplified) cookie-processing, loosely |
165 |
|
|
based on the original netscape specification. |
166 |
|
|
|
167 |
|
|
The C<$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 |
|
|
persistent storage with something like JSON or Storable, but this is not |
170 |
|
|
recommended, as expire times are currently being ignored. |
171 |
|
|
|
172 |
|
|
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 |
|
|
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 |
|
|
them unless required to. |
177 |
|
|
|
178 |
root |
1.1 |
=back |
179 |
|
|
|
180 |
root |
1.9 |
Example: make a simple HTTP GET request for http://www.nethype.de/ |
181 |
|
|
|
182 |
|
|
http_request GET => "http://www.nethype.de/", sub { |
183 |
|
|
my ($body, $hdr) = @_; |
184 |
|
|
print "$body\n"; |
185 |
|
|
}; |
186 |
|
|
|
187 |
|
|
Example: make a HTTP HEAD request on https://www.google.com/, use a |
188 |
|
|
timeout of 30 seconds. |
189 |
|
|
|
190 |
|
|
http_request |
191 |
|
|
GET => "https://www.google.com", |
192 |
|
|
timeout => 30, |
193 |
|
|
sub { |
194 |
|
|
my ($body, $hdr) = @_; |
195 |
|
|
use Data::Dumper; |
196 |
|
|
print Dumper $hdr; |
197 |
|
|
} |
198 |
|
|
; |
199 |
root |
1.1 |
|
200 |
root |
1.29 |
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 $request; |
209 |
|
|
|
210 |
root |
1.1 |
=cut |
211 |
|
|
|
212 |
root |
1.12 |
sub _slot_schedule; |
213 |
root |
1.11 |
sub _slot_schedule($) { |
214 |
|
|
my $host = shift; |
215 |
|
|
|
216 |
|
|
while ($CO_SLOT{$host}[0] < $MAX_PER_HOST) { |
217 |
|
|
if (my $cb = shift @{ $CO_SLOT{$host}[1] }) { |
218 |
root |
1.12 |
# somebody wants that slot |
219 |
root |
1.11 |
++$CO_SLOT{$host}[0]; |
220 |
root |
1.14 |
++$ACTIVE; |
221 |
root |
1.11 |
|
222 |
|
|
$cb->(AnyEvent::Util::guard { |
223 |
root |
1.14 |
--$ACTIVE; |
224 |
root |
1.11 |
--$CO_SLOT{$host}[0]; |
225 |
|
|
_slot_schedule $host; |
226 |
|
|
}); |
227 |
|
|
} else { |
228 |
|
|
# nobody wants the slot, maybe we can forget about it |
229 |
|
|
delete $CO_SLOT{$host} unless $CO_SLOT{$host}[0]; |
230 |
|
|
last; |
231 |
|
|
} |
232 |
|
|
} |
233 |
|
|
} |
234 |
|
|
|
235 |
|
|
# wait for a free slot on host, call callback |
236 |
|
|
sub _get_slot($$) { |
237 |
|
|
push @{ $CO_SLOT{$_[0]}[1] }, $_[1]; |
238 |
|
|
|
239 |
|
|
_slot_schedule $_[0]; |
240 |
|
|
} |
241 |
|
|
|
242 |
root |
1.34 |
our $qr_nl = qr<\015?\012>; |
243 |
|
|
our $qr_nlnl = qr<\015?\012\015?\012>; |
244 |
|
|
|
245 |
elmex |
1.15 |
sub http_request($$@) { |
246 |
root |
1.1 |
my $cb = pop; |
247 |
|
|
my ($method, $url, %arg) = @_; |
248 |
|
|
|
249 |
|
|
my %hdr; |
250 |
|
|
|
251 |
root |
1.3 |
$method = uc $method; |
252 |
|
|
|
253 |
root |
1.8 |
if (my $hdr = $arg{headers}) { |
254 |
root |
1.1 |
while (my ($k, $v) = each %$hdr) { |
255 |
|
|
$hdr{lc $k} = $v; |
256 |
|
|
} |
257 |
|
|
} |
258 |
|
|
|
259 |
root |
1.23 |
my $recurse = exists $arg{recurse} ? delete $arg{recurse} : $MAX_RECURSE; |
260 |
root |
1.8 |
|
261 |
elmex |
1.19 |
return $cb->(undef, { Status => 599, Reason => "recursion limit reached", URL => $url }) |
262 |
root |
1.8 |
if $recurse < 0; |
263 |
|
|
|
264 |
root |
1.2 |
my $proxy = $arg{proxy} || $PROXY; |
265 |
root |
1.1 |
my $timeout = $arg{timeout} || $TIMEOUT; |
266 |
|
|
|
267 |
|
|
$hdr{"user-agent"} ||= $USERAGENT; |
268 |
|
|
|
269 |
root |
1.31 |
my ($uscheme, $uauthority, $upath, $query, $fragment) = |
270 |
root |
1.10 |
$url =~ m|(?:([^:/?#]+):)?(?://([^/?#]*))?([^?#]*)(?:\?([^#]*))?(?:#(.*))?|; |
271 |
root |
1.2 |
|
272 |
root |
1.31 |
$uscheme = lc $uscheme; |
273 |
root |
1.1 |
|
274 |
root |
1.31 |
my $uport = $uscheme eq "http" ? 80 |
275 |
|
|
: $uscheme eq "https" ? 443 |
276 |
elmex |
1.19 |
: return $cb->(undef, { Status => 599, Reason => "only http and https URL schemes supported", URL => $url }); |
277 |
root |
1.10 |
|
278 |
root |
1.31 |
$hdr{referer} ||= "$uscheme://$uauthority$upath"; # leave out fragment and query string, just a heuristic |
279 |
root |
1.13 |
|
280 |
root |
1.31 |
$uauthority =~ /^(?: .*\@ )? ([^\@:]+) (?: : (\d+) )?$/x |
281 |
elmex |
1.19 |
or return $cb->(undef, { Status => 599, Reason => "unparsable URL", URL => $url }); |
282 |
root |
1.10 |
|
283 |
|
|
my $uhost = $1; |
284 |
|
|
$uport = $2 if defined $2; |
285 |
|
|
|
286 |
|
|
$uhost =~ s/^\[(.*)\]$/$1/; |
287 |
|
|
$upath .= "?$query" if length $query; |
288 |
|
|
|
289 |
|
|
$upath =~ s%^/?%/%; |
290 |
|
|
|
291 |
|
|
# cookie processing |
292 |
|
|
if (my $jar = $arg{cookie_jar}) { |
293 |
root |
1.31 |
%$jar = () if $jar->{version} != 1; |
294 |
root |
1.10 |
|
295 |
|
|
my @cookie; |
296 |
|
|
|
297 |
|
|
while (my ($chost, $v) = each %$jar) { |
298 |
root |
1.30 |
if ($chost =~ /^\./) { |
299 |
|
|
next unless $chost eq substr $uhost, -length $chost; |
300 |
|
|
} elsif ($chost =~ /\./) { |
301 |
|
|
next unless $chost eq $uhost; |
302 |
|
|
} else { |
303 |
|
|
next; |
304 |
|
|
} |
305 |
root |
1.10 |
|
306 |
|
|
while (my ($cpath, $v) = each %$v) { |
307 |
|
|
next unless $cpath eq substr $upath, 0, length $cpath; |
308 |
|
|
|
309 |
|
|
while (my ($k, $v) = each %$v) { |
310 |
root |
1.31 |
next if $uscheme ne "https" && exists $v->{secure}; |
311 |
|
|
my $value = $v->{value}; |
312 |
|
|
$value =~ s/([\\"])/\\$1/g; |
313 |
|
|
push @cookie, "$k=\"$value\""; |
314 |
root |
1.10 |
} |
315 |
|
|
} |
316 |
|
|
} |
317 |
|
|
|
318 |
|
|
$hdr{cookie} = join "; ", @cookie |
319 |
|
|
if @cookie; |
320 |
|
|
} |
321 |
root |
1.1 |
|
322 |
root |
1.31 |
my ($rhost, $rport, $rscheme, $rpath); # request host, port, path |
323 |
root |
1.2 |
|
324 |
root |
1.10 |
if ($proxy) { |
325 |
root |
1.38 |
($rpath, $rhost, $rport, $rscheme) = ($url, @$proxy); |
326 |
root |
1.31 |
|
327 |
|
|
# don't support https requests over https-proxy transport, |
328 |
root |
1.38 |
# can't be done with tls as spec'ed, unless you double-encrypt. |
329 |
root |
1.31 |
$rscheme = "http" if $uscheme eq "https" && $rscheme eq "https"; |
330 |
root |
1.10 |
} else { |
331 |
root |
1.31 |
($rhost, $rport, $rscheme, $rpath) = ($uhost, $uport, $uscheme, $upath); |
332 |
root |
1.2 |
} |
333 |
|
|
|
334 |
root |
1.31 |
$hdr{host} = $uhost; |
335 |
root |
1.10 |
$hdr{"content-length"} = length $arg{body}; |
336 |
root |
1.1 |
|
337 |
root |
1.11 |
my %state = (connect_guard => 1); |
338 |
|
|
|
339 |
|
|
_get_slot $uhost, sub { |
340 |
|
|
$state{slot_guard} = shift; |
341 |
root |
1.1 |
|
342 |
root |
1.11 |
return unless $state{connect_guard}; |
343 |
root |
1.1 |
|
344 |
root |
1.11 |
$state{connect_guard} = AnyEvent::Socket::tcp_connect $rhost, $rport, sub { |
345 |
|
|
$state{fh} = shift |
346 |
elmex |
1.19 |
or return $cb->(undef, { Status => 599, Reason => "$!", URL => $url }); |
347 |
root |
1.34 |
pop; # free memory, save a tree |
348 |
root |
1.11 |
|
349 |
root |
1.34 |
return unless delete $state{connect_guard}; |
350 |
root |
1.11 |
|
351 |
|
|
# get handle |
352 |
|
|
$state{handle} = new AnyEvent::Handle |
353 |
root |
1.34 |
fh => $state{fh}, |
354 |
|
|
timeout => $timeout; |
355 |
root |
1.11 |
|
356 |
|
|
# limit the number of persistent connections |
357 |
root |
1.34 |
# keepalive not yet supported |
358 |
root |
1.11 |
if ($KA_COUNT{$_[1]} < $MAX_PERSISTENT_PER_HOST) { |
359 |
|
|
++$KA_COUNT{$_[1]}; |
360 |
root |
1.34 |
$state{handle}{ka_count_guard} = AnyEvent::Util::guard { |
361 |
|
|
--$KA_COUNT{$_[1]} |
362 |
|
|
}; |
363 |
root |
1.11 |
$hdr{connection} = "keep-alive"; |
364 |
|
|
} else { |
365 |
|
|
delete $hdr{connection}; |
366 |
|
|
} |
367 |
root |
1.1 |
|
368 |
root |
1.11 |
# (re-)configure handle |
369 |
|
|
$state{handle}->on_error (sub { |
370 |
root |
1.14 |
my $errno = "$!"; |
371 |
root |
1.11 |
%state = (); |
372 |
elmex |
1.19 |
$cb->(undef, { Status => 599, Reason => $errno, URL => $url }); |
373 |
root |
1.11 |
}); |
374 |
|
|
$state{handle}->on_eof (sub { |
375 |
|
|
%state = (); |
376 |
elmex |
1.19 |
$cb->(undef, { Status => 599, Reason => "unexpected end-of-file", URL => $url }); |
377 |
root |
1.11 |
}); |
378 |
root |
1.1 |
|
379 |
root |
1.31 |
$state{handle}->starttls ("connect") if $rscheme eq "https"; |
380 |
|
|
|
381 |
|
|
# handle actual, non-tunneled, request |
382 |
|
|
my $handle_actual_request = sub { |
383 |
root |
1.34 |
$state{handle}->starttls ("connect") if $uscheme eq "https" && !exists $state{handle}{tls}; |
384 |
root |
1.31 |
|
385 |
|
|
# send request |
386 |
|
|
$state{handle}->push_write ( |
387 |
|
|
"$method $rpath HTTP/1.0\015\012" |
388 |
|
|
. (join "", map "\u$_: $hdr{$_}\015\012", keys %hdr) |
389 |
|
|
. "\015\012" |
390 |
|
|
. (delete $arg{body}) |
391 |
root |
1.11 |
); |
392 |
|
|
|
393 |
root |
1.31 |
%hdr = (); # reduce memory usage, save a kitten |
394 |
root |
1.10 |
|
395 |
root |
1.31 |
# status line |
396 |
root |
1.34 |
$state{handle}->push_read (line => $qr_nl, sub { |
397 |
root |
1.31 |
$_[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 })); |
399 |
|
|
|
400 |
|
|
my %hdr = ( # response headers |
401 |
|
|
HTTPVersion => ",$1", |
402 |
|
|
Status => ",$2", |
403 |
|
|
Reason => ",$3", |
404 |
|
|
URL => ",$url" |
405 |
|
|
); |
406 |
|
|
|
407 |
|
|
# headers, could be optimized a bit |
408 |
root |
1.34 |
$state{handle}->unshift_read (line => $qr_nlnl, sub { |
409 |
root |
1.31 |
for ("$_[1]\012") { |
410 |
|
|
y/\015//d; # weed out any \015, as they show up in the weirdest of places. |
411 |
|
|
|
412 |
|
|
# we support spaces in field names, as lotus domino |
413 |
|
|
# creates them (actually spaces around seperators |
414 |
|
|
# are strictly allowed in http, they are a security issue). |
415 |
|
|
$hdr{lc $1} .= ",$2" |
416 |
|
|
while /\G |
417 |
|
|
([^:\000-\037]+): |
418 |
|
|
[\011\040]* |
419 |
|
|
((?: [^\012]+ | \012[\011\040] )*) |
420 |
|
|
\012 |
421 |
|
|
/gxc; |
422 |
|
|
|
423 |
|
|
/\G$/ |
424 |
|
|
or return (%state = (), $cb->(undef, { Status => 599, Reason => "garbled response headers", URL => $url })); |
425 |
|
|
} |
426 |
|
|
|
427 |
|
|
substr $_, 0, 1, "" |
428 |
|
|
for values %hdr; |
429 |
|
|
|
430 |
|
|
my $finish = sub { |
431 |
root |
1.35 |
$state{handle}->destroy; |
432 |
root |
1.31 |
%state = (); |
433 |
|
|
|
434 |
|
|
# set-cookie processing |
435 |
|
|
if ($arg{cookie_jar}) { |
436 |
|
|
for ($hdr{"set-cookie"}) { |
437 |
|
|
# parse NAME=VALUE |
438 |
|
|
my @kv; |
439 |
|
|
|
440 |
|
|
while (/\G\s* ([^=;,[:space:]]+) \s*=\s* (?: "((?:[^\\"]+|\\.)*)" | ([^=;,[:space:]]*) )/gcxs) { |
441 |
|
|
my $name = $1; |
442 |
|
|
my $value = $3; |
443 |
|
|
|
444 |
|
|
unless ($value) { |
445 |
|
|
$value = $2; |
446 |
|
|
$value =~ s/\\(.)/$1/gs; |
447 |
|
|
} |
448 |
|
|
|
449 |
|
|
push @kv, $name => $value; |
450 |
|
|
|
451 |
|
|
last unless /\G\s*;/gc; |
452 |
|
|
} |
453 |
|
|
|
454 |
|
|
last unless @kv; |
455 |
root |
1.10 |
|
456 |
root |
1.31 |
my $name = shift @kv; |
457 |
|
|
my %kv = (value => shift @kv, @kv); |
458 |
root |
1.11 |
|
459 |
root |
1.31 |
my $cdom; |
460 |
|
|
my $cpath = (delete $kv{path}) || "/"; |
461 |
root |
1.10 |
|
462 |
root |
1.31 |
if (exists $kv{domain}) { |
463 |
|
|
$cdom = delete $kv{domain}; |
464 |
|
|
|
465 |
|
|
$cdom =~ s/^\.?/./; # make sure it starts with a "." |
466 |
root |
1.11 |
|
467 |
root |
1.31 |
next if $cdom =~ /\.$/; |
468 |
|
|
|
469 |
|
|
# this is not rfc-like and not netscape-like. go figure. |
470 |
|
|
my $ndots = $cdom =~ y/.//; |
471 |
|
|
next if $ndots < ($cdom =~ /\.[^.][^.]\.[^.][^.]$/ ? 3 : 2); |
472 |
|
|
} else { |
473 |
|
|
$cdom = $uhost; |
474 |
|
|
} |
475 |
root |
1.30 |
|
476 |
root |
1.31 |
# store it |
477 |
|
|
$arg{cookie_jar}{version} = 1; |
478 |
|
|
$arg{cookie_jar}{$cdom}{$cpath}{$name} = \%kv; |
479 |
|
|
|
480 |
|
|
redo if /\G\s*,/gc; |
481 |
root |
1.30 |
} |
482 |
root |
1.11 |
} |
483 |
root |
1.8 |
|
484 |
root |
1.31 |
# microsoft and other shitheads don't give a shit for following standards, |
485 |
|
|
# try to support some common forms of broken Location headers. |
486 |
|
|
if ($_[1]{location} !~ /^(?: $ | [^:\/?\#]+ : )/x) { |
487 |
|
|
$_[1]{location} =~ s/^\.\/+//; |
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/$_[1]{location}"; |
497 |
root |
1.25 |
} |
498 |
|
|
|
499 |
root |
1.31 |
if ($_[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, $cb); |
507 |
|
|
} elsif ($_[1]{Status} == 307 && $recurse && $method =~ /^(?:GET|HEAD)$/) { |
508 |
|
|
http_request ($method => $_[1]{location}, %arg, recurse => $recurse - 1, $cb); |
509 |
|
|
} else { |
510 |
|
|
$cb->($_[0], $_[1]); |
511 |
|
|
} |
512 |
|
|
}; |
513 |
root |
1.24 |
|
514 |
root |
1.31 |
if ($hdr{Status} =~ /^(?:1..|204|304)$/ or $method eq "HEAD") { |
515 |
|
|
$finish->(undef, \%hdr); |
516 |
root |
1.11 |
} else { |
517 |
root |
1.31 |
if (exists $hdr{"content-length"}) { |
518 |
|
|
$_[0]->unshift_read (chunk => $hdr{"content-length"}, sub { |
519 |
|
|
# could cache persistent connection now |
520 |
|
|
if ($hdr{connection} =~ /\bkeep-alive\b/i) { |
521 |
|
|
# but we don't, due to misdesigns, this is annoyingly complex |
522 |
|
|
}; |
523 |
|
|
|
524 |
|
|
$finish->($_[1], \%hdr); |
525 |
|
|
}); |
526 |
|
|
} 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 |
root |
1.37 |
# 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 |
root |
1.31 |
}); |
534 |
|
|
$_[0]->on_eof (undef); |
535 |
|
|
$_[0]->on_read (sub { }); |
536 |
|
|
} |
537 |
root |
1.11 |
} |
538 |
root |
1.31 |
}); |
539 |
|
|
}); |
540 |
|
|
}; |
541 |
root |
1.3 |
|
542 |
root |
1.31 |
# now handle proxy-CONNECT method |
543 |
|
|
if ($proxy && $uscheme eq "https") { |
544 |
|
|
# oh dear, we have to wrap it into a connect request |
545 |
|
|
|
546 |
|
|
# 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"); |
548 |
root |
1.34 |
$state{handle}->push_read (line => $qr_nlnl, sub { |
549 |
root |
1.31 |
$_[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 })); |
551 |
|
|
|
552 |
|
|
if ($2 == 200) { |
553 |
|
|
$rpath = $upath; |
554 |
|
|
&$handle_actual_request; |
555 |
root |
1.3 |
} else { |
556 |
root |
1.31 |
%state = (); |
557 |
|
|
$cb->(undef, { Status => $2, Reason => $3, URL => $url }); |
558 |
root |
1.3 |
} |
559 |
root |
1.11 |
}); |
560 |
root |
1.31 |
} else { |
561 |
|
|
&$handle_actual_request; |
562 |
|
|
} |
563 |
|
|
|
564 |
root |
1.11 |
}, sub { |
565 |
|
|
$timeout |
566 |
|
|
}; |
567 |
root |
1.1 |
}; |
568 |
|
|
|
569 |
|
|
defined wantarray && AnyEvent::Util::guard { %state = () } |
570 |
|
|
} |
571 |
|
|
|
572 |
elmex |
1.15 |
sub http_get($@) { |
573 |
root |
1.1 |
unshift @_, "GET"; |
574 |
|
|
&http_request |
575 |
|
|
} |
576 |
|
|
|
577 |
elmex |
1.15 |
sub http_head($@) { |
578 |
root |
1.4 |
unshift @_, "HEAD"; |
579 |
|
|
&http_request |
580 |
|
|
} |
581 |
|
|
|
582 |
elmex |
1.15 |
sub http_post($$@) { |
583 |
root |
1.22 |
my $url = shift; |
584 |
|
|
unshift @_, "POST", $url, "body"; |
585 |
root |
1.3 |
&http_request |
586 |
|
|
} |
587 |
|
|
|
588 |
root |
1.9 |
=back |
589 |
|
|
|
590 |
root |
1.2 |
=head2 GLOBAL FUNCTIONS AND VARIABLES |
591 |
root |
1.1 |
|
592 |
|
|
=over 4 |
593 |
|
|
|
594 |
root |
1.2 |
=item AnyEvent::HTTP::set_proxy "proxy-url" |
595 |
|
|
|
596 |
|
|
Sets the default proxy server to use. The proxy-url must begin with a |
597 |
|
|
string of the form C<http://host:port> (optionally C<https:...>). |
598 |
|
|
|
599 |
root |
1.3 |
=item $AnyEvent::HTTP::MAX_RECURSE |
600 |
root |
1.1 |
|
601 |
root |
1.3 |
The default value for the C<recurse> request parameter (default: C<10>). |
602 |
root |
1.1 |
|
603 |
|
|
=item $AnyEvent::HTTP::USERAGENT |
604 |
|
|
|
605 |
|
|
The default value for the C<User-Agent> header (the default is |
606 |
|
|
C<Mozilla/5.0 (compatible; AnyEvent::HTTP/$VERSION; +http://software.schmorp.de/pkg/AnyEvent)>). |
607 |
|
|
|
608 |
|
|
=item $AnyEvent::HTTP::MAX_PERSISTENT |
609 |
|
|
|
610 |
|
|
The maximum number of persistent connections to keep open (default: 8). |
611 |
|
|
|
612 |
root |
1.3 |
Not implemented currently. |
613 |
|
|
|
614 |
root |
1.1 |
=item $AnyEvent::HTTP::PERSISTENT_TIMEOUT |
615 |
|
|
|
616 |
root |
1.2 |
The maximum time to cache a persistent connection, in seconds (default: 2). |
617 |
root |
1.1 |
|
618 |
root |
1.3 |
Not implemented currently. |
619 |
|
|
|
620 |
root |
1.14 |
=item $AnyEvent::HTTP::ACTIVE |
621 |
|
|
|
622 |
|
|
The number of active connections. This is not the number of currently |
623 |
|
|
running requests, but the number of currently open and non-idle TCP |
624 |
|
|
connections. This number of can be useful for load-leveling. |
625 |
|
|
|
626 |
root |
1.1 |
=back |
627 |
|
|
|
628 |
|
|
=cut |
629 |
|
|
|
630 |
root |
1.2 |
sub set_proxy($) { |
631 |
|
|
$PROXY = [$2, $3 || 3128, $1] if $_[0] =~ m%^(https?):// ([^:/]+) (?: : (\d*) )?%ix; |
632 |
|
|
} |
633 |
|
|
|
634 |
|
|
# initialise proxy from environment |
635 |
|
|
set_proxy $ENV{http_proxy}; |
636 |
|
|
|
637 |
root |
1.1 |
=head1 SEE ALSO |
638 |
|
|
|
639 |
|
|
L<AnyEvent>. |
640 |
|
|
|
641 |
|
|
=head1 AUTHOR |
642 |
|
|
|
643 |
root |
1.18 |
Marc Lehmann <schmorp@schmorp.de> |
644 |
|
|
http://home.schmorp.de/ |
645 |
root |
1.1 |
|
646 |
root |
1.36 |
With many thanks to Дмитрий Шалашов, who provided countless |
647 |
|
|
testcases and bugreports. |
648 |
|
|
|
649 |
root |
1.1 |
=cut |
650 |
|
|
|
651 |
|
|
1 |
652 |
|
|
|