ViewVC Help
View File | Revision Log | Show Annotations | Download File
/cvs/AnyEvent/lib/AnyEvent/Handle.pm
(Generate patch)

Comparing AnyEvent/lib/AnyEvent/Handle.pm (file contents):
Revision 1.115 by root, Tue Feb 10 13:58:49 2009 UTC vs.
Revision 1.247 by root, Thu Jan 7 10:03:46 2016 UTC

1package AnyEvent::Handle;
2
3no warnings;
4use strict qw(subs vars);
5
6use AnyEvent ();
7use AnyEvent::Util qw(WSAEWOULDBLOCK);
8use Scalar::Util ();
9use Carp ();
10use Fcntl ();
11use Errno qw(EAGAIN EINTR);
12
13=head1 NAME 1=head1 NAME
14 2
15AnyEvent::Handle - non-blocking I/O on file handles via AnyEvent 3AnyEvent::Handle - non-blocking I/O on streaming handles via AnyEvent
16
17=cut
18
19our $VERSION = 4.331;
20 4
21=head1 SYNOPSIS 5=head1 SYNOPSIS
22 6
23 use AnyEvent; 7 use AnyEvent;
24 use AnyEvent::Handle; 8 use AnyEvent::Handle;
25 9
26 my $cv = AnyEvent->condvar; 10 my $cv = AnyEvent->condvar;
27 11
28 my $handle = 12 my $hdl; $hdl = new AnyEvent::Handle
29 AnyEvent::Handle->new (
30 fh => \*STDIN, 13 fh => \*STDIN,
31 on_eof => sub { 14 on_error => sub {
15 my ($hdl, $fatal, $msg) = @_;
16 AE::log error => $msg;
17 $hdl->destroy;
32 $cv->send; 18 $cv->send;
33 },
34 ); 19 };
35 20
36 # send some request line 21 # send some request line
37 $handle->push_write ("getinfo\015\012"); 22 $hdl->push_write ("getinfo\015\012");
38 23
39 # read the response line 24 # read the response line
40 $handle->push_read (line => sub { 25 $hdl->push_read (line => sub {
41 my ($handle, $line) = @_; 26 my ($hdl, $line) = @_;
42 warn "read line <$line>\n"; 27 say "got line <$line>";
43 $cv->send; 28 $cv->send;
44 }); 29 });
45 30
46 $cv->recv; 31 $cv->recv;
47 32
48=head1 DESCRIPTION 33=head1 DESCRIPTION
49 34
50This module is a helper module to make it easier to do event-based I/O on 35This is a helper module to make it easier to do event-based I/O
51filehandles. For utility functions for doing non-blocking connects and accepts 36on stream-based filehandles (sockets, pipes, and other stream
52on sockets see L<AnyEvent::Util>. 37things). Specifically, it doesn't work as expected on files, packet-based
38sockets or similar things.
53 39
54The L<AnyEvent::Intro> tutorial contains some well-documented 40The L<AnyEvent::Intro> tutorial contains some well-documented
55AnyEvent::Handle examples. 41AnyEvent::Handle examples.
56 42
57In the following, when the documentation refers to of "bytes" then this 43In the following, where the documentation refers to "bytes", it means
58means characters. As sysread and syswrite are used for all I/O, their 44characters. As sysread and syswrite are used for all I/O, their
59treatment of characters applies to this module as well. 45treatment of characters applies to this module as well.
46
47At the very minimum, you should specify C<fh> or C<connect>, and the
48C<on_error> callback.
60 49
61All callbacks will be invoked with the handle object as their first 50All callbacks will be invoked with the handle object as their first
62argument. 51argument.
63 52
53=cut
54
55package AnyEvent::Handle;
56
57use Scalar::Util ();
58use List::Util ();
59use Carp ();
60use Errno qw(EAGAIN EWOULDBLOCK EINTR);
61
62use AnyEvent (); BEGIN { AnyEvent::common_sense }
63use AnyEvent::Util qw(WSAEWOULDBLOCK);
64
65our $VERSION = $AnyEvent::VERSION;
66
67sub _load_func($) {
68 my $func = $_[0];
69
70 unless (defined &$func) {
71 my $pkg = $func;
72 do {
73 $pkg =~ s/::[^:]+$//
74 or return;
75 eval "require $pkg";
76 } until defined &$func;
77 }
78
79 \&$func
80}
81
82sub MAX_READ_SIZE() { 131072 }
83
64=head1 METHODS 84=head1 METHODS
65 85
66=over 4 86=over 4
67 87
68=item B<new (%args)> 88=item $handle = B<new> AnyEvent::Handle fh => $filehandle, key => value...
69 89
70The constructor supports these arguments (all as key => value pairs). 90The constructor supports these arguments (all as C<< key => value >> pairs).
71 91
72=over 4 92=over 4
73 93
74=item fh => $filehandle [MANDATORY] 94=item fh => $filehandle [C<fh> or C<connect> MANDATORY]
75 95
76The filehandle this L<AnyEvent::Handle> object will operate on. 96The filehandle this L<AnyEvent::Handle> object will operate on.
77
78NOTE: The filehandle will be set to non-blocking mode (using 97NOTE: The filehandle will be set to non-blocking mode (using
79C<AnyEvent::Util::fh_nonblocking>) by the constructor and needs to stay in 98C<AnyEvent::fh_unblock>) by the constructor and needs to stay in
80that mode. 99that mode.
100
101=item connect => [$host, $service] [C<fh> or C<connect> MANDATORY]
102
103Try to connect to the specified host and service (port), using
104C<AnyEvent::Socket::tcp_connect>. The C<$host> additionally becomes the
105default C<peername>.
106
107You have to specify either this parameter, or C<fh>, above.
108
109It is possible to push requests on the read and write queues, and modify
110properties of the stream, even while AnyEvent::Handle is connecting.
111
112When this parameter is specified, then the C<on_prepare>,
113C<on_connect_error> and C<on_connect> callbacks will be called under the
114appropriate circumstances:
115
116=over 4
117
118=item on_prepare => $cb->($handle)
119
120This (rarely used) callback is called before a new connection is
121attempted, but after the file handle has been created (you can access that
122file handle via C<< $handle->{fh} >>). It could be used to prepare the
123file handle with parameters required for the actual connect (as opposed to
124settings that can be changed when the connection is already established).
125
126The return value of this callback should be the connect timeout value in
127seconds (or C<0>, or C<undef>, or the empty list, to indicate that the
128default timeout is to be used).
129
130=item on_connect => $cb->($handle, $host, $port, $retry->())
131
132This callback is called when a connection has been successfully established.
133
134The peer's numeric host and port (the socket peername) are passed as
135parameters, together with a retry callback. At the time it is called the
136read and write queues, EOF status, TLS status and similar properties of
137the handle will have been reset.
138
139If, for some reason, the handle is not acceptable, calling C<$retry> will
140continue with the next connection target (in case of multi-homed hosts or
141SRV records there can be multiple connection endpoints). The C<$retry>
142callback can be invoked after the connect callback returns, i.e. one can
143start a handshake and then decide to retry with the next host if the
144handshake fails.
145
146In most cases, you should ignore the C<$retry> parameter.
147
148=item on_connect_error => $cb->($handle, $message)
149
150This callback is called when the connection could not be
151established. C<$!> will contain the relevant error code, and C<$message> a
152message describing it (usually the same as C<"$!">).
153
154If this callback isn't specified, then C<on_error> will be called with a
155fatal error instead.
156
157=back
158
159=item on_error => $cb->($handle, $fatal, $message)
160
161This is the error callback, which is called when, well, some error
162occured, such as not being able to resolve the hostname, failure to
163connect, or a read error.
164
165Some errors are fatal (which is indicated by C<$fatal> being true). On
166fatal errors the handle object will be destroyed (by a call to C<< ->
167destroy >>) after invoking the error callback (which means you are free to
168examine the handle object). Examples of fatal errors are an EOF condition
169with active (but unsatisfiable) read watchers (C<EPIPE>) or I/O errors. In
170cases where the other side can close the connection at will, it is
171often easiest to not report C<EPIPE> errors in this callback.
172
173AnyEvent::Handle tries to find an appropriate error code for you to check
174against, but in some cases (TLS errors), this does not work well.
175
176If you report the error to the user, it is recommended to always output
177the C<$message> argument in human-readable error messages (you don't need
178to report C<"$!"> if you report C<$message>).
179
180If you want to react programmatically to the error, then looking at C<$!>
181and comparing it against some of the documented C<Errno> values is usually
182better than looking at the C<$message>.
183
184Non-fatal errors can be retried by returning, but it is recommended
185to simply ignore this parameter and instead abondon the handle object
186when this callback is invoked. Examples of non-fatal errors are timeouts
187C<ETIMEDOUT>) or badly-formatted data (C<EBADMSG>).
188
189On entry to the callback, the value of C<$!> contains the operating
190system error code (or C<ENOSPC>, C<EPIPE>, C<ETIMEDOUT>, C<EBADMSG> or
191C<EPROTO>).
192
193While not mandatory, it is I<highly> recommended to set this callback, as
194you will not be notified of errors otherwise. The default just calls
195C<croak>.
196
197=item on_read => $cb->($handle)
198
199This sets the default read callback, which is called when data arrives
200and no read request is in the queue (unlike read queue callbacks, this
201callback will only be called when at least one octet of data is in the
202read buffer).
203
204To access (and remove data from) the read buffer, use the C<< ->rbuf >>
205method or access the C<< $handle->{rbuf} >> member directly. Note that you
206must not enlarge or modify the read buffer, you can only remove data at
207the beginning from it.
208
209You can also call C<< ->push_read (...) >> or any other function that
210modifies the read queue. Or do both. Or ...
211
212When an EOF condition is detected, AnyEvent::Handle will first try to
213feed all the remaining data to the queued callbacks and C<on_read> before
214calling the C<on_eof> callback. If no progress can be made, then a fatal
215error will be raised (with C<$!> set to C<EPIPE>).
216
217Note that, unlike requests in the read queue, an C<on_read> callback
218doesn't mean you I<require> some data: if there is an EOF and there
219are outstanding read requests then an error will be flagged. With an
220C<on_read> callback, the C<on_eof> callback will be invoked.
81 221
82=item on_eof => $cb->($handle) 222=item on_eof => $cb->($handle)
83 223
84Set the callback to be called when an end-of-file condition is detected, 224Set the callback to be called when an end-of-file condition is detected,
85i.e. in the case of a socket, when the other side has closed the 225i.e. in the case of a socket, when the other side has closed the
86connection cleanly. 226connection cleanly, and there are no outstanding read requests in the
227queue (if there are read requests, then an EOF counts as an unexpected
228connection close and will be flagged as an error).
87 229
88For sockets, this just means that the other side has stopped sending data, 230For sockets, this just means that the other side has stopped sending data,
89you can still try to write data, and, in fact, one can return from the EOF 231you can still try to write data, and, in fact, one can return from the EOF
90callback and continue writing data, as only the read part has been shut 232callback and continue writing data, as only the read part has been shut
91down. 233down.
92 234
93While not mandatory, it is I<highly> recommended to set an EOF callback,
94otherwise you might end up with a closed socket while you are still
95waiting for data.
96
97If an EOF condition has been detected but no C<on_eof> callback has been 235If an EOF condition has been detected but no C<on_eof> callback has been
98set, then a fatal error will be raised with C<$!> set to <0>. 236set, then a fatal error will be raised with C<$!> set to <0>.
99 237
100=item on_error => $cb->($handle, $fatal)
101
102This is the error callback, which is called when, well, some error
103occured, such as not being able to resolve the hostname, failure to
104connect or a read error.
105
106Some errors are fatal (which is indicated by C<$fatal> being true). On
107fatal errors the handle object will be shut down and will not be usable
108(but you are free to look at the current C<< ->rbuf >>). Examples of fatal
109errors are an EOF condition with active (but unsatisifable) read watchers
110(C<EPIPE>) or I/O errors.
111
112Non-fatal errors can be retried by simply returning, but it is recommended
113to simply ignore this parameter and instead abondon the handle object
114when this callback is invoked. Examples of non-fatal errors are timeouts
115C<ETIMEDOUT>) or badly-formatted data (C<EBADMSG>).
116
117On callback entrance, the value of C<$!> contains the operating system
118error (or C<ENOSPC>, C<EPIPE>, C<ETIMEDOUT> or C<EBADMSG>).
119
120While not mandatory, it is I<highly> recommended to set this callback, as
121you will not be notified of errors otherwise. The default simply calls
122C<croak>.
123
124=item on_read => $cb->($handle)
125
126This sets the default read callback, which is called when data arrives
127and no read request is in the queue (unlike read queue callbacks, this
128callback will only be called when at least one octet of data is in the
129read buffer).
130
131To access (and remove data from) the read buffer, use the C<< ->rbuf >>
132method or access the C<$handle->{rbuf}> member directly.
133
134When an EOF condition is detected then AnyEvent::Handle will first try to
135feed all the remaining data to the queued callbacks and C<on_read> before
136calling the C<on_eof> callback. If no progress can be made, then a fatal
137error will be raised (with C<$!> set to C<EPIPE>).
138
139=item on_drain => $cb->($handle) 238=item on_drain => $cb->($handle)
140 239
141This sets the callback that is called when the write buffer becomes empty 240This sets the callback that is called once when the write buffer becomes
142(or when the callback is set and the buffer is empty already). 241empty (and immediately when the handle object is created).
143 242
144To append to the write buffer, use the C<< ->push_write >> method. 243To append to the write buffer, use the C<< ->push_write >> method.
145 244
146This callback is useful when you don't want to put all of your write data 245This callback is useful when you don't want to put all of your write data
147into the queue at once, for example, when you want to write the contents 246into the queue at once, for example, when you want to write the contents
149memory and push it into the queue, but instead only read more data from 248memory and push it into the queue, but instead only read more data from
150the file when the write queue becomes empty. 249the file when the write queue becomes empty.
151 250
152=item timeout => $fractional_seconds 251=item timeout => $fractional_seconds
153 252
253=item rtimeout => $fractional_seconds
254
255=item wtimeout => $fractional_seconds
256
154If non-zero, then this enables an "inactivity" timeout: whenever this many 257If non-zero, then these enables an "inactivity" timeout: whenever this
155seconds pass without a successful read or write on the underlying file 258many seconds pass without a successful read or write on the underlying
156handle, the C<on_timeout> callback will be invoked (and if that one is 259file handle (or a call to C<timeout_reset>), the C<on_timeout> callback
157missing, a non-fatal C<ETIMEDOUT> error will be raised). 260will be invoked (and if that one is missing, a non-fatal C<ETIMEDOUT>
261error will be raised).
158 262
263There are three variants of the timeouts that work independently of each
264other, for both read and write (triggered when nothing was read I<OR>
265written), just read (triggered when nothing was read), and just write:
266C<timeout>, C<rtimeout> and C<wtimeout>, with corresponding callbacks
267C<on_timeout>, C<on_rtimeout> and C<on_wtimeout>, and reset functions
268C<timeout_reset>, C<rtimeout_reset>, and C<wtimeout_reset>.
269
159Note that timeout processing is also active when you currently do not have 270Note that timeout processing is active even when you do not have any
160any outstanding read or write requests: If you plan to keep the connection 271outstanding read or write requests: If you plan to keep the connection
161idle then you should disable the timout temporarily or ignore the timeout 272idle then you should disable the timeout temporarily or ignore the
162in the C<on_timeout> callback, in which case AnyEvent::Handle will simply 273timeout in the corresponding C<on_timeout> callback, in which case
163restart the timeout. 274AnyEvent::Handle will simply restart the timeout.
164 275
165Zero (the default) disables this timeout. 276Zero (the default) disables the corresponding timeout.
166 277
167=item on_timeout => $cb->($handle) 278=item on_timeout => $cb->($handle)
279
280=item on_rtimeout => $cb->($handle)
281
282=item on_wtimeout => $cb->($handle)
168 283
169Called whenever the inactivity timeout passes. If you return from this 284Called whenever the inactivity timeout passes. If you return from this
170callback, then the timeout will be reset as if some activity had happened, 285callback, then the timeout will be reset as if some activity had happened,
171so this condition is not fatal in any way. 286so this condition is not fatal in any way.
172 287
180be configured to accept only so-and-so much data that it cannot act on 295be configured to accept only so-and-so much data that it cannot act on
181(for example, when expecting a line, an attacker could send an unlimited 296(for example, when expecting a line, an attacker could send an unlimited
182amount of data without a callback ever being called as long as the line 297amount of data without a callback ever being called as long as the line
183isn't finished). 298isn't finished).
184 299
300=item wbuf_max => <bytes>
301
302If defined, then a fatal error will be raised (with C<$!> set to C<ENOSPC>)
303when the write buffer ever (strictly) exceeds this size. This is useful to
304avoid some forms of denial-of-service attacks.
305
306Although the units of this parameter is bytes, this is the I<raw> number
307of bytes not yet accepted by the kernel. This can make a difference when
308you e.g. use TLS, as TLS typically makes your write data larger (but it
309can also make it smaller due to compression).
310
311As an example of when this limit is useful, take a chat server that sends
312chat messages to a client. If the client does not read those in a timely
313manner then the send buffer in the server would grow unbounded.
314
185=item autocork => <boolean> 315=item autocork => <boolean>
186 316
187When disabled (the default), then C<push_write> will try to immediately 317When disabled (the default), C<push_write> will try to immediately
188write the data to the handle, if possible. This avoids having to register 318write the data to the handle if possible. This avoids having to register
189a write watcher and wait for the next event loop iteration, but can 319a write watcher and wait for the next event loop iteration, but can
190be inefficient if you write multiple small chunks (on the wire, this 320be inefficient if you write multiple small chunks (on the wire, this
191disadvantage is usually avoided by your kernel's nagle algorithm, see 321disadvantage is usually avoided by your kernel's nagle algorithm, see
192C<no_delay>, but this option can save costly syscalls). 322C<no_delay>, but this option can save costly syscalls).
193 323
194When enabled, then writes will always be queued till the next event loop 324When enabled, writes will always be queued till the next event loop
195iteration. This is efficient when you do many small writes per iteration, 325iteration. This is efficient when you do many small writes per iteration,
196but less efficient when you do a single write only per iteration (or when 326but less efficient when you do a single write only per iteration (or when
197the write buffer often is full). It also increases write latency. 327the write buffer often is full). It also increases write latency.
198 328
199=item no_delay => <boolean> 329=item no_delay => <boolean>
203the Nagle algorithm, and usually it is beneficial. 333the Nagle algorithm, and usually it is beneficial.
204 334
205In some situations you want as low a delay as possible, which can be 335In some situations you want as low a delay as possible, which can be
206accomplishd by setting this option to a true value. 336accomplishd by setting this option to a true value.
207 337
208The default is your opertaing system's default behaviour (most likely 338The default is your operating system's default behaviour (most likely
209enabled), this option explicitly enables or disables it, if possible. 339enabled). This option explicitly enables or disables it, if possible.
340
341=item keepalive => <boolean>
342
343Enables (default disable) the SO_KEEPALIVE option on the stream socket:
344normally, TCP connections have no time-out once established, so TCP
345connections, once established, can stay alive forever even when the other
346side has long gone. TCP keepalives are a cheap way to take down long-lived
347TCP connections when the other side becomes unreachable. While the default
348is OS-dependent, TCP keepalives usually kick in after around two hours,
349and, if the other side doesn't reply, take down the TCP connection some 10
350to 15 minutes later.
351
352It is harmless to specify this option for file handles that do not support
353keepalives, and enabling it on connections that are potentially long-lived
354is usually a good idea.
355
356=item oobinline => <boolean>
357
358BSD majorly fucked up the implementation of TCP urgent data. The result
359is that almost no OS implements TCP according to the specs, and every OS
360implements it slightly differently.
361
362If you want to handle TCP urgent data, then setting this flag (the default
363is enabled) gives you the most portable way of getting urgent data, by
364putting it into the stream.
365
366Since BSD emulation of OOB data on top of TCP's urgent data can have
367security implications, AnyEvent::Handle sets this flag automatically
368unless explicitly specified. Note that setting this flag after
369establishing a connection I<may> be a bit too late (data loss could
370already have occured on BSD systems), but at least it will protect you
371from most attacks.
210 372
211=item read_size => <bytes> 373=item read_size => <bytes>
212 374
213The default read block size (the amount of bytes this module will 375The initial read block size, the number of bytes this module will try
214try to read during each loop iteration, which affects memory 376to read during each loop iteration. Each handle object will consume
215requirements). Default: C<8192>. 377at least this amount of memory for the read buffer as well, so when
378handling many connections watch out for memory requirements). See also
379C<max_read_size>. Default: C<2048>.
380
381=item max_read_size => <bytes>
382
383The maximum read buffer size used by the dynamic adjustment
384algorithm: Each time AnyEvent::Handle can read C<read_size> bytes in
385one go it will double C<read_size> up to the maximum given by this
386option. Default: C<131072> or C<read_size>, whichever is higher.
216 387
217=item low_water_mark => <bytes> 388=item low_water_mark => <bytes>
218 389
219Sets the amount of bytes (default: C<0>) that make up an "empty" write 390Sets the number of bytes (default: C<0>) that make up an "empty" write
220buffer: If the write reaches this size or gets even samller it is 391buffer: If the buffer reaches this size or gets even samller it is
221considered empty. 392considered empty.
222 393
223Sometimes it can be beneficial (for performance reasons) to add data to 394Sometimes it can be beneficial (for performance reasons) to add data to
224the write buffer before it is fully drained, but this is a rare case, as 395the write buffer before it is fully drained, but this is a rare case, as
225the operating system kernel usually buffers data as well, so the default 396the operating system kernel usually buffers data as well, so the default
226is good in almost all cases. 397is good in almost all cases.
227 398
228=item linger => <seconds> 399=item linger => <seconds>
229 400
230If non-zero (default: C<3600>), then the destructor of the 401If this is non-zero (default: C<3600>), the destructor of the
231AnyEvent::Handle object will check whether there is still outstanding 402AnyEvent::Handle object will check whether there is still outstanding
232write data and will install a watcher that will write this data to the 403write data and will install a watcher that will write this data to the
233socket. No errors will be reported (this mostly matches how the operating 404socket. No errors will be reported (this mostly matches how the operating
234system treats outstanding data at socket close time). 405system treats outstanding data at socket close time).
235 406
236This will not work for partial TLS data that could not be encoded 407This will not work for partial TLS data that could not be encoded
237yet. This data will be lost. Calling the C<stoptls> method in time might 408yet. This data will be lost. Calling the C<stoptls> method in time might
238help. 409help.
239 410
411=item peername => $string
412
413A string used to identify the remote site - usually the DNS hostname
414(I<not> IDN!) used to create the connection, rarely the IP address.
415
416Apart from being useful in error messages, this string is also used in TLS
417peername verification (see C<verify_peername> in L<AnyEvent::TLS>). This
418verification will be skipped when C<peername> is not specified or is
419C<undef>.
420
240=item tls => "accept" | "connect" | Net::SSLeay::SSL object 421=item tls => "accept" | "connect" | Net::SSLeay::SSL object
241 422
242When this parameter is given, it enables TLS (SSL) mode, that means 423When this parameter is given, it enables TLS (SSL) mode, that means
243AnyEvent will start a TLS handshake as soon as the conenction has been 424AnyEvent will start a TLS handshake as soon as the connection has been
244established and will transparently encrypt/decrypt data afterwards. 425established and will transparently encrypt/decrypt data afterwards.
426
427All TLS protocol errors will be signalled as C<EPROTO>, with an
428appropriate error message.
245 429
246TLS mode requires Net::SSLeay to be installed (it will be loaded 430TLS mode requires Net::SSLeay to be installed (it will be loaded
247automatically when you try to create a TLS handle): this module doesn't 431automatically when you try to create a TLS handle): this module doesn't
248have a dependency on that module, so if your module requires it, you have 432have a dependency on that module, so if your module requires it, you have
249to add the dependency yourself. 433to add the dependency yourself. If Net::SSLeay cannot be loaded or is too
434old, you get an C<EPROTO> error.
250 435
251Unlike TCP, TLS has a server and client side: for the TLS server side, use 436Unlike TCP, TLS has a server and client side: for the TLS server side, use
252C<accept>, and for the TLS client side of a connection, use C<connect> 437C<accept>, and for the TLS client side of a connection, use C<connect>
253mode. 438mode.
254 439
255You can also provide your own TLS connection object, but you have 440You can also provide your own TLS connection object, but you have
256to make sure that you call either C<Net::SSLeay::set_connect_state> 441to make sure that you call either C<Net::SSLeay::set_connect_state>
257or C<Net::SSLeay::set_accept_state> on it before you pass it to 442or C<Net::SSLeay::set_accept_state> on it before you pass it to
258AnyEvent::Handle. 443AnyEvent::Handle. Also, this module will take ownership of this connection
444object.
445
446At some future point, AnyEvent::Handle might switch to another TLS
447implementation, then the option to use your own session object will go
448away.
259 449
260B<IMPORTANT:> since Net::SSLeay "objects" are really only integers, 450B<IMPORTANT:> since Net::SSLeay "objects" are really only integers,
261passing in the wrong integer will lead to certain crash. This most often 451passing in the wrong integer will lead to certain crash. This most often
262happens when one uses a stylish C<< tls => 1 >> and is surprised about the 452happens when one uses a stylish C<< tls => 1 >> and is surprised about the
263segmentation fault. 453segmentation fault.
264 454
265See the C<< ->starttls >> method for when need to start TLS negotiation later. 455Use the C<< ->starttls >> method if you need to start TLS negotiation later.
266 456
267=item tls_ctx => $ssl_ctx 457=item tls_ctx => $anyevent_tls
268 458
269Use the given C<Net::SSLeay::CTX> object to create the new TLS connection 459Use the given C<AnyEvent::TLS> object to create the new TLS connection
270(unless a connection object was specified directly). If this parameter is 460(unless a connection object was specified directly). If this
271missing, then AnyEvent::Handle will use C<AnyEvent::Handle::TLS_CTX>. 461parameter is missing (or C<undef>), then AnyEvent::Handle will use
462C<AnyEvent::Handle::TLS_CTX>.
272 463
464Instead of an object, you can also specify a hash reference with C<< key
465=> value >> pairs. Those will be passed to L<AnyEvent::TLS> to create a
466new TLS context object.
467
468=item on_starttls => $cb->($handle, $success[, $error_message])
469
470This callback will be invoked when the TLS/SSL handshake has finished. If
471C<$success> is true, then the TLS handshake succeeded, otherwise it failed
472(C<on_stoptls> will not be called in this case).
473
474The session in C<< $handle->{tls} >> can still be examined in this
475callback, even when the handshake was not successful.
476
477TLS handshake failures will not cause C<on_error> to be invoked when this
478callback is in effect, instead, the error message will be passed to C<on_starttls>.
479
480Without this callback, handshake failures lead to C<on_error> being
481called as usual.
482
483Note that you cannot just call C<starttls> again in this callback. If you
484need to do that, start an zero-second timer instead whose callback can
485then call C<< ->starttls >> again.
486
487=item on_stoptls => $cb->($handle)
488
489When a SSLv3/TLS shutdown/close notify/EOF is detected and this callback is
490set, then it will be invoked after freeing the TLS session. If it is not,
491then a TLS shutdown condition will be treated like a normal EOF condition
492on the handle.
493
494The session in C<< $handle->{tls} >> can still be examined in this
495callback.
496
497This callback will only be called on TLS shutdowns, not when the
498underlying handle signals EOF.
499
273=item json => JSON or JSON::XS object 500=item json => L<JSON>, L<JSON::PP> or L<JSON::XS> object
274 501
275This is the json coder object used by the C<json> read and write types. 502This is the json coder object used by the C<json> read and write types.
276 503
277If you don't supply it, then AnyEvent::Handle will create and use a 504If you don't supply it, then AnyEvent::Handle will create and use a
278suitable one (on demand), which will write and expect UTF-8 encoded JSON 505suitable one (on demand), which will write and expect UTF-8 encoded
506JSON texts (either using L<JSON::XS> or L<JSON>). The written texts are
507guaranteed not to contain any newline character.
508
509For security reasons, this encoder will likely I<not> handle numbers and
510strings, only arrays and objects/hashes. The reason is that originally
511JSON was self-delimited, but Dougles Crockford thought it was a splendid
512idea to redefine JSON incompatibly, so this is no longer true.
513
514For protocols that used back-to-back JSON texts, this might lead to
515run-ins, where two or more JSON texts will be interpreted as one JSON
279texts. 516text.
280 517
518For this reason, if the default encoder uses L<JSON::XS>, it will default
519to not allowing anything but arrays and objects/hashes, at least for the
520forseeable future (it will change at some point). This might or might not
521be true for the L<JSON> module, so this might cause a security issue.
522
523If you depend on either behaviour, you should create your own json object
524and pass it in explicitly.
525
526=item cbor => L<CBOR::XS> object
527
528This is the cbor coder object used by the C<cbor> read and write types.
529
530If you don't supply it, then AnyEvent::Handle will create and use a
531suitable one (on demand), which will write CBOR without using extensions,
532if possible.
533
281Note that you are responsible to depend on the JSON module if you want to 534Note that you are responsible to depend on the L<CBOR::XS> module if you
282use this functionality, as AnyEvent does not have a dependency itself. 535want to use this functionality, as AnyEvent does not have a dependency on
536it itself.
283 537
284=back 538=back
285 539
286=cut 540=cut
287 541
288sub new { 542sub new {
289 my $class = shift; 543 my $class = shift;
290
291 my $self = bless { @_ }, $class; 544 my $self = bless { @_ }, $class;
292 545
293 $self->{fh} or Carp::croak "mandatory argument fh is missing"; 546 if ($self->{fh}) {
547 $self->_start;
548 return unless $self->{fh}; # could be gone by now
294 549
550 } elsif ($self->{connect}) {
551 require AnyEvent::Socket;
552
553 $self->{peername} = $self->{connect}[0]
554 unless exists $self->{peername};
555
556 $self->{_skip_drain_rbuf} = 1;
557
558 {
559 Scalar::Util::weaken (my $self = $self);
560
561 $self->{_connect} =
562 AnyEvent::Socket::tcp_connect (
563 $self->{connect}[0],
564 $self->{connect}[1],
565 sub {
566 my ($fh, $host, $port, $retry) = @_;
567
568 delete $self->{_connect}; # no longer needed
569
570 if ($fh) {
571 $self->{fh} = $fh;
572
573 delete $self->{_skip_drain_rbuf};
574 $self->_start;
575
576 $self->{on_connect}
577 and $self->{on_connect}($self, $host, $port, sub {
578 delete @$self{qw(fh _tw _rtw _wtw _ww _rw _eof _queue rbuf _wbuf tls _tls_rbuf _tls_wbuf)};
579 $self->{_skip_drain_rbuf} = 1;
580 &$retry;
581 });
582
583 } else {
584 if ($self->{on_connect_error}) {
585 $self->{on_connect_error}($self, "$!");
586 $self->destroy if $self;
587 } else {
588 $self->_error ($!, 1);
589 }
590 }
591 },
592 sub {
593 local $self->{fh} = $_[0];
594
595 $self->{on_prepare}
596 ? $self->{on_prepare}->($self)
597 : ()
598 }
599 );
600 }
601
602 } else {
603 Carp::croak "AnyEvent::Handle: either an existing fh or the connect parameter must be specified";
604 }
605
606 $self
607}
608
609sub _start {
610 my ($self) = @_;
611
612 # too many clueless people try to use udp and similar sockets
613 # with AnyEvent::Handle, do them a favour.
614 my $type = getsockopt $self->{fh}, Socket::SOL_SOCKET (), Socket::SO_TYPE ();
615 Carp::croak "AnyEvent::Handle: only stream sockets supported, anything else will NOT work!"
616 if Socket::SOCK_STREAM () != (unpack "I", $type) && defined $type;
617
295 AnyEvent::Util::fh_nonblocking $self->{fh}, 1; 618 AnyEvent::fh_unblock $self->{fh};
296 619
620 $self->{_activity} =
621 $self->{_ractivity} =
622 $self->{_wactivity} = AE::now;
623
624 $self->{read_size} ||= 2048;
625 $self->{max_read_size} = $self->{read_size}
626 if $self->{read_size} > ($self->{max_read_size} || MAX_READ_SIZE);
627
628 $self->timeout (delete $self->{timeout} ) if $self->{timeout};
629 $self->rtimeout (delete $self->{rtimeout} ) if $self->{rtimeout};
630 $self->wtimeout (delete $self->{wtimeout} ) if $self->{wtimeout};
631
632 $self->no_delay (delete $self->{no_delay} ) if exists $self->{no_delay} && $self->{no_delay};
633 $self->keepalive (delete $self->{keepalive}) if exists $self->{keepalive} && $self->{keepalive};
634
635 $self->oobinline (exists $self->{oobinline} ? delete $self->{oobinline} : 1);
636
297 $self->starttls (delete $self->{tls}, delete $self->{tls_ctx}) 637 $self->starttls (delete $self->{tls}, delete $self->{tls_ctx})
298 if $self->{tls}; 638 if $self->{tls};
299 639
300 $self->{_activity} = AnyEvent->now;
301 $self->_timeout;
302
303 $self->on_drain (delete $self->{on_drain}) if exists $self->{on_drain}; 640 $self->on_drain (delete $self->{on_drain} ) if $self->{on_drain};
304 $self->no_delay (delete $self->{no_delay}) if exists $self->{no_delay};
305 641
306 $self->start_read 642 $self->start_read
307 if $self->{on_read}; 643 if $self->{on_read} || @{ $self->{_queue} };
308 644
309 $self 645 $self->_drain_wbuf;
310}
311
312sub _shutdown {
313 my ($self) = @_;
314
315 delete $self->{_tw};
316 delete $self->{_rw};
317 delete $self->{_ww};
318 delete $self->{fh};
319
320 &_freetls;
321
322 delete $self->{on_read};
323 delete $self->{_queue};
324} 646}
325 647
326sub _error { 648sub _error {
327 my ($self, $errno, $fatal) = @_; 649 my ($self, $errno, $fatal, $message) = @_;
328
329 $self->_shutdown
330 if $fatal;
331 650
332 $! = $errno; 651 $! = $errno;
652 $message ||= "$!";
333 653
334 if ($self->{on_error}) { 654 if ($self->{on_error}) {
335 $self->{on_error}($self, $fatal); 655 $self->{on_error}($self, $fatal, $message);
336 } elsif ($self->{fh}) { 656 $self->destroy if $fatal;
657 } elsif ($self->{fh} || $self->{connect}) {
658 $self->destroy;
337 Carp::croak "AnyEvent::Handle uncaught error: $!"; 659 Carp::croak "AnyEvent::Handle uncaught error: $message";
338 } 660 }
339} 661}
340 662
341=item $fh = $handle->fh 663=item $fh = $handle->fh
342 664
366 $_[0]{on_eof} = $_[1]; 688 $_[0]{on_eof} = $_[1];
367} 689}
368 690
369=item $handle->on_timeout ($cb) 691=item $handle->on_timeout ($cb)
370 692
371Replace the current C<on_timeout> callback, or disables the callback (but 693=item $handle->on_rtimeout ($cb)
372not the timeout) if C<$cb> = C<undef>. See the C<timeout> constructor
373argument and method.
374 694
375=cut 695=item $handle->on_wtimeout ($cb)
376 696
377sub on_timeout { 697Replace the current C<on_timeout>, C<on_rtimeout> or C<on_wtimeout>
378 $_[0]{on_timeout} = $_[1]; 698callback, or disables the callback (but not the timeout) if C<$cb> =
379} 699C<undef>. See the C<timeout> constructor argument and method.
700
701=cut
702
703# see below
380 704
381=item $handle->autocork ($boolean) 705=item $handle->autocork ($boolean)
382 706
383Enables or disables the current autocork behaviour (see C<autocork> 707Enables or disables the current autocork behaviour (see C<autocork>
384constructor argument). Changes will only take effect on the next write. 708constructor argument). Changes will only take effect on the next write.
397=cut 721=cut
398 722
399sub no_delay { 723sub no_delay {
400 $_[0]{no_delay} = $_[1]; 724 $_[0]{no_delay} = $_[1];
401 725
726 setsockopt $_[0]{fh}, Socket::IPPROTO_TCP (), Socket::TCP_NODELAY (), int $_[1]
727 if $_[0]{fh};
728}
729
730=item $handle->keepalive ($boolean)
731
732Enables or disables the C<keepalive> setting (see constructor argument of
733the same name for details).
734
735=cut
736
737sub keepalive {
738 $_[0]{keepalive} = $_[1];
739
402 eval { 740 eval {
403 local $SIG{__DIE__}; 741 local $SIG{__DIE__};
404 setsockopt $_[0]{fh}, &Socket::IPPROTO_TCP, &Socket::TCP_NODELAY, int $_[1]; 742 setsockopt $_[0]{fh}, Socket::SOL_SOCKET (), Socket::SO_KEEPALIVE (), int $_[1]
743 if $_[0]{fh};
405 }; 744 };
406} 745}
407 746
747=item $handle->oobinline ($boolean)
748
749Enables or disables the C<oobinline> setting (see constructor argument of
750the same name for details).
751
752=cut
753
754sub oobinline {
755 $_[0]{oobinline} = $_[1];
756
757 eval {
758 local $SIG{__DIE__};
759 setsockopt $_[0]{fh}, Socket::SOL_SOCKET (), Socket::SO_OOBINLINE (), int $_[1]
760 if $_[0]{fh};
761 };
762}
763
764=item $handle->keepalive ($boolean)
765
766Enables or disables the C<keepalive> setting (see constructor argument of
767the same name for details).
768
769=cut
770
771sub keepalive {
772 $_[0]{keepalive} = $_[1];
773
774 eval {
775 local $SIG{__DIE__};
776 setsockopt $_[0]{fh}, Socket::SOL_SOCKET (), Socket::SO_KEEPALIVE (), int $_[1]
777 if $_[0]{fh};
778 };
779}
780
781=item $handle->on_starttls ($cb)
782
783Replace the current C<on_starttls> callback (see the C<on_starttls> constructor argument).
784
785=cut
786
787sub on_starttls {
788 $_[0]{on_starttls} = $_[1];
789}
790
791=item $handle->on_stoptls ($cb)
792
793Replace the current C<on_stoptls> callback (see the C<on_stoptls> constructor argument).
794
795=cut
796
797sub on_stoptls {
798 $_[0]{on_stoptls} = $_[1];
799}
800
801=item $handle->rbuf_max ($max_octets)
802
803Configures the C<rbuf_max> setting (C<undef> disables it).
804
805=item $handle->wbuf_max ($max_octets)
806
807Configures the C<wbuf_max> setting (C<undef> disables it).
808
809=cut
810
811sub rbuf_max {
812 $_[0]{rbuf_max} = $_[1];
813}
814
815sub wbuf_max {
816 $_[0]{wbuf_max} = $_[1];
817}
818
408############################################################################# 819#############################################################################
409 820
410=item $handle->timeout ($seconds) 821=item $handle->timeout ($seconds)
411 822
823=item $handle->rtimeout ($seconds)
824
825=item $handle->wtimeout ($seconds)
826
412Configures (or disables) the inactivity timeout. 827Configures (or disables) the inactivity timeout.
413 828
414=cut 829The timeout will be checked instantly, so this method might destroy the
830handle before it returns.
415 831
416sub timeout { 832=item $handle->timeout_reset
833
834=item $handle->rtimeout_reset
835
836=item $handle->wtimeout_reset
837
838Reset the activity timeout, as if data was received or sent.
839
840These methods are cheap to call.
841
842=cut
843
844for my $dir ("", "r", "w") {
845 my $timeout = "${dir}timeout";
846 my $tw = "_${dir}tw";
847 my $on_timeout = "on_${dir}timeout";
848 my $activity = "_${dir}activity";
849 my $cb;
850
851 *$on_timeout = sub {
852 $_[0]{$on_timeout} = $_[1];
853 };
854
855 *$timeout = sub {
417 my ($self, $timeout) = @_; 856 my ($self, $new_value) = @_;
418 857
858 $new_value >= 0
859 or Carp::croak "AnyEvent::Handle->$timeout called with negative timeout ($new_value), caught";
860
419 $self->{timeout} = $timeout; 861 $self->{$timeout} = $new_value;
420 $self->_timeout; 862 delete $self->{$tw}; &$cb;
421} 863 };
422 864
865 *{"${dir}timeout_reset"} = sub {
866 $_[0]{$activity} = AE::now;
867 };
868
869 # main workhorse:
423# reset the timeout watcher, as neccessary 870 # reset the timeout watcher, as neccessary
424# also check for time-outs 871 # also check for time-outs
425sub _timeout { 872 $cb = sub {
426 my ($self) = @_; 873 my ($self) = @_;
427 874
428 if ($self->{timeout}) { 875 if ($self->{$timeout} && $self->{fh}) {
429 my $NOW = AnyEvent->now; 876 my $NOW = AE::now;
430 877
431 # when would the timeout trigger? 878 # when would the timeout trigger?
432 my $after = $self->{_activity} + $self->{timeout} - $NOW; 879 my $after = $self->{$activity} + $self->{$timeout} - $NOW;
433 880
434 # now or in the past already? 881 # now or in the past already?
435 if ($after <= 0) { 882 if ($after <= 0) {
436 $self->{_activity} = $NOW; 883 $self->{$activity} = $NOW;
437 884
438 if ($self->{on_timeout}) { 885 if ($self->{$on_timeout}) {
439 $self->{on_timeout}($self); 886 $self->{$on_timeout}($self);
440 } else { 887 } else {
441 $self->_error (&Errno::ETIMEDOUT); 888 $self->_error (Errno::ETIMEDOUT);
889 }
890
891 # callback could have changed timeout value, optimise
892 return unless $self->{$timeout};
893
894 # calculate new after
895 $after = $self->{$timeout};
442 } 896 }
443 897
444 # callback could have changed timeout value, optimise 898 Scalar::Util::weaken $self;
445 return unless $self->{timeout}; 899 return unless $self; # ->error could have destroyed $self
446 900
447 # calculate new after 901 $self->{$tw} ||= AE::timer $after, 0, sub {
448 $after = $self->{timeout}; 902 delete $self->{$tw};
903 $cb->($self);
904 };
905 } else {
906 delete $self->{$tw};
449 } 907 }
450
451 Scalar::Util::weaken $self;
452 return unless $self; # ->error could have destroyed $self
453
454 $self->{_tw} ||= AnyEvent->timer (after => $after, cb => sub {
455 delete $self->{_tw};
456 $self->_timeout;
457 });
458 } else {
459 delete $self->{_tw};
460 } 908 }
461} 909}
462 910
463############################################################################# 911#############################################################################
464 912
471 919
472The write queue is very simple: you can add data to its end, and 920The write queue is very simple: you can add data to its end, and
473AnyEvent::Handle will automatically try to get rid of it for you. 921AnyEvent::Handle will automatically try to get rid of it for you.
474 922
475When data could be written and the write buffer is shorter then the low 923When data could be written and the write buffer is shorter then the low
476water mark, the C<on_drain> callback will be invoked. 924water mark, the C<on_drain> callback will be invoked once.
477 925
478=over 4 926=over 4
479 927
480=item $handle->on_drain ($cb) 928=item $handle->on_drain ($cb)
481 929
482Sets the C<on_drain> callback or clears it (see the description of 930Sets the C<on_drain> callback or clears it (see the description of
483C<on_drain> in the constructor). 931C<on_drain> in the constructor).
484 932
933This method may invoke callbacks (and therefore the handle might be
934destroyed after it returns).
935
485=cut 936=cut
486 937
487sub on_drain { 938sub on_drain {
488 my ($self, $cb) = @_; 939 my ($self, $cb) = @_;
489 940
493 if $cb && $self->{low_water_mark} >= (length $self->{wbuf}) + (length $self->{_tls_wbuf}); 944 if $cb && $self->{low_water_mark} >= (length $self->{wbuf}) + (length $self->{_tls_wbuf});
494} 945}
495 946
496=item $handle->push_write ($data) 947=item $handle->push_write ($data)
497 948
498Queues the given scalar to be written. You can push as much data as you 949Queues the given scalar to be written. You can push as much data as
499want (only limited by the available memory), as C<AnyEvent::Handle> 950you want (only limited by the available memory and C<wbuf_max>), as
500buffers it independently of the kernel. 951C<AnyEvent::Handle> buffers it independently of the kernel.
952
953This method may invoke callbacks (and therefore the handle might be
954destroyed after it returns).
501 955
502=cut 956=cut
503 957
504sub _drain_wbuf { 958sub _drain_wbuf {
505 my ($self) = @_; 959 my ($self) = @_;
509 Scalar::Util::weaken $self; 963 Scalar::Util::weaken $self;
510 964
511 my $cb = sub { 965 my $cb = sub {
512 my $len = syswrite $self->{fh}, $self->{wbuf}; 966 my $len = syswrite $self->{fh}, $self->{wbuf};
513 967
514 if ($len >= 0) { 968 if (defined $len) {
515 substr $self->{wbuf}, 0, $len, ""; 969 substr $self->{wbuf}, 0, $len, "";
516 970
517 $self->{_activity} = AnyEvent->now; 971 $self->{_activity} = $self->{_wactivity} = AE::now;
518 972
519 $self->{on_drain}($self) 973 $self->{on_drain}($self)
520 if $self->{low_water_mark} >= (length $self->{wbuf}) + (length $self->{_tls_wbuf}) 974 if $self->{low_water_mark} >= (length $self->{wbuf}) + (length $self->{_tls_wbuf})
521 && $self->{on_drain}; 975 && $self->{on_drain};
522 976
523 delete $self->{_ww} unless length $self->{wbuf}; 977 delete $self->{_ww} unless length $self->{wbuf};
524 } elsif ($! != EAGAIN && $! != EINTR && $! != WSAEWOULDBLOCK) { 978 } elsif ($! != EAGAIN && $! != EINTR && $! != EWOULDBLOCK && $! != WSAEWOULDBLOCK) {
525 $self->_error ($!, 1); 979 $self->_error ($!, 1);
526 } 980 }
527 }; 981 };
528 982
529 # try to write data immediately 983 # try to write data immediately
530 $cb->() unless $self->{autocork}; 984 $cb->() unless $self->{autocork};
531 985
532 # if still data left in wbuf, we need to poll 986 # if still data left in wbuf, we need to poll
533 $self->{_ww} = AnyEvent->io (fh => $self->{fh}, poll => "w", cb => $cb) 987 $self->{_ww} = AE::io $self->{fh}, 1, $cb
534 if length $self->{wbuf}; 988 if length $self->{wbuf};
989
990 if (
991 defined $self->{wbuf_max}
992 && $self->{wbuf_max} < length $self->{wbuf}
993 ) {
994 $self->_error (Errno::ENOSPC, 1), return;
995 }
535 }; 996 };
536} 997}
537 998
538our %WH; 999our %WH;
539 1000
1001# deprecated
540sub register_write_type($$) { 1002sub register_write_type($$) {
541 $WH{$_[0]} = $_[1]; 1003 $WH{$_[0]} = $_[1];
542} 1004}
543 1005
544sub push_write { 1006sub push_write {
545 my $self = shift; 1007 my $self = shift;
546 1008
547 if (@_ > 1) { 1009 if (@_ > 1) {
548 my $type = shift; 1010 my $type = shift;
549 1011
1012 @_ = ($WH{$type} ||= _load_func "$type\::anyevent_write_type"
550 @_ = ($WH{$type} or Carp::croak "unsupported type passed to AnyEvent::Handle::push_write") 1013 or Carp::croak "unsupported/unloadable type '$type' passed to AnyEvent::Handle::push_write")
551 ->($self, @_); 1014 ->($self, @_);
552 } 1015 }
553 1016
1017 # we downgrade here to avoid hard-to-track-down bugs,
1018 # and diagnose the problem earlier and better.
1019
554 if ($self->{tls}) { 1020 if ($self->{tls}) {
555 $self->{_tls_wbuf} .= $_[0]; 1021 utf8::downgrade $self->{_tls_wbuf} .= $_[0];
556 1022 &_dotls ($self) if $self->{fh};
557 &_dotls ($self);
558 } else { 1023 } else {
559 $self->{wbuf} .= $_[0]; 1024 utf8::downgrade $self->{wbuf} .= $_[0];
560 $self->_drain_wbuf; 1025 $self->_drain_wbuf if $self->{fh};
561 } 1026 }
562} 1027}
563 1028
564=item $handle->push_write (type => @args) 1029=item $handle->push_write (type => @args)
565 1030
566Instead of formatting your data yourself, you can also let this module do 1031Instead of formatting your data yourself, you can also let this module
567the job by specifying a type and type-specific arguments. 1032do the job by specifying a type and type-specific arguments. You
1033can also specify the (fully qualified) name of a package, in which
1034case AnyEvent tries to load the package and then expects to find the
1035C<anyevent_write_type> function inside (see "custom write types", below).
568 1036
569Predefined types are (if you have ideas for additional types, feel free to 1037Predefined types are (if you have ideas for additional types, feel free to
570drop by and tell us): 1038drop by and tell us):
571 1039
572=over 4 1040=over 4
603 1071
604Encodes the given hash or array reference into a JSON object. Unless you 1072Encodes the given hash or array reference into a JSON object. Unless you
605provide your own JSON object, this means it will be encoded to JSON text 1073provide your own JSON object, this means it will be encoded to JSON text
606in UTF-8. 1074in UTF-8.
607 1075
1076The default encoder might or might not handle every type of JSON value -
1077it might be limited to arrays and objects for security reasons. See the
1078C<json> constructor attribute for more details.
1079
608JSON objects (and arrays) are self-delimiting, so you can write JSON at 1080JSON objects (and arrays) are self-delimiting, so if you only use arrays
609one end of a handle and read them at the other end without using any 1081and hashes, you can write JSON at one end of a handle and read them at the
610additional framing. 1082other end without using any additional framing.
611 1083
612The generated JSON text is guaranteed not to contain any newlines: While 1084The JSON text generated by the default encoder is guaranteed not to
613this module doesn't need delimiters after or between JSON texts to be 1085contain any newlines: While this module doesn't need delimiters after or
614able to read them, many other languages depend on that. 1086between JSON texts to be able to read them, many other languages depend on
1087them.
615 1088
616A simple RPC protocol that interoperates easily with others is to send 1089A simple RPC protocol that interoperates easily with other languages is
617JSON arrays (or objects, although arrays are usually the better choice as 1090to send JSON arrays (or objects, although arrays are usually the better
618they mimic how function argument passing works) and a newline after each 1091choice as they mimic how function argument passing works) and a newline
619JSON text: 1092after each JSON text:
620 1093
621 $handle->push_write (json => ["method", "arg1", "arg2"]); # whatever 1094 $handle->push_write (json => ["method", "arg1", "arg2"]); # whatever
622 $handle->push_write ("\012"); 1095 $handle->push_write ("\012");
623 1096
624An AnyEvent::Handle receiver would simply use the C<json> read type and 1097An AnyEvent::Handle receiver would simply use the C<json> read type and
627 $handle->push_read (json => sub { my $array = $_[1]; ... }); 1100 $handle->push_read (json => sub { my $array = $_[1]; ... });
628 1101
629Other languages could read single lines terminated by a newline and pass 1102Other languages could read single lines terminated by a newline and pass
630this line into their JSON decoder of choice. 1103this line into their JSON decoder of choice.
631 1104
1105=item cbor => $perl_scalar
1106
1107Encodes the given scalar into a CBOR value. Unless you provide your own
1108L<CBOR::XS> object, this means it will be encoded to a CBOR string not
1109using any extensions, if possible.
1110
1111CBOR values are self-delimiting, so you can write CBOR at one end of
1112a handle and read them at the other end without using any additional
1113framing.
1114
1115A simple nd very very fast RPC protocol that interoperates with
1116other languages is to send CBOR and receive CBOR values (arrays are
1117recommended):
1118
1119 $handle->push_write (cbor => ["method", "arg1", "arg2"]); # whatever
1120
1121An AnyEvent::Handle receiver would simply use the C<cbor> read type:
1122
1123 $handle->push_read (cbor => sub { my $array = $_[1]; ... });
1124
632=cut 1125=cut
1126
1127sub json_coder() {
1128 eval { require JSON::XS; JSON::XS->new->utf8 }
1129 || do { require JSON::PP; JSON::PP->new->utf8 }
1130}
633 1131
634register_write_type json => sub { 1132register_write_type json => sub {
635 my ($self, $ref) = @_; 1133 my ($self, $ref) = @_;
636 1134
637 require JSON; 1135 ($self->{json} ||= json_coder)
1136 ->encode ($ref)
1137};
638 1138
639 $self->{json} ? $self->{json}->encode ($ref) 1139sub cbor_coder() {
640 : JSON::encode_json ($ref) 1140 require CBOR::XS;
1141 CBOR::XS->new
1142}
1143
1144register_write_type cbor => sub {
1145 my ($self, $scalar) = @_;
1146
1147 ($self->{cbor} ||= cbor_coder)
1148 ->encode ($scalar)
641}; 1149};
642 1150
643=item storable => $reference 1151=item storable => $reference
644 1152
645Freezes the given reference using L<Storable> and writes it to the 1153Freezes the given reference using L<Storable> and writes it to the
648=cut 1156=cut
649 1157
650register_write_type storable => sub { 1158register_write_type storable => sub {
651 my ($self, $ref) = @_; 1159 my ($self, $ref) = @_;
652 1160
653 require Storable; 1161 require Storable unless $Storable::VERSION;
654 1162
655 pack "w/a*", Storable::nfreeze ($ref) 1163 pack "w/a*", Storable::nfreeze ($ref)
656}; 1164};
657 1165
658=back 1166=back
659 1167
660=item AnyEvent::Handle::register_write_type type => $coderef->($handle, @args) 1168=item $handle->push_shutdown
661 1169
662This function (not method) lets you add your own types to C<push_write>. 1170Sometimes you know you want to close the socket after writing your data
1171before it was actually written. One way to do that is to replace your
1172C<on_drain> handler by a callback that shuts down the socket (and set
1173C<low_water_mark> to C<0>). This method is a shorthand for just that, and
1174replaces the C<on_drain> callback with:
1175
1176 sub { shutdown $_[0]{fh}, 1 }
1177
1178This simply shuts down the write side and signals an EOF condition to the
1179the peer.
1180
1181You can rely on the normal read queue and C<on_eof> handling
1182afterwards. This is the cleanest way to close a connection.
1183
1184This method may invoke callbacks (and therefore the handle might be
1185destroyed after it returns).
1186
1187=cut
1188
1189sub push_shutdown {
1190 my ($self) = @_;
1191
1192 delete $self->{low_water_mark};
1193 $self->on_drain (sub { shutdown $_[0]{fh}, 1 });
1194}
1195
1196=item custom write types - Package::anyevent_write_type $handle, @args
1197
1198Instead of one of the predefined types, you can also specify the name of
1199a package. AnyEvent will try to load the package and then expects to find
1200a function named C<anyevent_write_type> inside. If it isn't found, it
1201progressively tries to load the parent package until it either finds the
1202function (good) or runs out of packages (bad).
1203
663Whenever the given C<type> is used, C<push_write> will invoke the code 1204Whenever the given C<type> is used, C<push_write> will the function with
664reference with the handle object and the remaining arguments. 1205the handle object and the remaining arguments.
665 1206
666The code reference is supposed to return a single octet string that will 1207The function is supposed to return a single octet string that will be
667be appended to the write buffer. 1208appended to the write buffer, so you can mentally treat this function as a
1209"arguments to on-the-wire-format" converter.
668 1210
669Note that this is a function, and all types registered this way will be 1211Example: implement a custom write type C<join> that joins the remaining
670global, so try to use unique names. 1212arguments using the first one.
1213
1214 $handle->push_write (My::Type => " ", 1,2,3);
1215
1216 # uses the following package, which can be defined in the "My::Type" or in
1217 # the "My" modules to be auto-loaded, or just about anywhere when the
1218 # My::Type::anyevent_write_type is defined before invoking it.
1219
1220 package My::Type;
1221
1222 sub anyevent_write_type {
1223 my ($handle, $delim, @args) = @_;
1224
1225 join $delim, @args
1226 }
671 1227
672=cut 1228=cut
673 1229
674############################################################################# 1230#############################################################################
675 1231
684ways, the "simple" way, using only C<on_read> and the "complex" way, using 1240ways, the "simple" way, using only C<on_read> and the "complex" way, using
685a queue. 1241a queue.
686 1242
687In the simple case, you just install an C<on_read> callback and whenever 1243In the simple case, you just install an C<on_read> callback and whenever
688new data arrives, it will be called. You can then remove some data (if 1244new data arrives, it will be called. You can then remove some data (if
689enough is there) from the read buffer (C<< $handle->rbuf >>). Or you cna 1245enough is there) from the read buffer (C<< $handle->rbuf >>). Or you can
690leave the data there if you want to accumulate more (e.g. when only a 1246leave the data there if you want to accumulate more (e.g. when only a
691partial message has been received so far). 1247partial message has been received so far), or change the read queue with
1248e.g. C<push_read>.
692 1249
693In the more complex case, you want to queue multiple callbacks. In this 1250In the more complex case, you want to queue multiple callbacks. In this
694case, AnyEvent::Handle will call the first queued callback each time new 1251case, AnyEvent::Handle will call the first queued callback each time new
695data arrives (also the first time it is queued) and removes it when it has 1252data arrives (also the first time it is queued) and remove it when it has
696done its job (see C<push_read>, below). 1253done its job (see C<push_read>, below).
697 1254
698This way you can, for example, push three line-reads, followed by reading 1255This way you can, for example, push three line-reads, followed by reading
699a chunk of data, and AnyEvent::Handle will execute them in order. 1256a chunk of data, and AnyEvent::Handle will execute them in order.
700 1257
757=cut 1314=cut
758 1315
759sub _drain_rbuf { 1316sub _drain_rbuf {
760 my ($self) = @_; 1317 my ($self) = @_;
761 1318
1319 # avoid recursion
1320 return if $self->{_skip_drain_rbuf};
762 local $self->{_in_drain} = 1; 1321 local $self->{_skip_drain_rbuf} = 1;
763
764 if (
765 defined $self->{rbuf_max}
766 && $self->{rbuf_max} < length $self->{rbuf}
767 ) {
768 $self->_error (&Errno::ENOSPC, 1), return;
769 }
770 1322
771 while () { 1323 while () {
772 $self->{rbuf} .= delete $self->{tls_rbuf} if exists $self->{tls_rbuf};#d# 1324 # we need to use a separate tls read buffer, as we must not receive data while
1325 # we are draining the buffer, and this can only happen with TLS.
1326 $self->{rbuf} .= delete $self->{_tls_rbuf}
1327 if exists $self->{_tls_rbuf};
773 1328
774 my $len = length $self->{rbuf}; 1329 my $len = length $self->{rbuf};
775 1330
776 if (my $cb = shift @{ $self->{_queue} }) { 1331 if (my $cb = shift @{ $self->{_queue} }) {
777 unless ($cb->($self)) { 1332 unless ($cb->($self)) {
778 if ($self->{_eof}) { 1333 # no progress can be made
779 # no progress can be made (not enough data and no data forthcoming) 1334 # (not enough data and no data forthcoming)
780 $self->_error (&Errno::EPIPE, 1), return; 1335 $self->_error (Errno::EPIPE, 1), return
781 } 1336 if $self->{_eof};
782 1337
783 unshift @{ $self->{_queue} }, $cb; 1338 unshift @{ $self->{_queue} }, $cb;
784 last; 1339 last;
785 } 1340 }
786 } elsif ($self->{on_read}) { 1341 } elsif ($self->{on_read}) {
793 && !@{ $self->{_queue} } # and the queue is still empty 1348 && !@{ $self->{_queue} } # and the queue is still empty
794 && $self->{on_read} # but we still have on_read 1349 && $self->{on_read} # but we still have on_read
795 ) { 1350 ) {
796 # no further data will arrive 1351 # no further data will arrive
797 # so no progress can be made 1352 # so no progress can be made
798 $self->_error (&Errno::EPIPE, 1), return 1353 $self->_error (Errno::EPIPE, 1), return
799 if $self->{_eof}; 1354 if $self->{_eof};
800 1355
801 last; # more data might arrive 1356 last; # more data might arrive
802 } 1357 }
803 } else { 1358 } else {
806 last; 1361 last;
807 } 1362 }
808 } 1363 }
809 1364
810 if ($self->{_eof}) { 1365 if ($self->{_eof}) {
811 if ($self->{on_eof}) { 1366 $self->{on_eof}
812 $self->{on_eof}($self) 1367 ? $self->{on_eof}($self)
813 } else { 1368 : $self->_error (0, 1, "Unexpected end-of-file");
814 $self->_error (0, 1); 1369
815 } 1370 return;
1371 }
1372
1373 if (
1374 defined $self->{rbuf_max}
1375 && $self->{rbuf_max} < length $self->{rbuf}
1376 ) {
1377 $self->_error (Errno::ENOSPC, 1), return;
816 } 1378 }
817 1379
818 # may need to restart read watcher 1380 # may need to restart read watcher
819 unless ($self->{_rw}) { 1381 unless ($self->{_rw}) {
820 $self->start_read 1382 $self->start_read
826 1388
827This replaces the currently set C<on_read> callback, or clears it (when 1389This replaces the currently set C<on_read> callback, or clears it (when
828the new callback is C<undef>). See the description of C<on_read> in the 1390the new callback is C<undef>). See the description of C<on_read> in the
829constructor. 1391constructor.
830 1392
1393This method may invoke callbacks (and therefore the handle might be
1394destroyed after it returns).
1395
831=cut 1396=cut
832 1397
833sub on_read { 1398sub on_read {
834 my ($self, $cb) = @_; 1399 my ($self, $cb) = @_;
835 1400
836 $self->{on_read} = $cb; 1401 $self->{on_read} = $cb;
837 $self->_drain_rbuf if $cb && !$self->{_in_drain}; 1402 $self->_drain_rbuf if $cb;
838} 1403}
839 1404
840=item $handle->rbuf 1405=item $handle->rbuf
841 1406
842Returns the read buffer (as a modifiable lvalue). 1407Returns the read buffer (as a modifiable lvalue). You can also access the
1408read buffer directly as the C<< ->{rbuf} >> member, if you want (this is
1409much faster, and no less clean).
843 1410
844You can access the read buffer directly as the C<< ->{rbuf} >> member, if 1411The only operation allowed on the read buffer (apart from looking at it)
845you want. 1412is removing data from its beginning. Otherwise modifying or appending to
1413it is not allowed and will lead to hard-to-track-down bugs.
846 1414
847NOTE: The read buffer should only be used or modified if the C<on_read>, 1415NOTE: The read buffer should only be used or modified in the C<on_read>
848C<push_read> or C<unshift_read> methods are used. The other read methods 1416callback or when C<push_read> or C<unshift_read> are used with a single
849automatically manage the read buffer. 1417callback (i.e. untyped). Typed C<push_read> and C<unshift_read> methods
1418will manage the read buffer on their own.
850 1419
851=cut 1420=cut
852 1421
853sub rbuf : lvalue { 1422sub rbuf : lvalue {
854 $_[0]{rbuf} 1423 $_[0]{rbuf}
871 1440
872If enough data was available, then the callback must remove all data it is 1441If enough data was available, then the callback must remove all data it is
873interested in (which can be none at all) and return a true value. After returning 1442interested in (which can be none at all) and return a true value. After returning
874true, it will be removed from the queue. 1443true, it will be removed from the queue.
875 1444
1445These methods may invoke callbacks (and therefore the handle might be
1446destroyed after it returns).
1447
876=cut 1448=cut
877 1449
878our %RH; 1450our %RH;
879 1451
880sub register_read_type($$) { 1452sub register_read_type($$) {
886 my $cb = pop; 1458 my $cb = pop;
887 1459
888 if (@_) { 1460 if (@_) {
889 my $type = shift; 1461 my $type = shift;
890 1462
1463 $cb = ($RH{$type} ||= _load_func "$type\::anyevent_read_type"
891 $cb = ($RH{$type} or Carp::croak "unsupported type passed to AnyEvent::Handle::push_read") 1464 or Carp::croak "unsupported/unloadable type '$type' passed to AnyEvent::Handle::push_read")
892 ->($self, $cb, @_); 1465 ->($self, $cb, @_);
893 } 1466 }
894 1467
895 push @{ $self->{_queue} }, $cb; 1468 push @{ $self->{_queue} }, $cb;
896 $self->_drain_rbuf unless $self->{_in_drain}; 1469 $self->_drain_rbuf;
897} 1470}
898 1471
899sub unshift_read { 1472sub unshift_read {
900 my $self = shift; 1473 my $self = shift;
901 my $cb = pop; 1474 my $cb = pop;
902 1475
903 if (@_) { 1476 if (@_) {
904 my $type = shift; 1477 my $type = shift;
905 1478
1479 $cb = ($RH{$type} ||= _load_func "$type\::anyevent_read_type"
906 $cb = ($RH{$type} or Carp::croak "unsupported type passed to AnyEvent::Handle::unshift_read") 1480 or Carp::croak "unsupported/unloadable type '$type' passed to AnyEvent::Handle::unshift_read")
907 ->($self, $cb, @_); 1481 ->($self, $cb, @_);
908 } 1482 }
909 1483
910
911 unshift @{ $self->{_queue} }, $cb; 1484 unshift @{ $self->{_queue} }, $cb;
912 $self->_drain_rbuf unless $self->{_in_drain}; 1485 $self->_drain_rbuf;
913} 1486}
914 1487
915=item $handle->push_read (type => @args, $cb) 1488=item $handle->push_read (type => @args, $cb)
916 1489
917=item $handle->unshift_read (type => @args, $cb) 1490=item $handle->unshift_read (type => @args, $cb)
918 1491
919Instead of providing a callback that parses the data itself you can chose 1492Instead of providing a callback that parses the data itself you can chose
920between a number of predefined parsing formats, for chunks of data, lines 1493between a number of predefined parsing formats, for chunks of data, lines
921etc. 1494etc. You can also specify the (fully qualified) name of a package, in
1495which case AnyEvent tries to load the package and then expects to find the
1496C<anyevent_read_type> function inside (see "custom read types", below).
922 1497
923Predefined types are (if you have ideas for additional types, feel free to 1498Predefined types are (if you have ideas for additional types, feel free to
924drop by and tell us): 1499drop by and tell us):
925 1500
926=over 4 1501=over 4
932data. 1507data.
933 1508
934Example: read 2 bytes. 1509Example: read 2 bytes.
935 1510
936 $handle->push_read (chunk => 2, sub { 1511 $handle->push_read (chunk => 2, sub {
937 warn "yay ", unpack "H*", $_[1]; 1512 say "yay " . unpack "H*", $_[1];
938 }); 1513 });
939 1514
940=cut 1515=cut
941 1516
942register_read_type chunk => sub { 1517register_read_type chunk => sub {
972 1547
973register_read_type line => sub { 1548register_read_type line => sub {
974 my ($self, $cb, $eol) = @_; 1549 my ($self, $cb, $eol) = @_;
975 1550
976 if (@_ < 3) { 1551 if (@_ < 3) {
977 # this is more than twice as fast as the generic code below 1552 # this is faster then the generic code below
978 sub { 1553 sub {
979 $_[0]{rbuf} =~ s/^([^\015\012]*)(\015?\012)// or return; 1554 (my $pos = index $_[0]{rbuf}, "\012") >= 0
1555 or return;
980 1556
1557 (my $str = substr $_[0]{rbuf}, 0, $pos + 1, "") =~ s/(\015?\012)\Z// or die;
981 $cb->($_[0], $1, $2); 1558 $cb->($_[0], $str, "$1");
982 1 1559 1
983 } 1560 }
984 } else { 1561 } else {
985 $eol = quotemeta $eol unless ref $eol; 1562 $eol = quotemeta $eol unless ref $eol;
986 $eol = qr|^(.*?)($eol)|s; 1563 $eol = qr|^(.*?)($eol)|s;
987 1564
988 sub { 1565 sub {
989 $_[0]{rbuf} =~ s/$eol// or return; 1566 $_[0]{rbuf} =~ s/$eol// or return;
990 1567
991 $cb->($_[0], $1, $2); 1568 $cb->($_[0], "$1", "$2");
992 1 1569 1
993 } 1570 }
994 } 1571 }
995}; 1572};
996 1573
997=item regex => $accept[, $reject[, $skip], $cb->($handle, $data) 1574=item regex => $accept[, $reject[, $skip], $cb->($handle, $data)
998 1575
999Makes a regex match against the regex object C<$accept> and returns 1576Makes a regex match against the regex object C<$accept> and returns
1000everything up to and including the match. 1577everything up to and including the match. All the usual regex variables
1578($1, %+ etc.) from the regex match are available in the callback.
1001 1579
1002Example: read a single line terminated by '\n'. 1580Example: read a single line terminated by '\n'.
1003 1581
1004 $handle->push_read (regex => qr<\n>, sub { ... }); 1582 $handle->push_read (regex => qr<\n>, sub { ... });
1005 1583
1018the receive buffer when neither C<$accept> nor C<$reject> match, 1596the receive buffer when neither C<$accept> nor C<$reject> match,
1019and everything preceding and including the match will be accepted 1597and everything preceding and including the match will be accepted
1020unconditionally. This is useful to skip large amounts of data that you 1598unconditionally. This is useful to skip large amounts of data that you
1021know cannot be matched, so that the C<$accept> or C<$reject> regex do not 1599know cannot be matched, so that the C<$accept> or C<$reject> regex do not
1022have to start matching from the beginning. This is purely an optimisation 1600have to start matching from the beginning. This is purely an optimisation
1023and is usually worth only when you expect more than a few kilobytes. 1601and is usually worth it only when you expect more than a few kilobytes.
1024 1602
1025Example: expect a http header, which ends at C<\015\012\015\012>. Since we 1603Example: expect a http header, which ends at C<\015\012\015\012>. Since we
1026expect the header to be very large (it isn't in practise, but...), we use 1604expect the header to be very large (it isn't in practice, but...), we use
1027a skip regex to skip initial portions. The skip regex is tricky in that 1605a skip regex to skip initial portions. The skip regex is tricky in that
1028it only accepts something not ending in either \015 or \012, as these are 1606it only accepts something not ending in either \015 or \012, as these are
1029required for the accept regex. 1607required for the accept regex.
1030 1608
1031 $handle->push_read (regex => 1609 $handle->push_read (regex =>
1044 1622
1045 sub { 1623 sub {
1046 # accept 1624 # accept
1047 if ($$rbuf =~ $accept) { 1625 if ($$rbuf =~ $accept) {
1048 $data .= substr $$rbuf, 0, $+[0], ""; 1626 $data .= substr $$rbuf, 0, $+[0], "";
1049 $cb->($self, $data); 1627 $cb->($_[0], $data);
1050 return 1; 1628 return 1;
1051 } 1629 }
1052 1630
1053 # reject 1631 # reject
1054 if ($reject && $$rbuf =~ $reject) { 1632 if ($reject && $$rbuf =~ $reject) {
1055 $self->_error (&Errno::EBADMSG); 1633 $_[0]->_error (Errno::EBADMSG);
1056 } 1634 }
1057 1635
1058 # skip 1636 # skip
1059 if ($skip && $$rbuf =~ $skip) { 1637 if ($skip && $$rbuf =~ $skip) {
1060 $data .= substr $$rbuf, 0, $+[0], ""; 1638 $data .= substr $$rbuf, 0, $+[0], "";
1076 my ($self, $cb) = @_; 1654 my ($self, $cb) = @_;
1077 1655
1078 sub { 1656 sub {
1079 unless ($_[0]{rbuf} =~ s/^(0|[1-9][0-9]*)://) { 1657 unless ($_[0]{rbuf} =~ s/^(0|[1-9][0-9]*)://) {
1080 if ($_[0]{rbuf} =~ /[^0-9]/) { 1658 if ($_[0]{rbuf} =~ /[^0-9]/) {
1081 $self->_error (&Errno::EBADMSG); 1659 $_[0]->_error (Errno::EBADMSG);
1082 } 1660 }
1083 return; 1661 return;
1084 } 1662 }
1085 1663
1086 my $len = $1; 1664 my $len = $1;
1087 1665
1088 $self->unshift_read (chunk => $len, sub { 1666 $_[0]->unshift_read (chunk => $len, sub {
1089 my $string = $_[1]; 1667 my $string = $_[1];
1090 $_[0]->unshift_read (chunk => 1, sub { 1668 $_[0]->unshift_read (chunk => 1, sub {
1091 if ($_[1] eq ",") { 1669 if ($_[1] eq ",") {
1092 $cb->($_[0], $string); 1670 $cb->($_[0], $string);
1093 } else { 1671 } else {
1094 $self->_error (&Errno::EBADMSG); 1672 $_[0]->_error (Errno::EBADMSG);
1095 } 1673 }
1096 }); 1674 });
1097 }); 1675 });
1098 1676
1099 1 1677 1
1149=item json => $cb->($handle, $hash_or_arrayref) 1727=item json => $cb->($handle, $hash_or_arrayref)
1150 1728
1151Reads a JSON object or array, decodes it and passes it to the 1729Reads a JSON object or array, decodes it and passes it to the
1152callback. When a parse error occurs, an C<EBADMSG> error will be raised. 1730callback. When a parse error occurs, an C<EBADMSG> error will be raised.
1153 1731
1154If a C<json> object was passed to the constructor, then that will be used 1732If a C<json> object was passed to the constructor, then that will be
1155for the final decode, otherwise it will create a JSON coder expecting UTF-8. 1733used for the final decode, otherwise it will create a L<JSON::XS> or
1734L<JSON::PP> coder object expecting UTF-8.
1156 1735
1157This read type uses the incremental parser available with JSON version 1736This read type uses the incremental parser available with JSON version
11582.09 (and JSON::XS version 2.2) and above. You have to provide a 17372.09 (and JSON::XS version 2.2) and above.
1159dependency on your own: this module will load the JSON module, but
1160AnyEvent does not depend on it itself.
1161 1738
1162Since JSON texts are fully self-delimiting, the C<json> read and write 1739Since JSON texts are fully self-delimiting, the C<json> read and write
1163types are an ideal simple RPC protocol: just exchange JSON datagrams. See 1740types are an ideal simple RPC protocol: just exchange JSON datagrams. See
1164the C<json> write type description, above, for an actual example. 1741the C<json> write type description, above, for an actual example.
1165 1742
1166=cut 1743=cut
1167 1744
1168register_read_type json => sub { 1745register_read_type json => sub {
1169 my ($self, $cb) = @_; 1746 my ($self, $cb) = @_;
1170 1747
1171 require JSON; 1748 my $json = $self->{json} ||= json_coder;
1172 1749
1173 my $data; 1750 my $data;
1174 my $rbuf = \$self->{rbuf};
1175
1176 my $json = $self->{json} ||= JSON->new->utf8;
1177 1751
1178 sub { 1752 sub {
1179 my $ref = eval { $json->incr_parse ($self->{rbuf}) }; 1753 my $ref = eval { $json->incr_parse ($_[0]{rbuf}) };
1180 1754
1181 if ($ref) { 1755 if ($ref) {
1182 $self->{rbuf} = $json->incr_text; 1756 $_[0]{rbuf} = $json->incr_text;
1183 $json->incr_text = ""; 1757 $json->incr_text = "";
1184 $cb->($self, $ref); 1758 $cb->($_[0], $ref);
1185 1759
1186 1 1760 1
1187 } elsif ($@) { 1761 } elsif ($@) {
1188 # error case 1762 # error case
1189 $json->incr_skip; 1763 $json->incr_skip;
1190 1764
1191 $self->{rbuf} = $json->incr_text; 1765 $_[0]{rbuf} = $json->incr_text;
1192 $json->incr_text = ""; 1766 $json->incr_text = "";
1193 1767
1194 $self->_error (&Errno::EBADMSG); 1768 $_[0]->_error (Errno::EBADMSG);
1195 1769
1196 () 1770 ()
1197 } else { 1771 } else {
1198 $self->{rbuf} = ""; 1772 $_[0]{rbuf} = "";
1199 1773
1774 ()
1775 }
1776 }
1777};
1778
1779=item cbor => $cb->($handle, $scalar)
1780
1781Reads a CBOR value, decodes it and passes it to the callback. When a parse
1782error occurs, an C<EBADMSG> error will be raised.
1783
1784If a L<CBOR::XS> object was passed to the constructor, then that will be
1785used for the final decode, otherwise it will create a CBOR coder without
1786enabling any options.
1787
1788You have to provide a dependency to L<CBOR::XS> on your own: this module
1789will load the L<CBOR::XS> module, but AnyEvent does not depend on it
1790itself.
1791
1792Since CBOR values are fully self-delimiting, the C<cbor> read and write
1793types are an ideal simple RPC protocol: just exchange CBOR datagrams. See
1794the C<cbor> write type description, above, for an actual example.
1795
1796=cut
1797
1798register_read_type cbor => sub {
1799 my ($self, $cb) = @_;
1800
1801 my $cbor = $self->{cbor} ||= cbor_coder;
1802
1803 my $data;
1804
1805 sub {
1806 my (@value) = eval { $cbor->incr_parse ($_[0]{rbuf}) };
1807
1808 if (@value) {
1809 $cb->($_[0], @value);
1810
1811 1
1812 } elsif ($@) {
1813 # error case
1814 $cbor->incr_reset;
1815
1816 $_[0]->_error (Errno::EBADMSG);
1817
1818 ()
1819 } else {
1200 () 1820 ()
1201 } 1821 }
1202 } 1822 }
1203}; 1823};
1204 1824
1213=cut 1833=cut
1214 1834
1215register_read_type storable => sub { 1835register_read_type storable => sub {
1216 my ($self, $cb) = @_; 1836 my ($self, $cb) = @_;
1217 1837
1218 require Storable; 1838 require Storable unless $Storable::VERSION;
1219 1839
1220 sub { 1840 sub {
1221 # when we can use 5.10 we can use ".", but for 5.8 we use the re-pack method 1841 # when we can use 5.10 we can use ".", but for 5.8 we use the re-pack method
1222 defined (my $len = eval { unpack "w", $_[0]{rbuf} }) 1842 defined (my $len = eval { unpack "w", $_[0]{rbuf} })
1223 or return; 1843 or return;
1226 1846
1227 # bypass unshift if we already have the remaining chunk 1847 # bypass unshift if we already have the remaining chunk
1228 if ($format + $len <= length $_[0]{rbuf}) { 1848 if ($format + $len <= length $_[0]{rbuf}) {
1229 my $data = substr $_[0]{rbuf}, $format, $len; 1849 my $data = substr $_[0]{rbuf}, $format, $len;
1230 substr $_[0]{rbuf}, 0, $format + $len, ""; 1850 substr $_[0]{rbuf}, 0, $format + $len, "";
1851
1231 $cb->($_[0], Storable::thaw ($data)); 1852 eval { $cb->($_[0], Storable::thaw ($data)); 1 }
1853 or return $_[0]->_error (Errno::EBADMSG);
1232 } else { 1854 } else {
1233 # remove prefix 1855 # remove prefix
1234 substr $_[0]{rbuf}, 0, $format, ""; 1856 substr $_[0]{rbuf}, 0, $format, "";
1235 1857
1236 # read remaining chunk 1858 # read remaining chunk
1237 $_[0]->unshift_read (chunk => $len, sub { 1859 $_[0]->unshift_read (chunk => $len, sub {
1238 if (my $ref = eval { Storable::thaw ($_[1]) }) { 1860 eval { $cb->($_[0], Storable::thaw ($_[1])); 1 }
1239 $cb->($_[0], $ref);
1240 } else {
1241 $self->_error (&Errno::EBADMSG); 1861 or $_[0]->_error (Errno::EBADMSG);
1242 }
1243 }); 1862 });
1244 } 1863 }
1245 1864
1246 1 1865 1
1247 } 1866 }
1248}; 1867};
1249 1868
1869=item tls_detect => $cb->($handle, $detect, $major, $minor)
1870
1871Checks the input stream for a valid SSL or TLS handshake TLSPaintext
1872record without consuming anything. Only SSL version 3 or higher
1873is handled, up to the fictituous protocol 4.x (but both SSL3+ and
1874SSL2-compatible framing is supported).
1875
1876If it detects that the input data is likely TLS, it calls the callback
1877with a true value for C<$detect> and the (on-wire) TLS version as second
1878and third argument (C<$major> is C<3>, and C<$minor> is 0..3 for SSL
18793.0, TLS 1.0, 1.1 and 1.2, respectively). If it detects the input to
1880be definitely not TLS, it calls the callback with a false value for
1881C<$detect>.
1882
1883The callback could use this information to decide whether or not to start
1884TLS negotiation.
1885
1886In all cases the data read so far is passed to the following read
1887handlers.
1888
1889Usually you want to use the C<tls_autostart> read type instead.
1890
1891If you want to design a protocol that works in the presence of TLS
1892dtection, make sure that any non-TLS data doesn't start with the octet 22
1893(ASCII SYN, 16 hex) or 128-255 (i.e. highest bit set). The checks this
1894read type does are a bit more strict, but might losen in the future to
1895accomodate protocol changes.
1896
1897This read type does not rely on L<AnyEvent::TLS> (and thus, not on
1898L<Net::SSLeay>).
1899
1900=item tls_autostart => [$tls_ctx, ]$tls
1901
1902Tries to detect a valid SSL or TLS handshake. If one is detected, it tries
1903to start tls by calling C<starttls> with the given arguments.
1904
1905In practise, C<$tls> must be C<accept>, or a Net::SSLeay context that has
1906been configured to accept, as servers do not normally send a handshake on
1907their own and ths cannot be detected in this way.
1908
1909See C<tls_detect> above for more details.
1910
1911Example: give the client a chance to start TLS before accepting a text
1912line.
1913
1914 $hdl->push_read (tls_autostart => "accept");
1915 $hdl->push_read (line => sub {
1916 print "received ", ($_[0]{tls} ? "encrypted" : "cleartext"), " <$_[1]>\n";
1917 });
1918
1919=cut
1920
1921register_read_type tls_detect => sub {
1922 my ($self, $cb) = @_;
1923
1924 sub {
1925 # this regex matches a full or partial tls record
1926 if (
1927 # ssl3+: type(22=handshake) major(=3) minor(any) length_hi
1928 $self->{rbuf} =~ /^(?:\z| \x16 (\z| [\x03\x04] (?:\z| . (?:\z| [\x00-\x40] ))))/xs
1929 # ssl2 comapatible: len_hi len_lo type(1) major minor dummy(forlength)
1930 or $self->{rbuf} =~ /^(?:\z| [\x80-\xff] (?:\z| . (?:\z| \x01 (\z| [\x03\x04] (?:\z| . (?:\z| . ))))))/xs
1931 ) {
1932 return if 3 != length $1; # partial match, can't decide yet
1933
1934 # full match, valid TLS record
1935 my ($major, $minor) = unpack "CC", $1;
1936 $cb->($self, "accept", $major + $minor * 0.1);
1937 } else {
1938 # mismatch == guaranteed not TLS
1939 $cb->($self, undef);
1940 }
1941
1942 1
1943 }
1944};
1945
1946register_read_type tls_autostart => sub {
1947 my ($self, @tls) = @_;
1948
1949 $RH{tls_detect}($self, sub {
1950 return unless $_[1];
1951 $_[0]->starttls (@tls);
1952 })
1953};
1954
1250=back 1955=back
1251 1956
1252=item AnyEvent::Handle::register_read_type type => $coderef->($handle, $cb, @args) 1957=item custom read types - Package::anyevent_read_type $handle, $cb, @args
1253 1958
1254This function (not method) lets you add your own types to C<push_read>. 1959Instead of one of the predefined types, you can also specify the name
1960of a package. AnyEvent will try to load the package and then expects to
1961find a function named C<anyevent_read_type> inside. If it isn't found, it
1962progressively tries to load the parent package until it either finds the
1963function (good) or runs out of packages (bad).
1255 1964
1256Whenever the given C<type> is used, C<push_read> will invoke the code 1965Whenever this type is used, C<push_read> will invoke the function with the
1257reference with the handle object, the callback and the remaining 1966handle object, the original callback and the remaining arguments.
1258arguments.
1259 1967
1260The code reference is supposed to return a callback (usually a closure) 1968The function is supposed to return a callback (usually a closure) that
1261that works as a plain read callback (see C<< ->push_read ($cb) >>). 1969works as a plain read callback (see C<< ->push_read ($cb) >>), so you can
1970mentally treat the function as a "configurable read type to read callback"
1971converter.
1262 1972
1263It should invoke the passed callback when it is done reading (remember to 1973It should invoke the original callback when it is done reading (remember
1264pass C<$handle> as first argument as all other callbacks do that). 1974to pass C<$handle> as first argument as all other callbacks do that,
1975although there is no strict requirement on this).
1265 1976
1266Note that this is a function, and all types registered this way will be
1267global, so try to use unique names.
1268
1269For examples, see the source of this module (F<perldoc -m AnyEvent::Handle>, 1977For examples, see the source of this module (F<perldoc -m
1270search for C<register_read_type>)). 1978AnyEvent::Handle>, search for C<register_read_type>)).
1271 1979
1272=item $handle->stop_read 1980=item $handle->stop_read
1273 1981
1274=item $handle->start_read 1982=item $handle->start_read
1275 1983
1281Note that AnyEvent::Handle will automatically C<start_read> for you when 1989Note that AnyEvent::Handle will automatically C<start_read> for you when
1282you change the C<on_read> callback or push/unshift a read callback, and it 1990you change the C<on_read> callback or push/unshift a read callback, and it
1283will automatically C<stop_read> for you when neither C<on_read> is set nor 1991will automatically C<stop_read> for you when neither C<on_read> is set nor
1284there are any read requests in the queue. 1992there are any read requests in the queue.
1285 1993
1286These methods will have no effect when in TLS mode (as TLS doesn't support 1994In older versions of this module (<= 5.3), these methods had no effect,
1287half-duplex connections). 1995as TLS does not support half-duplex connections. In current versions they
1996work as expected, as this behaviour is required to avoid certain resource
1997attacks, where the program would be forced to read (and buffer) arbitrary
1998amounts of data before being able to send some data. The drawback is that
1999some readings of the the SSL/TLS specifications basically require this
2000attack to be working, as SSL/TLS implementations might stall sending data
2001during a rehandshake.
2002
2003As a guideline, during the initial handshake, you should not stop reading,
2004and as a client, it might cause problems, depending on your application.
1288 2005
1289=cut 2006=cut
1290 2007
1291sub stop_read { 2008sub stop_read {
1292 my ($self) = @_; 2009 my ($self) = @_;
1293 2010
1294 delete $self->{_rw} unless $self->{tls}; 2011 delete $self->{_rw};
1295} 2012}
1296 2013
1297sub start_read { 2014sub start_read {
1298 my ($self) = @_; 2015 my ($self) = @_;
1299 2016
1300 unless ($self->{_rw} || $self->{_eof}) { 2017 unless ($self->{_rw} || $self->{_eof} || !$self->{fh}) {
1301 Scalar::Util::weaken $self; 2018 Scalar::Util::weaken $self;
1302 2019
1303 $self->{_rw} = AnyEvent->io (fh => $self->{fh}, poll => "r", cb => sub { 2020 $self->{_rw} = AE::io $self->{fh}, 0, sub {
1304 my $rbuf = \($self->{tls} ? my $buf : $self->{rbuf}); 2021 my $rbuf = \($self->{tls} ? my $buf : $self->{rbuf});
1305 my $len = sysread $self->{fh}, $$rbuf, $self->{read_size} || 8192, length $$rbuf; 2022 my $len = sysread $self->{fh}, $$rbuf, $self->{read_size}, length $$rbuf;
1306 2023
1307 if ($len > 0) { 2024 if ($len > 0) {
1308 $self->{_activity} = AnyEvent->now; 2025 $self->{_activity} = $self->{_ractivity} = AE::now;
1309 2026
1310 if ($self->{tls}) { 2027 if ($self->{tls}) {
1311 Net::SSLeay::BIO_write ($self->{_rbio}, $$rbuf); 2028 Net::SSLeay::BIO_write ($self->{_rbio}, $$rbuf);
1312 2029
1313 &_dotls ($self); 2030 &_dotls ($self);
1314 } else { 2031 } else {
1315 $self->_drain_rbuf unless $self->{_in_drain}; 2032 $self->_drain_rbuf;
2033 }
2034
2035 if ($len == $self->{read_size}) {
2036 $self->{read_size} *= 2;
2037 $self->{read_size} = $self->{max_read_size} || MAX_READ_SIZE
2038 if $self->{read_size} > ($self->{max_read_size} || MAX_READ_SIZE);
1316 } 2039 }
1317 2040
1318 } elsif (defined $len) { 2041 } elsif (defined $len) {
1319 delete $self->{_rw}; 2042 delete $self->{_rw};
1320 $self->{_eof} = 1; 2043 $self->{_eof} = 1;
1321 $self->_drain_rbuf unless $self->{_in_drain}; 2044 $self->_drain_rbuf;
1322 2045
1323 } elsif ($! != EAGAIN && $! != EINTR && $! != WSAEWOULDBLOCK) { 2046 } elsif ($! != EAGAIN && $! != EINTR && $! != EWOULDBLOCK && $! != WSAEWOULDBLOCK) {
1324 return $self->_error ($!, 1); 2047 return $self->_error ($!, 1);
1325 } 2048 }
1326 }); 2049 };
2050 }
2051}
2052
2053our $ERROR_SYSCALL;
2054our $ERROR_WANT_READ;
2055
2056sub _tls_error {
2057 my ($self, $err) = @_;
2058
2059 return $self->_error ($!, 1)
2060 if $err == Net::SSLeay::ERROR_SYSCALL ();
2061
2062 my $err = Net::SSLeay::ERR_error_string (Net::SSLeay::ERR_get_error ());
2063
2064 # reduce error string to look less scary
2065 $err =~ s/^error:[0-9a-fA-F]{8}:[^:]+:([^:]+):/\L$1: /;
2066
2067 if ($self->{_on_starttls}) {
2068 (delete $self->{_on_starttls})->($self, undef, $err);
2069 &_freetls;
2070 } else {
2071 &_freetls;
2072 $self->_error (Errno::EPROTO, 1, $err);
1327 } 2073 }
1328} 2074}
1329 2075
1330# poll the write BIO and send the data if applicable 2076# poll the write BIO and send the data if applicable
2077# also decode read data if possible
2078# this is basiclaly our TLS state machine
2079# more efficient implementations are possible with openssl,
2080# but not with the buggy and incomplete Net::SSLeay.
1331sub _dotls { 2081sub _dotls {
1332 my ($self) = @_; 2082 my ($self) = @_;
1333 2083
1334 my $tmp; 2084 my $tmp;
1335 2085
1336 if (length $self->{_tls_wbuf}) { 2086 while (length $self->{_tls_wbuf}) {
1337 while (($tmp = Net::SSLeay::write ($self->{tls}, $self->{_tls_wbuf})) > 0) { 2087 if (($tmp = Net::SSLeay::write ($self->{tls}, $self->{_tls_wbuf})) <= 0) {
1338 substr $self->{_tls_wbuf}, 0, $tmp, ""; 2088 $tmp = Net::SSLeay::get_error ($self->{tls}, $tmp);
2089
2090 return $self->_tls_error ($tmp)
2091 if $tmp != $ERROR_WANT_READ
2092 && ($tmp != $ERROR_SYSCALL || $!);
2093
2094 last;
1339 } 2095 }
2096
2097 substr $self->{_tls_wbuf}, 0, $tmp, "";
1340 } 2098 }
1341 2099
1342 while (defined ($tmp = Net::SSLeay::read ($self->{tls}))) { 2100 while (defined ($tmp = Net::SSLeay::read ($self->{tls}))) {
1343 unless (length $tmp) { 2101 unless (length $tmp) {
1344 # let's treat SSL-eof as we treat normal EOF 2102 $self->{_on_starttls}
1345 delete $self->{_rw}; 2103 and (delete $self->{_on_starttls})->($self, undef, "EOF during handshake"); # ???
1346 $self->{_eof} = 1;
1347 &_freetls; 2104 &_freetls;
2105
2106 if ($self->{on_stoptls}) {
2107 $self->{on_stoptls}($self);
2108 return;
2109 } else {
2110 # let's treat SSL-eof as we treat normal EOF
2111 delete $self->{_rw};
2112 $self->{_eof} = 1;
2113 }
1348 } 2114 }
1349 2115
1350 $self->{tls_rbuf} .= $tmp;#d# 2116 $self->{_tls_rbuf} .= $tmp;
1351 $self->_drain_rbuf unless $self->{_in_drain}; 2117 $self->_drain_rbuf;
1352 $self->{tls} or return; # tls session might have gone away in callback 2118 $self->{tls} or return; # tls session might have gone away in callback
1353 } 2119 }
1354 2120
1355 $tmp = Net::SSLeay::get_error ($self->{tls}, -1); 2121 $tmp = Net::SSLeay::get_error ($self->{tls}, -1); # -1 is not neccessarily correct, but Net::SSLeay doesn't tell us
1356
1357 if ($tmp != Net::SSLeay::ERROR_WANT_READ ()) {
1358 if ($tmp == Net::SSLeay::ERROR_SYSCALL ()) {
1359 return $self->_error ($!, 1); 2122 return $self->_tls_error ($tmp)
1360 } elsif ($tmp == Net::SSLeay::ERROR_SSL ()) { 2123 if $tmp != $ERROR_WANT_READ
1361 return $self->_error (&Errno::EIO, 1); 2124 && ($tmp != $ERROR_SYSCALL || $!);
1362 }
1363
1364 # all other errors are fine for our purposes
1365 }
1366 2125
1367 while (length ($tmp = Net::SSLeay::BIO_read ($self->{_wbio}))) { 2126 while (length ($tmp = Net::SSLeay::BIO_read ($self->{_wbio}))) {
1368 $self->{wbuf} .= $tmp; 2127 $self->{wbuf} .= $tmp;
1369 $self->_drain_wbuf; 2128 $self->_drain_wbuf;
2129 $self->{tls} or return; # tls session might have gone away in callback
1370 } 2130 }
2131
2132 $self->{_on_starttls}
2133 and Net::SSLeay::state ($self->{tls}) == Net::SSLeay::ST_OK ()
2134 and (delete $self->{_on_starttls})->($self, 1, "TLS/SSL connection established");
1371} 2135}
1372 2136
1373=item $handle->starttls ($tls[, $tls_ctx]) 2137=item $handle->starttls ($tls[, $tls_ctx])
1374 2138
1375Instead of starting TLS negotiation immediately when the AnyEvent::Handle 2139Instead of starting TLS negotiation immediately when the AnyEvent::Handle
1376object is created, you can also do that at a later time by calling 2140object is created, you can also do that at a later time by calling
1377C<starttls>. 2141C<starttls>. See the C<tls> constructor argument for general info.
2142
2143Starting TLS is currently an asynchronous operation - when you push some
2144write data and then call C<< ->starttls >> then TLS negotiation will start
2145immediately, after which the queued write data is then sent. This might
2146change in future versions, so best make sure you have no outstanding write
2147data when calling this method.
1378 2148
1379The first argument is the same as the C<tls> constructor argument (either 2149The first argument is the same as the C<tls> constructor argument (either
1380C<"connect">, C<"accept"> or an existing Net::SSLeay object). 2150C<"connect">, C<"accept"> or an existing Net::SSLeay object).
1381 2151
1382The second argument is the optional C<Net::SSLeay::CTX> object that is 2152The second argument is the optional C<AnyEvent::TLS> object that is used
1383used when AnyEvent::Handle has to create its own TLS connection object. 2153when AnyEvent::Handle has to create its own TLS connection object, or
2154a hash reference with C<< key => value >> pairs that will be used to
2155construct a new context.
1384 2156
1385The TLS connection object will end up in C<< $handle->{tls} >> after this 2157The TLS connection object will end up in C<< $handle->{tls} >>, the TLS
1386call and can be used or changed to your liking. Note that the handshake 2158context in C<< $handle->{tls_ctx} >> after this call and can be used or
1387might have already started when this function returns. 2159changed to your liking. Note that the handshake might have already started
2160when this function returns.
1388 2161
1389If it an error to start a TLS handshake more than once per 2162Due to bugs in OpenSSL, it might or might not be possible to do multiple
1390AnyEvent::Handle object (this is due to bugs in OpenSSL). 2163handshakes on the same stream. It is best to not attempt to use the
2164stream after stopping TLS.
1391 2165
2166This method may invoke callbacks (and therefore the handle might be
2167destroyed after it returns).
2168
1392=cut 2169=cut
2170
2171our %TLS_CACHE; #TODO not yet documented, should we?
1393 2172
1394sub starttls { 2173sub starttls {
1395 my ($self, $ssl, $ctx) = @_; 2174 my ($self, $tls, $ctx) = @_;
1396 2175
1397 require Net::SSLeay; 2176 Carp::croak "It is an error to call starttls on an AnyEvent::Handle object while TLS is already active, caught"
1398
1399 Carp::croak "it is an error to call starttls more than once on an AnyEvent::Handle object"
1400 if $self->{tls}; 2177 if $self->{tls};
2178
2179 unless (defined $AnyEvent::TLS::VERSION) {
2180 eval {
2181 require Net::SSLeay;
2182 require AnyEvent::TLS;
2183 1
2184 } or return $self->_error (Errno::EPROTO, 1, "TLS support not available on this system");
2185 }
2186
2187 $self->{tls} = $tls;
2188 $self->{tls_ctx} = $ctx if @_ > 2;
2189
2190 return unless $self->{fh};
2191
2192 $ERROR_SYSCALL = Net::SSLeay::ERROR_SYSCALL ();
2193 $ERROR_WANT_READ = Net::SSLeay::ERROR_WANT_READ ();
2194
2195 $tls = delete $self->{tls};
2196 $ctx = $self->{tls_ctx};
2197
2198 local $Carp::CarpLevel = 1; # skip ourselves when creating a new context or session
2199
2200 if ("HASH" eq ref $ctx) {
2201 if ($ctx->{cache}) {
2202 my $key = $ctx+0;
2203 $ctx = $TLS_CACHE{$key} ||= new AnyEvent::TLS %$ctx;
2204 } else {
2205 $ctx = new AnyEvent::TLS %$ctx;
2206 }
2207 }
1401 2208
1402 if ($ssl eq "accept") { 2209 $self->{tls_ctx} = $ctx || TLS_CTX ();
1403 $ssl = Net::SSLeay::new ($ctx || TLS_CTX ()); 2210 $self->{tls} = $tls = $self->{tls_ctx}->_get_session ($tls, $self, $self->{peername});
1404 Net::SSLeay::set_accept_state ($ssl);
1405 } elsif ($ssl eq "connect") {
1406 $ssl = Net::SSLeay::new ($ctx || TLS_CTX ());
1407 Net::SSLeay::set_connect_state ($ssl);
1408 }
1409
1410 $self->{tls} = $ssl;
1411 2211
1412 # basically, this is deep magic (because SSL_read should have the same issues) 2212 # basically, this is deep magic (because SSL_read should have the same issues)
1413 # but the openssl maintainers basically said: "trust us, it just works". 2213 # but the openssl maintainers basically said: "trust us, it just works".
1414 # (unfortunately, we have to hardcode constants because the abysmally misdesigned 2214 # (unfortunately, we have to hardcode constants because the abysmally misdesigned
1415 # and mismaintained ssleay-module doesn't even offer them). 2215 # and mismaintained ssleay-module doesn't even offer them).
1419 # 2219 #
1420 # note that we do not try to keep the length constant between writes as we are required to do. 2220 # note that we do not try to keep the length constant between writes as we are required to do.
1421 # we assume that most (but not all) of this insanity only applies to non-blocking cases, 2221 # we assume that most (but not all) of this insanity only applies to non-blocking cases,
1422 # and we drive openssl fully in blocking mode here. Or maybe we don't - openssl seems to 2222 # and we drive openssl fully in blocking mode here. Or maybe we don't - openssl seems to
1423 # have identity issues in that area. 2223 # have identity issues in that area.
1424 Net::SSLeay::CTX_set_mode ($self->{tls}, 2224# Net::SSLeay::CTX_set_mode ($ssl,
1425 (eval { local $SIG{__DIE__}; Net::SSLeay::MODE_ENABLE_PARTIAL_WRITE () } || 1) 2225# (eval { local $SIG{__DIE__}; Net::SSLeay::MODE_ENABLE_PARTIAL_WRITE () } || 1)
1426 | (eval { local $SIG{__DIE__}; Net::SSLeay::MODE_ACCEPT_MOVING_WRITE_BUFFER () } || 2)); 2226# | (eval { local $SIG{__DIE__}; Net::SSLeay::MODE_ACCEPT_MOVING_WRITE_BUFFER () } || 2));
2227 Net::SSLeay::CTX_set_mode ($tls, 1|2);
1427 2228
1428 $self->{_rbio} = Net::SSLeay::BIO_new (Net::SSLeay::BIO_s_mem ()); 2229 $self->{_rbio} = Net::SSLeay::BIO_new (Net::SSLeay::BIO_s_mem ());
1429 $self->{_wbio} = Net::SSLeay::BIO_new (Net::SSLeay::BIO_s_mem ()); 2230 $self->{_wbio} = Net::SSLeay::BIO_new (Net::SSLeay::BIO_s_mem ());
1430 2231
2232 Net::SSLeay::BIO_write ($self->{_rbio}, $self->{rbuf});
2233 $self->{rbuf} = "";
2234
1431 Net::SSLeay::set_bio ($ssl, $self->{_rbio}, $self->{_wbio}); 2235 Net::SSLeay::set_bio ($tls, $self->{_rbio}, $self->{_wbio});
2236
2237 $self->{_on_starttls} = sub { $_[0]{on_starttls}(@_) }
2238 if $self->{on_starttls};
1432 2239
1433 &_dotls; # need to trigger the initial handshake 2240 &_dotls; # need to trigger the initial handshake
1434 $self->start_read; # make sure we actually do read 2241 $self->start_read; # make sure we actually do read
1435} 2242}
1436 2243
1437=item $handle->stoptls 2244=item $handle->stoptls
1438 2245
1439Shuts down the SSL connection - this makes a proper EOF handshake by 2246Shuts down the SSL connection - this makes a proper EOF handshake by
1440sending a close notify to the other side, but since OpenSSL doesn't 2247sending a close notify to the other side, but since OpenSSL doesn't
1441support non-blocking shut downs, it is not possible to re-use the stream 2248support non-blocking shut downs, it is not guaranteed that you can re-use
1442afterwards. 2249the stream afterwards.
2250
2251This method may invoke callbacks (and therefore the handle might be
2252destroyed after it returns).
1443 2253
1444=cut 2254=cut
1445 2255
1446sub stoptls { 2256sub stoptls {
1447 my ($self) = @_; 2257 my ($self) = @_;
1448 2258
1449 if ($self->{tls}) { 2259 if ($self->{tls} && $self->{fh}) {
1450 Net::SSLeay::shutdown ($self->{tls}); 2260 Net::SSLeay::shutdown ($self->{tls});
1451 2261
1452 &_dotls; 2262 &_dotls;
1453 2263
1454 # we don't give a shit. no, we do, but we can't. no... 2264# # we don't give a shit. no, we do, but we can't. no...#d#
1455 # we, we... have to use openssl :/ 2265# # we, we... have to use openssl :/#d#
1456 &_freetls; 2266# &_freetls;#d#
1457 } 2267 }
1458} 2268}
1459 2269
1460sub _freetls { 2270sub _freetls {
1461 my ($self) = @_; 2271 my ($self) = @_;
1462 2272
1463 return unless $self->{tls}; 2273 return unless $self->{tls};
1464 2274
1465 Net::SSLeay::free (delete $self->{tls}); 2275 $self->{tls_ctx}->_put_session (delete $self->{tls})
2276 if $self->{tls} > 0;
1466 2277
1467 delete @$self{qw(_rbio _wbio _tls_wbuf)}; 2278 delete @$self{qw(_rbio _wbio _tls_wbuf _on_starttls)};
1468} 2279}
2280
2281=item $handle->resettls
2282
2283This rarely-used method simply resets and TLS state on the handle, usually
2284causing data loss.
2285
2286One case where it may be useful is when you want to skip over the data in
2287the stream but you are not interested in interpreting it, so data loss is
2288no concern.
2289
2290=cut
2291
2292*resettls = \&_freetls;
1469 2293
1470sub DESTROY { 2294sub DESTROY {
1471 my $self = shift; 2295 my ($self) = @_;
1472 2296
1473 &_freetls; 2297 &_freetls;
1474 2298
1475 my $linger = exists $self->{linger} ? $self->{linger} : 3600; 2299 my $linger = exists $self->{linger} ? $self->{linger} : 3600;
1476 2300
1477 if ($linger && length $self->{wbuf}) { 2301 if ($linger && length $self->{wbuf} && $self->{fh}) {
1478 my $fh = delete $self->{fh}; 2302 my $fh = delete $self->{fh};
1479 my $wbuf = delete $self->{wbuf}; 2303 my $wbuf = delete $self->{wbuf};
1480 2304
1481 my @linger; 2305 my @linger;
1482 2306
1483 push @linger, AnyEvent->io (fh => $fh, poll => "w", cb => sub { 2307 push @linger, AE::io $fh, 1, sub {
1484 my $len = syswrite $fh, $wbuf, length $wbuf; 2308 my $len = syswrite $fh, $wbuf, length $wbuf;
1485 2309
1486 if ($len > 0) { 2310 if ($len > 0) {
1487 substr $wbuf, 0, $len, ""; 2311 substr $wbuf, 0, $len, "";
1488 } else { 2312 } elsif (defined $len || ($! != EAGAIN && $! != EINTR && $! != EWOULDBLOCK && $! != WSAEWOULDBLOCK)) {
1489 @linger = (); # end 2313 @linger = (); # end
1490 } 2314 }
1491 }); 2315 };
1492 push @linger, AnyEvent->timer (after => $linger, cb => sub { 2316 push @linger, AE::timer $linger, 0, sub {
1493 @linger = (); 2317 @linger = ();
1494 }); 2318 };
1495 } 2319 }
1496} 2320}
1497 2321
1498=item $handle->destroy 2322=item $handle->destroy
1499 2323
1500Shuts down the handle object as much as possible - this call ensures that 2324Shuts down the handle object as much as possible - this call ensures that
1501no further callbacks will be invoked and resources will be freed as much 2325no further callbacks will be invoked and as many resources as possible
1502as possible. You must not call any methods on the object afterwards. 2326will be freed. Any method you will call on the handle object after
2327destroying it in this way will be silently ignored (and it will return the
2328empty list).
1503 2329
1504Normally, you can just "forget" any references to an AnyEvent::Handle 2330Normally, you can just "forget" any references to an AnyEvent::Handle
1505object and it will simply shut down. This works in fatal error and EOF 2331object and it will simply shut down. This works in fatal error and EOF
1506callbacks, as well as code outside. It does I<NOT> work in a read or write 2332callbacks, as well as code outside. It does I<NOT> work in a read or write
1507callback, so when you want to destroy the AnyEvent::Handle object from 2333callback, so when you want to destroy the AnyEvent::Handle object from
1508within such an callback. You I<MUST> call C<< ->destroy >> explicitly in 2334within such an callback. You I<MUST> call C<< ->destroy >> explicitly in
1509that case. 2335that case.
1510 2336
2337Destroying the handle object in this way has the advantage that callbacks
2338will be removed as well, so if those are the only reference holders (as
2339is common), then one doesn't need to do anything special to break any
2340reference cycles.
2341
1511The handle might still linger in the background and write out remaining 2342The handle might still linger in the background and write out remaining
1512data, as specified by the C<linger> option, however. 2343data, as specified by the C<linger> option, however.
1513 2344
1514=cut 2345=cut
1515 2346
1516sub destroy { 2347sub destroy {
1517 my ($self) = @_; 2348 my ($self) = @_;
1518 2349
1519 $self->DESTROY; 2350 $self->DESTROY;
1520 %$self = (); 2351 %$self = ();
2352 bless $self, "AnyEvent::Handle::destroyed";
1521} 2353}
2354
2355sub AnyEvent::Handle::destroyed::AUTOLOAD {
2356 #nop
2357}
2358
2359=item $handle->destroyed
2360
2361Returns false as long as the handle hasn't been destroyed by a call to C<<
2362->destroy >>, true otherwise.
2363
2364Can be useful to decide whether the handle is still valid after some
2365callback possibly destroyed the handle. For example, C<< ->push_write >>,
2366C<< ->starttls >> and other methods can call user callbacks, which in turn
2367can destroy the handle, so work can be avoided by checking sometimes:
2368
2369 $hdl->starttls ("accept");
2370 return if $hdl->destroyed;
2371 $hdl->push_write (...
2372
2373Note that the call to C<push_write> will silently be ignored if the handle
2374has been destroyed, so often you can just ignore the possibility of the
2375handle being destroyed.
2376
2377=cut
2378
2379sub destroyed { 0 }
2380sub AnyEvent::Handle::destroyed::destroyed { 1 }
1522 2381
1523=item AnyEvent::Handle::TLS_CTX 2382=item AnyEvent::Handle::TLS_CTX
1524 2383
1525This function creates and returns the Net::SSLeay::CTX object used by 2384This function creates and returns the AnyEvent::TLS object used by default
1526default for TLS mode. 2385for TLS mode.
1527 2386
1528The context is created like this: 2387The context is created by calling L<AnyEvent::TLS> without any arguments.
1529
1530 Net::SSLeay::load_error_strings;
1531 Net::SSLeay::SSLeay_add_ssl_algorithms;
1532 Net::SSLeay::randomize;
1533
1534 my $CTX = Net::SSLeay::CTX_new;
1535
1536 Net::SSLeay::CTX_set_options $CTX, Net::SSLeay::OP_ALL
1537 2388
1538=cut 2389=cut
1539 2390
1540our $TLS_CTX; 2391our $TLS_CTX;
1541 2392
1542sub TLS_CTX() { 2393sub TLS_CTX() {
1543 $TLS_CTX || do { 2394 $TLS_CTX ||= do {
1544 require Net::SSLeay; 2395 require AnyEvent::TLS;
1545 2396
1546 Net::SSLeay::load_error_strings (); 2397 new AnyEvent::TLS
1547 Net::SSLeay::SSLeay_add_ssl_algorithms ();
1548 Net::SSLeay::randomize ();
1549
1550 $TLS_CTX = Net::SSLeay::CTX_new ();
1551
1552 Net::SSLeay::CTX_set_options ($TLS_CTX, Net::SSLeay::OP_ALL ());
1553
1554 $TLS_CTX
1555 } 2398 }
1556} 2399}
1557 2400
1558=back 2401=back
1559 2402
1570 2413
1571It is only safe to "forget" the reference inside EOF or error callbacks, 2414It is only safe to "forget" the reference inside EOF or error callbacks,
1572from within all other callbacks, you need to explicitly call the C<< 2415from within all other callbacks, you need to explicitly call the C<<
1573->destroy >> method. 2416->destroy >> method.
1574 2417
2418=item Why is my C<on_eof> callback never called?
2419
2420Probably because your C<on_error> callback is being called instead: When
2421you have outstanding requests in your read queue, then an EOF is
2422considered an error as you clearly expected some data.
2423
2424To avoid this, make sure you have an empty read queue whenever your handle
2425is supposed to be "idle" (i.e. connection closes are O.K.). You can set
2426an C<on_read> handler that simply pushes the first read requests in the
2427queue.
2428
2429See also the next question, which explains this in a bit more detail.
2430
2431=item How can I serve requests in a loop?
2432
2433Most protocols consist of some setup phase (authentication for example)
2434followed by a request handling phase, where the server waits for requests
2435and handles them, in a loop.
2436
2437There are two important variants: The first (traditional, better) variant
2438handles requests until the server gets some QUIT command, causing it to
2439close the connection first (highly desirable for a busy TCP server). A
2440client dropping the connection is an error, which means this variant can
2441detect an unexpected detection close.
2442
2443To handle this case, always make sure you have a non-empty read queue, by
2444pushing the "read request start" handler on it:
2445
2446 # we assume a request starts with a single line
2447 my @start_request; @start_request = (line => sub {
2448 my ($hdl, $line) = @_;
2449
2450 ... handle request
2451
2452 # push next request read, possibly from a nested callback
2453 $hdl->push_read (@start_request);
2454 });
2455
2456 # auth done, now go into request handling loop
2457 # now push the first @start_request
2458 $hdl->push_read (@start_request);
2459
2460By always having an outstanding C<push_read>, the handle always expects
2461some data and raises the C<EPIPE> error when the connction is dropped
2462unexpectedly.
2463
2464The second variant is a protocol where the client can drop the connection
2465at any time. For TCP, this means that the server machine may run out of
2466sockets easier, and in general, it means you cannot distinguish a protocl
2467failure/client crash from a normal connection close. Nevertheless, these
2468kinds of protocols are common (and sometimes even the best solution to the
2469problem).
2470
2471Having an outstanding read request at all times is possible if you ignore
2472C<EPIPE> errors, but this doesn't help with when the client drops the
2473connection during a request, which would still be an error.
2474
2475A better solution is to push the initial request read in an C<on_read>
2476callback. This avoids an error, as when the server doesn't expect data
2477(i.e. is idly waiting for the next request, an EOF will not raise an
2478error, but simply result in an C<on_eof> callback. It is also a bit slower
2479and simpler:
2480
2481 # auth done, now go into request handling loop
2482 $hdl->on_read (sub {
2483 my ($hdl) = @_;
2484
2485 # called each time we receive data but the read queue is empty
2486 # simply start read the request
2487
2488 $hdl->push_read (line => sub {
2489 my ($hdl, $line) = @_;
2490
2491 ... handle request
2492
2493 # do nothing special when the request has been handled, just
2494 # let the request queue go empty.
2495 });
2496 });
2497
1575=item I get different callback invocations in TLS mode/Why can't I pause 2498=item I get different callback invocations in TLS mode/Why can't I pause
1576reading? 2499reading?
1577 2500
1578Unlike, say, TCP, TLS connections do not consist of two independent 2501Unlike, say, TCP, TLS connections do not consist of two independent
1579communication channels, one for each direction. Or put differently. The 2502communication channels, one for each direction. Or put differently, the
1580read and write directions are not independent of each other: you cannot 2503read and write directions are not independent of each other: you cannot
1581write data unless you are also prepared to read, and vice versa. 2504write data unless you are also prepared to read, and vice versa.
1582 2505
1583This can mean than, in TLS mode, you might get C<on_error> or C<on_eof> 2506This means that, in TLS mode, you might get C<on_error> or C<on_eof>
1584callback invocations when you are not expecting any read data - the reason 2507callback invocations when you are not expecting any read data - the reason
1585is that AnyEvent::Handle always reads in TLS mode. 2508is that AnyEvent::Handle always reads in TLS mode.
1586 2509
1587During the connection, you have to make sure that you always have a 2510During the connection, you have to make sure that you always have a
1588non-empty read-queue, or an C<on_read> watcher. At the end of the 2511non-empty read-queue, or an C<on_read> watcher. At the end of the
1598 2521
1599 $handle->on_read (sub { }); 2522 $handle->on_read (sub { });
1600 $handle->on_eof (undef); 2523 $handle->on_eof (undef);
1601 $handle->on_error (sub { 2524 $handle->on_error (sub {
1602 my $data = delete $_[0]{rbuf}; 2525 my $data = delete $_[0]{rbuf};
1603 undef $handle;
1604 }); 2526 });
2527
2528Note that this example removes the C<rbuf> member from the handle object,
2529which is not normally allowed by the API. It is expressly permitted in
2530this case only, as the handle object needs to be destroyed afterwards.
1605 2531
1606The reason to use C<on_error> is that TCP connections, due to latencies 2532The reason to use C<on_error> is that TCP connections, due to latencies
1607and packets loss, might get closed quite violently with an error, when in 2533and packets loss, might get closed quite violently with an error, when in
1608fact, all data has been received. 2534fact all data has been received.
1609 2535
1610It is usually better to use acknowledgements when transferring data, 2536It is usually better to use acknowledgements when transferring data,
1611to make sure the other side hasn't just died and you got the data 2537to make sure the other side hasn't just died and you got the data
1612intact. This is also one reason why so many internet protocols have an 2538intact. This is also one reason why so many internet protocols have an
1613explicit QUIT command. 2539explicit QUIT command.
1620C<low_water_mark> this will be called precisely when all data has been 2546C<low_water_mark> this will be called precisely when all data has been
1621written to the socket: 2547written to the socket:
1622 2548
1623 $handle->push_write (...); 2549 $handle->push_write (...);
1624 $handle->on_drain (sub { 2550 $handle->on_drain (sub {
1625 warn "all data submitted to the kernel\n"; 2551 AE::log debug => "All data submitted to the kernel.";
1626 undef $handle; 2552 undef $handle;
1627 }); 2553 });
1628 2554
2555If you just want to queue some data and then signal EOF to the other side,
2556consider using C<< ->push_shutdown >> instead.
2557
2558=item I want to contact a TLS/SSL server, I don't care about security.
2559
2560If your TLS server is a pure TLS server (e.g. HTTPS) that only speaks TLS,
2561connect to it and then create the AnyEvent::Handle with the C<tls>
2562parameter:
2563
2564 tcp_connect $host, $port, sub {
2565 my ($fh) = @_;
2566
2567 my $handle = new AnyEvent::Handle
2568 fh => $fh,
2569 tls => "connect",
2570 on_error => sub { ... };
2571
2572 $handle->push_write (...);
2573 };
2574
2575=item I want to contact a TLS/SSL server, I do care about security.
2576
2577Then you should additionally enable certificate verification, including
2578peername verification, if the protocol you use supports it (see
2579L<AnyEvent::TLS>, C<verify_peername>).
2580
2581E.g. for HTTPS:
2582
2583 tcp_connect $host, $port, sub {
2584 my ($fh) = @_;
2585
2586 my $handle = new AnyEvent::Handle
2587 fh => $fh,
2588 peername => $host,
2589 tls => "connect",
2590 tls_ctx => { verify => 1, verify_peername => "https" },
2591 ...
2592
2593Note that you must specify the hostname you connected to (or whatever
2594"peername" the protocol needs) as the C<peername> argument, otherwise no
2595peername verification will be done.
2596
2597The above will use the system-dependent default set of trusted CA
2598certificates. If you want to check against a specific CA, add the
2599C<ca_file> (or C<ca_cert>) arguments to C<tls_ctx>:
2600
2601 tls_ctx => {
2602 verify => 1,
2603 verify_peername => "https",
2604 ca_file => "my-ca-cert.pem",
2605 },
2606
2607=item I want to create a TLS/SSL server, how do I do that?
2608
2609Well, you first need to get a server certificate and key. You have
2610three options: a) ask a CA (buy one, use cacert.org etc.) b) create a
2611self-signed certificate (cheap. check the search engine of your choice,
2612there are many tutorials on the net) or c) make your own CA (tinyca2 is a
2613nice program for that purpose).
2614
2615Then create a file with your private key (in PEM format, see
2616L<AnyEvent::TLS>), followed by the certificate (also in PEM format). The
2617file should then look like this:
2618
2619 -----BEGIN RSA PRIVATE KEY-----
2620 ...header data
2621 ... lots of base64'y-stuff
2622 -----END RSA PRIVATE KEY-----
2623
2624 -----BEGIN CERTIFICATE-----
2625 ... lots of base64'y-stuff
2626 -----END CERTIFICATE-----
2627
2628The important bits are the "PRIVATE KEY" and "CERTIFICATE" parts. Then
2629specify this file as C<cert_file>:
2630
2631 tcp_server undef, $port, sub {
2632 my ($fh) = @_;
2633
2634 my $handle = new AnyEvent::Handle
2635 fh => $fh,
2636 tls => "accept",
2637 tls_ctx => { cert_file => "my-server-keycert.pem" },
2638 ...
2639
2640When you have intermediate CA certificates that your clients might not
2641know about, just append them to the C<cert_file>.
2642
1629=back 2643=back
1630
1631 2644
1632=head1 SUBCLASSING AnyEvent::Handle 2645=head1 SUBCLASSING AnyEvent::Handle
1633 2646
1634In many cases, you might want to subclass AnyEvent::Handle. 2647In many cases, you might want to subclass AnyEvent::Handle.
1635 2648
1652 2665
1653=item * all members not documented here and not prefixed with an underscore 2666=item * all members not documented here and not prefixed with an underscore
1654are free to use in subclasses. 2667are free to use in subclasses.
1655 2668
1656Of course, new versions of AnyEvent::Handle may introduce more "public" 2669Of course, new versions of AnyEvent::Handle may introduce more "public"
1657member variables, but thats just life, at least it is documented. 2670member variables, but that's just life. At least it is documented.
1658 2671
1659=back 2672=back
1660 2673
1661=head1 AUTHOR 2674=head1 AUTHOR
1662 2675
1663Robin Redeker C<< <elmex at ta-sa.org> >>, Marc Lehmann <schmorp@schmorp.de>. 2676Robin Redeker C<< <elmex at ta-sa.org> >>, Marc Lehmann <schmorp@schmorp.de>.
1664 2677
1665=cut 2678=cut
1666 2679
16671; # End of AnyEvent::Handle 26801
2681

Diff Legend

Removed lines
+ Added lines
< Changed lines
> Changed lines