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 1.37 by root, Mon May 26 20:02:22 2008 UTC vs.
Revision 1.95 by root, Thu Oct 2 06:42:39 2008 UTC

1package AnyEvent::Handle; 1package AnyEvent::Handle;
2 2
3no warnings; 3no warnings;
4use strict; 4use strict qw(subs vars);
5 5
6use AnyEvent (); 6use AnyEvent ();
7use AnyEvent::Util qw(WSAWOULDBLOCK); 7use AnyEvent::Util qw(WSAEWOULDBLOCK);
8use Scalar::Util (); 8use Scalar::Util ();
9use Carp (); 9use Carp ();
10use Fcntl (); 10use Fcntl ();
11use Errno qw/EAGAIN EINTR/; 11use Errno qw(EAGAIN EINTR);
12 12
13=head1 NAME 13=head1 NAME
14 14
15AnyEvent::Handle - non-blocking I/O on file handles via AnyEvent 15AnyEvent::Handle - non-blocking I/O on file handles via AnyEvent
16 16
17=cut 17=cut
18 18
19our $VERSION = '0.04'; 19our $VERSION = 4.3;
20 20
21=head1 SYNOPSIS 21=head1 SYNOPSIS
22 22
23 use AnyEvent; 23 use AnyEvent;
24 use AnyEvent::Handle; 24 use AnyEvent::Handle;
49 49
50This module is a helper module to make it easier to do event-based I/O on 50This module is a helper module to make it easier to do event-based I/O on
51filehandles. For utility functions for doing non-blocking connects and accepts 51filehandles. For utility functions for doing non-blocking connects and accepts
52on sockets see L<AnyEvent::Util>. 52on sockets see L<AnyEvent::Util>.
53 53
54The L<AnyEvent::Intro> tutorial contains some well-documented
55AnyEvent::Handle examples.
56
54In the following, when the documentation refers to of "bytes" then this 57In the following, when the documentation refers to of "bytes" then this
55means characters. As sysread and syswrite are used for all I/O, their 58means characters. As sysread and syswrite are used for all I/O, their
56treatment of characters applies to this module as well. 59treatment of characters applies to this module as well.
57 60
58All callbacks will be invoked with the handle object as their first 61All callbacks will be invoked with the handle object as their first
59argument. 62argument.
60 63
64=head2 SIGPIPE is not handled by this module
65
66SIGPIPE is not handled by this module, so one of the practical
67requirements of using it is to ignore SIGPIPE (C<$SIG{PIPE} =
68'IGNORE'>). At least, this is highly recommend in a networked program: If
69you use AnyEvent::Handle in a filter program (like sort), exiting on
70SIGPIPE is probably the right thing to do.
71
61=head1 METHODS 72=head1 METHODS
62 73
63=over 4 74=over 4
64 75
65=item B<new (%args)> 76=item B<new (%args)>
70 81
71=item fh => $filehandle [MANDATORY] 82=item fh => $filehandle [MANDATORY]
72 83
73The filehandle this L<AnyEvent::Handle> object will operate on. 84The filehandle this L<AnyEvent::Handle> object will operate on.
74 85
75NOTE: The filehandle will be set to non-blocking (using 86NOTE: The filehandle will be set to non-blocking mode (using
76AnyEvent::Util::fh_nonblocking). 87C<AnyEvent::Util::fh_nonblocking>) by the constructor and needs to stay in
88that mode.
77 89
78=item on_eof => $cb->($self) 90=item on_eof => $cb->($handle)
79 91
80Set the callback to be called on EOF. 92Set the callback to be called when an end-of-file condition is detected,
93i.e. in the case of a socket, when the other side has closed the
94connection cleanly.
81 95
96For sockets, this just means that the other side has stopped sending data,
97you can still try to write data, and, in fact, one can return from the eof
98callback and continue writing data, as only the read part has been shut
99down.
100
82While not mandatory, it is highly recommended to set an eof callback, 101While not mandatory, it is I<highly> recommended to set an eof callback,
83otherwise you might end up with a closed socket while you are still 102otherwise you might end up with a closed socket while you are still
84waiting for data. 103waiting for data.
85 104
105If an EOF condition has been detected but no C<on_eof> callback has been
106set, then a fatal error will be raised with C<$!> set to <0>.
107
86=item on_error => $cb->($self) 108=item on_error => $cb->($handle, $fatal)
87 109
88This is the fatal error callback, that is called when, well, a fatal error 110This is the error callback, which is called when, well, some error
89occurs, such as not being able to resolve the hostname, failure to connect 111occured, such as not being able to resolve the hostname, failure to
90or a read error. 112connect or a read error.
91 113
92The object will not be in a usable state when this callback has been 114Some errors are fatal (which is indicated by C<$fatal> being true). On
93called. 115fatal errors the handle object will be shut down and will not be usable
116(but you are free to look at the current C<< ->rbuf >>). Examples of fatal
117errors are an EOF condition with active (but unsatisifable) read watchers
118(C<EPIPE>) or I/O errors.
119
120Non-fatal errors can be retried by simply returning, but it is recommended
121to simply ignore this parameter and instead abondon the handle object
122when this callback is invoked. Examples of non-fatal errors are timeouts
123C<ETIMEDOUT>) or badly-formatted data (C<EBADMSG>).
94 124
95On callback entrance, the value of C<$!> contains the operating system 125On callback entrance, the value of C<$!> contains the operating system
96error (or C<ENOSPC>, C<EPIPE> or C<EBADMSG>). 126error (or C<ENOSPC>, C<EPIPE>, C<ETIMEDOUT> or C<EBADMSG>).
97
98The callbakc should throw an exception. If it returns, then
99AnyEvent::Handle will C<croak> for you.
100 127
101While not mandatory, it is I<highly> recommended to set this callback, as 128While not mandatory, it is I<highly> recommended to set this callback, as
102you will not be notified of errors otherwise. The default simply calls 129you will not be notified of errors otherwise. The default simply calls
103die. 130C<croak>.
104 131
105=item on_read => $cb->($self) 132=item on_read => $cb->($handle)
106 133
107This sets the default read callback, which is called when data arrives 134This sets the default read callback, which is called when data arrives
108and no read request is in the queue. 135and no read request is in the queue (unlike read queue callbacks, this
136callback will only be called when at least one octet of data is in the
137read buffer).
109 138
110To access (and remove data from) the read buffer, use the C<< ->rbuf >> 139To access (and remove data from) the read buffer, use the C<< ->rbuf >>
111method or access the C<$self->{rbuf}> member directly. 140method or access the C<$handle->{rbuf}> member directly.
112 141
113When an EOF condition is detected then AnyEvent::Handle will first try to 142When an EOF condition is detected then AnyEvent::Handle will first try to
114feed all the remaining data to the queued callbacks and C<on_read> before 143feed all the remaining data to the queued callbacks and C<on_read> before
115calling the C<on_eof> callback. If no progress can be made, then a fatal 144calling the C<on_eof> callback. If no progress can be made, then a fatal
116error will be raised (with C<$!> set to C<EPIPE>). 145error will be raised (with C<$!> set to C<EPIPE>).
117 146
118=item on_drain => $cb->() 147=item on_drain => $cb->($handle)
119 148
120This sets the callback that is called when the write buffer becomes empty 149This sets the callback that is called when the write buffer becomes empty
121(or when the callback is set and the buffer is empty already). 150(or when the callback is set and the buffer is empty already).
122 151
123To append to the write buffer, use the C<< ->push_write >> method. 152To append to the write buffer, use the C<< ->push_write >> method.
124 153
154This callback is useful when you don't want to put all of your write data
155into the queue at once, for example, when you want to write the contents
156of some file to the socket you might not want to read the whole file into
157memory and push it into the queue, but instead only read more data from
158the file when the write queue becomes empty.
159
160=item timeout => $fractional_seconds
161
162If non-zero, then this enables an "inactivity" timeout: whenever this many
163seconds pass without a successful read or write on the underlying file
164handle, the C<on_timeout> callback will be invoked (and if that one is
165missing, a non-fatal C<ETIMEDOUT> error will be raised).
166
167Note that timeout processing is also active when you currently do not have
168any outstanding read or write requests: If you plan to keep the connection
169idle then you should disable the timout temporarily or ignore the timeout
170in the C<on_timeout> callback, in which case AnyEvent::Handle will simply
171restart the timeout.
172
173Zero (the default) disables this timeout.
174
175=item on_timeout => $cb->($handle)
176
177Called whenever the inactivity timeout passes. If you return from this
178callback, then the timeout will be reset as if some activity had happened,
179so this condition is not fatal in any way.
180
125=item rbuf_max => <bytes> 181=item rbuf_max => <bytes>
126 182
127If defined, then a fatal error will be raised (with C<$!> set to C<ENOSPC>) 183If defined, then a fatal error will be raised (with C<$!> set to C<ENOSPC>)
128when the read buffer ever (strictly) exceeds this size. This is useful to 184when the read buffer ever (strictly) exceeds this size. This is useful to
129avoid denial-of-service attacks. 185avoid some forms of denial-of-service attacks.
130 186
131For example, a server accepting connections from untrusted sources should 187For example, a server accepting connections from untrusted sources should
132be configured to accept only so-and-so much data that it cannot act on 188be configured to accept only so-and-so much data that it cannot act on
133(for example, when expecting a line, an attacker could send an unlimited 189(for example, when expecting a line, an attacker could send an unlimited
134amount of data without a callback ever being called as long as the line 190amount of data without a callback ever being called as long as the line
135isn't finished). 191isn't finished).
136 192
193=item autocork => <boolean>
194
195When disabled (the default), then C<push_write> will try to immediately
196write the data to the handle, if possible. This avoids having to register
197a write watcher and wait for the next event loop iteration, but can
198be inefficient if you write multiple small chunks (on the wire, this
199disadvantage is usually avoided by your kernel's nagle algorithm, see
200C<no_delay>, but this option can save costly syscalls).
201
202When enabled, then writes will always be queued till the next event loop
203iteration. This is efficient when you do many small writes per iteration,
204but less efficient when you do a single write only per iteration (or when
205the write buffer often is full). It also increases write latency.
206
207=item no_delay => <boolean>
208
209When doing small writes on sockets, your operating system kernel might
210wait a bit for more data before actually sending it out. This is called
211the Nagle algorithm, and usually it is beneficial.
212
213In some situations you want as low a delay as possible, which can be
214accomplishd by setting this option to a true value.
215
216The default is your opertaing system's default behaviour (most likely
217enabled), this option explicitly enables or disables it, if possible.
218
137=item read_size => <bytes> 219=item read_size => <bytes>
138 220
139The default read block size (the amount of bytes this module will try to read 221The default read block size (the amount of bytes this module will
140on each [loop iteration). Default: C<4096>. 222try to read during each loop iteration, which affects memory
223requirements). Default: C<8192>.
141 224
142=item low_water_mark => <bytes> 225=item low_water_mark => <bytes>
143 226
144Sets the amount of bytes (default: C<0>) that make up an "empty" write 227Sets the amount of bytes (default: C<0>) that make up an "empty" write
145buffer: If the write reaches this size or gets even samller it is 228buffer: If the write reaches this size or gets even samller it is
146considered empty. 229considered empty.
147 230
231Sometimes it can be beneficial (for performance reasons) to add data to
232the write buffer before it is fully drained, but this is a rare case, as
233the operating system kernel usually buffers data as well, so the default
234is good in almost all cases.
235
236=item linger => <seconds>
237
238If non-zero (default: C<3600>), then the destructor of the
239AnyEvent::Handle object will check whether there is still outstanding
240write data and will install a watcher that will write this data to the
241socket. No errors will be reported (this mostly matches how the operating
242system treats outstanding data at socket close time).
243
244This will not work for partial TLS data that could not be encoded
245yet. This data will be lost. Calling the C<stoptls> method in time might
246help.
247
148=item tls => "accept" | "connect" | Net::SSLeay::SSL object 248=item tls => "accept" | "connect" | Net::SSLeay::SSL object
149 249
150When this parameter is given, it enables TLS (SSL) mode, that means it 250When this parameter is given, it enables TLS (SSL) mode, that means
151will start making tls handshake and will transparently encrypt/decrypt 251AnyEvent will start a TLS handshake as soon as the conenction has been
152data. 252established and will transparently encrypt/decrypt data afterwards.
153 253
154TLS mode requires Net::SSLeay to be installed (it will be loaded 254TLS mode requires Net::SSLeay to be installed (it will be loaded
155automatically when you try to create a TLS handle). 255automatically when you try to create a TLS handle): this module doesn't
256have a dependency on that module, so if your module requires it, you have
257to add the dependency yourself.
156 258
157For the TLS server side, use C<accept>, and for the TLS client side of a 259Unlike TCP, TLS has a server and client side: for the TLS server side, use
158connection, use C<connect> mode. 260C<accept>, and for the TLS client side of a connection, use C<connect>
261mode.
159 262
160You can also provide your own TLS connection object, but you have 263You can also provide your own TLS connection object, but you have
161to make sure that you call either C<Net::SSLeay::set_connect_state> 264to make sure that you call either C<Net::SSLeay::set_connect_state>
162or C<Net::SSLeay::set_accept_state> on it before you pass it to 265or C<Net::SSLeay::set_accept_state> on it before you pass it to
163AnyEvent::Handle. 266AnyEvent::Handle.
164 267
165See the C<starttls> method if you need to start TLs negotiation later. 268See the C<< ->starttls >> method for when need to start TLS negotiation later.
166 269
167=item tls_ctx => $ssl_ctx 270=item tls_ctx => $ssl_ctx
168 271
169Use the given Net::SSLeay::CTX object to create the new TLS connection 272Use the given C<Net::SSLeay::CTX> object to create the new TLS connection
170(unless a connection object was specified directly). If this parameter is 273(unless a connection object was specified directly). If this parameter is
171missing, then AnyEvent::Handle will use C<AnyEvent::Handle::TLS_CTX>. 274missing, then AnyEvent::Handle will use C<AnyEvent::Handle::TLS_CTX>.
172 275
276=item json => JSON or JSON::XS object
277
278This is the json coder object used by the C<json> read and write types.
279
280If you don't supply it, then AnyEvent::Handle will create and use a
281suitable one (on demand), which will write and expect UTF-8 encoded JSON
282texts.
283
284Note that you are responsible to depend on the JSON module if you want to
285use this functionality, as AnyEvent does not have a dependency itself.
286
173=back 287=back
174 288
175=cut 289=cut
176 290
177sub new { 291sub new {
181 295
182 $self->{fh} or Carp::croak "mandatory argument fh is missing"; 296 $self->{fh} or Carp::croak "mandatory argument fh is missing";
183 297
184 AnyEvent::Util::fh_nonblocking $self->{fh}, 1; 298 AnyEvent::Util::fh_nonblocking $self->{fh}, 1;
185 299
186 if ($self->{tls}) {
187 require Net::SSLeay;
188 $self->starttls (delete $self->{tls}, delete $self->{tls_ctx}); 300 $self->starttls (delete $self->{tls}, delete $self->{tls_ctx})
189 } 301 if $self->{tls};
190 302
191 $self->on_eof (delete $self->{on_eof} ) if $self->{on_eof}; 303 $self->{_activity} = AnyEvent->now;
192 $self->on_error (delete $self->{on_error}) if $self->{on_error}; 304 $self->_timeout;
305
193 $self->on_drain (delete $self->{on_drain}) if $self->{on_drain}; 306 $self->on_drain (delete $self->{on_drain}) if exists $self->{on_drain};
194 $self->on_read (delete $self->{on_read} ) if $self->{on_read}; 307 $self->no_delay (delete $self->{no_delay}) if exists $self->{no_delay};
195 308
196 $self->start_read; 309 $self->start_read
310 if $self->{on_read};
197 311
198 $self 312 $self
199} 313}
200 314
201sub _shutdown { 315sub _shutdown {
202 my ($self) = @_; 316 my ($self) = @_;
203 317
318 delete $self->{_tw};
204 delete $self->{rw}; 319 delete $self->{_rw};
205 delete $self->{ww}; 320 delete $self->{_ww};
206 delete $self->{fh}; 321 delete $self->{fh};
207}
208 322
323 &_freetls;
324
325 delete $self->{on_read};
326 delete $self->{_queue};
327}
328
209sub error { 329sub _error {
210 my ($self) = @_; 330 my ($self, $errno, $fatal) = @_;
211 331
212 {
213 local $!;
214 $self->_shutdown; 332 $self->_shutdown
215 } 333 if $fatal;
216 334
217 $self->{on_error}($self) 335 $! = $errno;
336
218 if $self->{on_error}; 337 if ($self->{on_error}) {
219 338 $self->{on_error}($self, $fatal);
339 } else {
220 Carp::croak "AnyEvent::Handle uncaught fatal error: $!"; 340 Carp::croak "AnyEvent::Handle uncaught error: $!";
341 }
221} 342}
222 343
223=item $fh = $handle->fh 344=item $fh = $handle->fh
224 345
225This method returns the file handle of the L<AnyEvent::Handle> object. 346This method returns the file handle used to create the L<AnyEvent::Handle> object.
226 347
227=cut 348=cut
228 349
229sub fh { $_[0]->{fh} } 350sub fh { $_[0]{fh} }
230 351
231=item $handle->on_error ($cb) 352=item $handle->on_error ($cb)
232 353
233Replace the current C<on_error> callback (see the C<on_error> constructor argument). 354Replace the current C<on_error> callback (see the C<on_error> constructor argument).
234 355
246 367
247sub on_eof { 368sub on_eof {
248 $_[0]{on_eof} = $_[1]; 369 $_[0]{on_eof} = $_[1];
249} 370}
250 371
372=item $handle->on_timeout ($cb)
373
374Replace the current C<on_timeout> callback, or disables the callback (but
375not the timeout) if C<$cb> = C<undef>. See the C<timeout> constructor
376argument and method.
377
378=cut
379
380sub on_timeout {
381 $_[0]{on_timeout} = $_[1];
382}
383
384=item $handle->autocork ($boolean)
385
386Enables or disables the current autocork behaviour (see C<autocork>
387constructor argument).
388
389=cut
390
391=item $handle->no_delay ($boolean)
392
393Enables or disables the C<no_delay> setting (see constructor argument of
394the same name for details).
395
396=cut
397
398sub no_delay {
399 $_[0]{no_delay} = $_[1];
400
401 eval {
402 local $SIG{__DIE__};
403 setsockopt $_[0]{fh}, &Socket::IPPROTO_TCP, &Socket::TCP_NODELAY, int $_[1];
404 };
405}
406
407#############################################################################
408
409=item $handle->timeout ($seconds)
410
411Configures (or disables) the inactivity timeout.
412
413=cut
414
415sub timeout {
416 my ($self, $timeout) = @_;
417
418 $self->{timeout} = $timeout;
419 $self->_timeout;
420}
421
422# reset the timeout watcher, as neccessary
423# also check for time-outs
424sub _timeout {
425 my ($self) = @_;
426
427 if ($self->{timeout}) {
428 my $NOW = AnyEvent->now;
429
430 # when would the timeout trigger?
431 my $after = $self->{_activity} + $self->{timeout} - $NOW;
432
433 # now or in the past already?
434 if ($after <= 0) {
435 $self->{_activity} = $NOW;
436
437 if ($self->{on_timeout}) {
438 $self->{on_timeout}($self);
439 } else {
440 $self->_error (&Errno::ETIMEDOUT);
441 }
442
443 # callback could have changed timeout value, optimise
444 return unless $self->{timeout};
445
446 # calculate new after
447 $after = $self->{timeout};
448 }
449
450 Scalar::Util::weaken $self;
451 return unless $self; # ->error could have destroyed $self
452
453 $self->{_tw} ||= AnyEvent->timer (after => $after, cb => sub {
454 delete $self->{_tw};
455 $self->_timeout;
456 });
457 } else {
458 delete $self->{_tw};
459 }
460}
461
251############################################################################# 462#############################################################################
252 463
253=back 464=back
254 465
255=head2 WRITE QUEUE 466=head2 WRITE QUEUE
276 my ($self, $cb) = @_; 487 my ($self, $cb) = @_;
277 488
278 $self->{on_drain} = $cb; 489 $self->{on_drain} = $cb;
279 490
280 $cb->($self) 491 $cb->($self)
281 if $cb && $self->{low_water_mark} >= length $self->{wbuf}; 492 if $cb && $self->{low_water_mark} >= (length $self->{wbuf}) + (length $self->{_tls_wbuf});
282} 493}
283 494
284=item $handle->push_write ($data) 495=item $handle->push_write ($data)
285 496
286Queues the given scalar to be written. You can push as much data as you 497Queues the given scalar to be written. You can push as much data as you
290=cut 501=cut
291 502
292sub _drain_wbuf { 503sub _drain_wbuf {
293 my ($self) = @_; 504 my ($self) = @_;
294 505
295 if (!$self->{ww} && length $self->{wbuf}) { 506 if (!$self->{_ww} && length $self->{wbuf}) {
296 507
297 Scalar::Util::weaken $self; 508 Scalar::Util::weaken $self;
298 509
299 my $cb = sub { 510 my $cb = sub {
300 my $len = syswrite $self->{fh}, $self->{wbuf}; 511 my $len = syswrite $self->{fh}, $self->{wbuf};
301 512
302 if ($len >= 0) { 513 if ($len >= 0) {
303 substr $self->{wbuf}, 0, $len, ""; 514 substr $self->{wbuf}, 0, $len, "";
304 515
516 $self->{_activity} = AnyEvent->now;
517
305 $self->{on_drain}($self) 518 $self->{on_drain}($self)
306 if $self->{low_water_mark} >= length $self->{wbuf} 519 if $self->{low_water_mark} >= (length $self->{wbuf}) + (length $self->{_tls_wbuf})
307 && $self->{on_drain}; 520 && $self->{on_drain};
308 521
309 delete $self->{ww} unless length $self->{wbuf}; 522 delete $self->{_ww} unless length $self->{wbuf};
310 } elsif ($! != EAGAIN && $! != EINTR && $! != WSAWOULDBLOCK) { 523 } elsif ($! != EAGAIN && $! != EINTR && $! != WSAEWOULDBLOCK) {
311 $self->error; 524 $self->_error ($!, 1);
312 } 525 }
313 }; 526 };
314 527
315 # try to write data immediately 528 # try to write data immediately
316 $cb->(); 529 $cb->() unless $self->{autocork};
317 530
318 # if still data left in wbuf, we need to poll 531 # if still data left in wbuf, we need to poll
319 $self->{ww} = AnyEvent->io (fh => $self->{fh}, poll => "w", cb => $cb) 532 $self->{_ww} = AnyEvent->io (fh => $self->{fh}, poll => "w", cb => $cb)
320 if length $self->{wbuf}; 533 if length $self->{wbuf};
321 }; 534 };
322} 535}
323 536
324our %WH; 537our %WH;
335 548
336 @_ = ($WH{$type} or Carp::croak "unsupported type passed to AnyEvent::Handle::push_write") 549 @_ = ($WH{$type} or Carp::croak "unsupported type passed to AnyEvent::Handle::push_write")
337 ->($self, @_); 550 ->($self, @_);
338 } 551 }
339 552
340 if ($self->{filter_w}) { 553 if ($self->{tls}) {
341 $self->{filter_w}->($self, \$_[0]); 554 $self->{_tls_wbuf} .= $_[0];
555 &_dotls ($self);
342 } else { 556 } else {
343 $self->{wbuf} .= $_[0]; 557 $self->{wbuf} .= $_[0];
344 $self->_drain_wbuf; 558 $self->_drain_wbuf;
345 } 559 }
346} 560}
347 561
348=item $handle->push_write (type => @args) 562=item $handle->push_write (type => @args)
349 563
350=item $handle->unshift_write (type => @args)
351
352Instead of formatting your data yourself, you can also let this module do 564Instead of formatting your data yourself, you can also let this module do
353the job by specifying a type and type-specific arguments. 565the job by specifying a type and type-specific arguments.
354 566
355Predefined types are (if you have ideas for additional types, feel free to 567Predefined types are (if you have ideas for additional types, feel free to
356drop by and tell us): 568drop by and tell us):
360=item netstring => $string 572=item netstring => $string
361 573
362Formats the given value as netstring 574Formats the given value as netstring
363(http://cr.yp.to/proto/netstrings.txt, this is not a recommendation to use them). 575(http://cr.yp.to/proto/netstrings.txt, this is not a recommendation to use them).
364 576
365=back
366
367=cut 577=cut
368 578
369register_write_type netstring => sub { 579register_write_type netstring => sub {
370 my ($self, $string) = @_; 580 my ($self, $string) = @_;
371 581
372 sprintf "%d:%s,", (length $string), $string 582 sprintf "%d:%s,", (length $string), $string
373}; 583};
374 584
585=item packstring => $format, $data
586
587An octet string prefixed with an encoded length. The encoding C<$format>
588uses the same format as a Perl C<pack> format, but must specify a single
589integer only (only one of C<cCsSlLqQiInNvVjJw> is allowed, plus an
590optional C<!>, C<< < >> or C<< > >> modifier).
591
592=cut
593
594register_write_type packstring => sub {
595 my ($self, $format, $string) = @_;
596
597 pack "$format/a*", $string
598};
599
600=item json => $array_or_hashref
601
602Encodes the given hash or array reference into a JSON object. Unless you
603provide your own JSON object, this means it will be encoded to JSON text
604in UTF-8.
605
606JSON objects (and arrays) are self-delimiting, so you can write JSON at
607one end of a handle and read them at the other end without using any
608additional framing.
609
610The generated JSON text is guaranteed not to contain any newlines: While
611this module doesn't need delimiters after or between JSON texts to be
612able to read them, many other languages depend on that.
613
614A simple RPC protocol that interoperates easily with others is to send
615JSON arrays (or objects, although arrays are usually the better choice as
616they mimic how function argument passing works) and a newline after each
617JSON text:
618
619 $handle->push_write (json => ["method", "arg1", "arg2"]); # whatever
620 $handle->push_write ("\012");
621
622An AnyEvent::Handle receiver would simply use the C<json> read type and
623rely on the fact that the newline will be skipped as leading whitespace:
624
625 $handle->push_read (json => sub { my $array = $_[1]; ... });
626
627Other languages could read single lines terminated by a newline and pass
628this line into their JSON decoder of choice.
629
630=cut
631
632register_write_type json => sub {
633 my ($self, $ref) = @_;
634
635 require JSON;
636
637 $self->{json} ? $self->{json}->encode ($ref)
638 : JSON::encode_json ($ref)
639};
640
641=item storable => $reference
642
643Freezes the given reference using L<Storable> and writes it to the
644handle. Uses the C<nfreeze> format.
645
646=cut
647
648register_write_type storable => sub {
649 my ($self, $ref) = @_;
650
651 require Storable;
652
653 pack "w/a*", Storable::nfreeze ($ref)
654};
655
656=back
657
375=item AnyEvent::Handle::register_write_type type => $coderef->($self, @args) 658=item AnyEvent::Handle::register_write_type type => $coderef->($handle, @args)
376 659
377This function (not method) lets you add your own types to C<push_write>. 660This function (not method) lets you add your own types to C<push_write>.
378Whenever the given C<type> is used, C<push_write> will invoke the code 661Whenever the given C<type> is used, C<push_write> will invoke the code
379reference with the handle object and the remaining arguments. 662reference with the handle object and the remaining arguments.
380 663
399ways, the "simple" way, using only C<on_read> and the "complex" way, using 682ways, the "simple" way, using only C<on_read> and the "complex" way, using
400a queue. 683a queue.
401 684
402In the simple case, you just install an C<on_read> callback and whenever 685In the simple case, you just install an C<on_read> callback and whenever
403new data arrives, it will be called. You can then remove some data (if 686new data arrives, it will be called. You can then remove some data (if
404enough is there) from the read buffer (C<< $handle->rbuf >>) if you want 687enough is there) from the read buffer (C<< $handle->rbuf >>). Or you cna
405or not. 688leave the data there if you want to accumulate more (e.g. when only a
689partial message has been received so far).
406 690
407In the more complex case, you want to queue multiple callbacks. In this 691In the more complex case, you want to queue multiple callbacks. In this
408case, AnyEvent::Handle will call the first queued callback each time new 692case, AnyEvent::Handle will call the first queued callback each time new
409data arrives and removes it when it has done its job (see C<push_read>, 693data arrives (also the first time it is queued) and removes it when it has
410below). 694done its job (see C<push_read>, below).
411 695
412This way you can, for example, push three line-reads, followed by reading 696This way you can, for example, push three line-reads, followed by reading
413a chunk of data, and AnyEvent::Handle will execute them in order. 697a chunk of data, and AnyEvent::Handle will execute them in order.
414 698
415Example 1: EPP protocol parser. EPP sends 4 byte length info, followed by 699Example 1: EPP protocol parser. EPP sends 4 byte length info, followed by
416the specified number of bytes which give an XML datagram. 700the specified number of bytes which give an XML datagram.
417 701
418 # in the default state, expect some header bytes 702 # in the default state, expect some header bytes
419 $handle->on_read (sub { 703 $handle->on_read (sub {
420 # some data is here, now queue the length-header-read (4 octets) 704 # some data is here, now queue the length-header-read (4 octets)
421 shift->unshift_read_chunk (4, sub { 705 shift->unshift_read (chunk => 4, sub {
422 # header arrived, decode 706 # header arrived, decode
423 my $len = unpack "N", $_[1]; 707 my $len = unpack "N", $_[1];
424 708
425 # now read the payload 709 # now read the payload
426 shift->unshift_read_chunk ($len, sub { 710 shift->unshift_read (chunk => $len, sub {
427 my $xml = $_[1]; 711 my $xml = $_[1];
428 # handle xml 712 # handle xml
429 }); 713 });
430 }); 714 });
431 }); 715 });
432 716
433Example 2: Implement a client for a protocol that replies either with 717Example 2: Implement a client for a protocol that replies either with "OK"
434"OK" and another line or "ERROR" for one request, and 64 bytes for the 718and another line or "ERROR" for the first request that is sent, and 64
435second request. Due tot he availability of a full queue, we can just 719bytes for the second request. Due to the availability of a queue, we can
436pipeline sending both requests and manipulate the queue as necessary in 720just pipeline sending both requests and manipulate the queue as necessary
437the callbacks: 721in the callbacks.
438 722
439 # request one 723When the first callback is called and sees an "OK" response, it will
724C<unshift> another line-read. This line-read will be queued I<before> the
72564-byte chunk callback.
726
727 # request one, returns either "OK + extra line" or "ERROR"
440 $handle->push_write ("request 1\015\012"); 728 $handle->push_write ("request 1\015\012");
441 729
442 # we expect "ERROR" or "OK" as response, so push a line read 730 # we expect "ERROR" or "OK" as response, so push a line read
443 $handle->push_read_line (sub { 731 $handle->push_read (line => sub {
444 # if we got an "OK", we have to _prepend_ another line, 732 # if we got an "OK", we have to _prepend_ another line,
445 # so it will be read before the second request reads its 64 bytes 733 # so it will be read before the second request reads its 64 bytes
446 # which are already in the queue when this callback is called 734 # which are already in the queue when this callback is called
447 # we don't do this in case we got an error 735 # we don't do this in case we got an error
448 if ($_[1] eq "OK") { 736 if ($_[1] eq "OK") {
449 $_[0]->unshift_read_line (sub { 737 $_[0]->unshift_read (line => sub {
450 my $response = $_[1]; 738 my $response = $_[1];
451 ... 739 ...
452 }); 740 });
453 } 741 }
454 }); 742 });
455 743
456 # request two 744 # request two, simply returns 64 octets
457 $handle->push_write ("request 2\015\012"); 745 $handle->push_write ("request 2\015\012");
458 746
459 # simply read 64 bytes, always 747 # simply read 64 bytes, always
460 $handle->push_read_chunk (64, sub { 748 $handle->push_read (chunk => 64, sub {
461 my $response = $_[1]; 749 my $response = $_[1];
462 ... 750 ...
463 }); 751 });
464 752
465=over 4 753=over 4
466 754
467=cut 755=cut
468 756
469sub _drain_rbuf { 757sub _drain_rbuf {
470 my ($self) = @_; 758 my ($self) = @_;
759
760 local $self->{_in_drain} = 1;
471 761
472 if ( 762 if (
473 defined $self->{rbuf_max} 763 defined $self->{rbuf_max}
474 && $self->{rbuf_max} < length $self->{rbuf} 764 && $self->{rbuf_max} < length $self->{rbuf}
475 ) { 765 ) {
476 $! = &Errno::ENOSPC; 766 $self->_error (&Errno::ENOSPC, 1), return;
477 $self->error;
478 } 767 }
479 768
480 return if $self->{in_drain}; 769 while () {
481 local $self->{in_drain} = 1;
482
483 while (my $len = length $self->{rbuf}) { 770 my $len = length $self->{rbuf};
484 no strict 'refs'; 771
485 if (my $cb = shift @{ $self->{queue} }) { 772 if (my $cb = shift @{ $self->{_queue} }) {
486 unless ($cb->($self)) { 773 unless ($cb->($self)) {
487 if ($self->{eof}) { 774 if ($self->{_eof}) {
488 # no progress can be made (not enough data and no data forthcoming) 775 # no progress can be made (not enough data and no data forthcoming)
489 $! = &Errno::EPIPE; 776 $self->_error (&Errno::EPIPE, 1), return;
490 $self->error;
491 } 777 }
492 778
493 unshift @{ $self->{queue} }, $cb; 779 unshift @{ $self->{_queue} }, $cb;
494 return; 780 last;
495 } 781 }
496 } elsif ($self->{on_read}) { 782 } elsif ($self->{on_read}) {
783 last unless $len;
784
497 $self->{on_read}($self); 785 $self->{on_read}($self);
498 786
499 if ( 787 if (
500 $self->{eof} # if no further data will arrive
501 && $len == length $self->{rbuf} # and no data has been consumed 788 $len == length $self->{rbuf} # if no data has been consumed
502 && !@{ $self->{queue} } # and the queue is still empty 789 && !@{ $self->{_queue} } # and the queue is still empty
503 && $self->{on_read} # and we still want to read data 790 && $self->{on_read} # but we still have on_read
504 ) { 791 ) {
792 # no further data will arrive
505 # then no progress can be made 793 # so no progress can be made
506 $! = &Errno::EPIPE; 794 $self->_error (&Errno::EPIPE, 1), return
507 $self->error; 795 if $self->{_eof};
796
797 last; # more data might arrive
508 } 798 }
509 } else { 799 } else {
510 # read side becomes idle 800 # read side becomes idle
511 delete $self->{rw}; 801 delete $self->{_rw} unless $self->{tls};
512 return; 802 last;
513 } 803 }
514 } 804 }
515 805
516 if ($self->{eof}) { 806 if ($self->{_eof}) {
517 $self->_shutdown; 807 if ($self->{on_eof}) {
518 $self->{on_eof}($self) 808 $self->{on_eof}($self)
519 if $self->{on_eof}; 809 } else {
810 $self->_error (0, 1);
811 }
812 }
813
814 # may need to restart read watcher
815 unless ($self->{_rw}) {
816 $self->start_read
817 if $self->{on_read} || @{ $self->{_queue} };
520 } 818 }
521} 819}
522 820
523=item $handle->on_read ($cb) 821=item $handle->on_read ($cb)
524 822
530 828
531sub on_read { 829sub on_read {
532 my ($self, $cb) = @_; 830 my ($self, $cb) = @_;
533 831
534 $self->{on_read} = $cb; 832 $self->{on_read} = $cb;
833 $self->_drain_rbuf if $cb && !$self->{_in_drain};
535} 834}
536 835
537=item $handle->rbuf 836=item $handle->rbuf
538 837
539Returns the read buffer (as a modifiable lvalue). 838Returns the read buffer (as a modifiable lvalue).
587 886
588 $cb = ($RH{$type} or Carp::croak "unsupported type passed to AnyEvent::Handle::push_read") 887 $cb = ($RH{$type} or Carp::croak "unsupported type passed to AnyEvent::Handle::push_read")
589 ->($self, $cb, @_); 888 ->($self, $cb, @_);
590 } 889 }
591 890
592 push @{ $self->{queue} }, $cb; 891 push @{ $self->{_queue} }, $cb;
593 $self->_drain_rbuf; 892 $self->_drain_rbuf unless $self->{_in_drain};
594} 893}
595 894
596sub unshift_read { 895sub unshift_read {
597 my $self = shift; 896 my $self = shift;
598 my $cb = pop; 897 my $cb = pop;
603 $cb = ($RH{$type} or Carp::croak "unsupported type passed to AnyEvent::Handle::unshift_read") 902 $cb = ($RH{$type} or Carp::croak "unsupported type passed to AnyEvent::Handle::unshift_read")
604 ->($self, $cb, @_); 903 ->($self, $cb, @_);
605 } 904 }
606 905
607 906
608 unshift @{ $self->{queue} }, $cb; 907 unshift @{ $self->{_queue} }, $cb;
609 $self->_drain_rbuf; 908 $self->_drain_rbuf unless $self->{_in_drain};
610} 909}
611 910
612=item $handle->push_read (type => @args, $cb) 911=item $handle->push_read (type => @args, $cb)
613 912
614=item $handle->unshift_read (type => @args, $cb) 913=item $handle->unshift_read (type => @args, $cb)
620Predefined types are (if you have ideas for additional types, feel free to 919Predefined types are (if you have ideas for additional types, feel free to
621drop by and tell us): 920drop by and tell us):
622 921
623=over 4 922=over 4
624 923
625=item chunk => $octets, $cb->($self, $data) 924=item chunk => $octets, $cb->($handle, $data)
626 925
627Invoke the callback only once C<$octets> bytes have been read. Pass the 926Invoke the callback only once C<$octets> bytes have been read. Pass the
628data read to the callback. The callback will never be called with less 927data read to the callback. The callback will never be called with less
629data. 928data.
630 929
644 $cb->($_[0], substr $_[0]{rbuf}, 0, $len, ""); 943 $cb->($_[0], substr $_[0]{rbuf}, 0, $len, "");
645 1 944 1
646 } 945 }
647}; 946};
648 947
649# compatibility with older API
650sub push_read_chunk {
651 $_[0]->push_read (chunk => $_[1], $_[2]);
652}
653
654sub unshift_read_chunk {
655 $_[0]->unshift_read (chunk => $_[1], $_[2]);
656}
657
658=item line => [$eol, ]$cb->($self, $line, $eol) 948=item line => [$eol, ]$cb->($handle, $line, $eol)
659 949
660The callback will be called only once a full line (including the end of 950The callback will be called only once a full line (including the end of
661line marker, C<$eol>) has been read. This line (excluding the end of line 951line marker, C<$eol>) has been read. This line (excluding the end of line
662marker) will be passed to the callback as second argument (C<$line>), and 952marker) will be passed to the callback as second argument (C<$line>), and
663the end of line marker as the third argument (C<$eol>). 953the end of line marker as the third argument (C<$eol>).
677=cut 967=cut
678 968
679register_read_type line => sub { 969register_read_type line => sub {
680 my ($self, $cb, $eol) = @_; 970 my ($self, $cb, $eol) = @_;
681 971
682 $eol = qr|(\015?\012)| if @_ < 3; 972 if (@_ < 3) {
973 # this is more than twice as fast as the generic code below
974 sub {
975 $_[0]{rbuf} =~ s/^([^\015\012]*)(\015?\012)// or return;
976
977 $cb->($_[0], $1, $2);
978 1
979 }
980 } else {
683 $eol = quotemeta $eol unless ref $eol; 981 $eol = quotemeta $eol unless ref $eol;
684 $eol = qr|^(.*?)($eol)|s; 982 $eol = qr|^(.*?)($eol)|s;
685 983
686 sub { 984 sub {
687 $_[0]{rbuf} =~ s/$eol// or return; 985 $_[0]{rbuf} =~ s/$eol// or return;
688 986
689 $cb->($_[0], $1, $2); 987 $cb->($_[0], $1, $2);
988 1
690 1 989 }
691 } 990 }
692}; 991};
693 992
694# compatibility with older API
695sub push_read_line {
696 my $self = shift;
697 $self->push_read (line => @_);
698}
699
700sub unshift_read_line {
701 my $self = shift;
702 $self->unshift_read (line => @_);
703}
704
705=item netstring => $cb->($string)
706
707A netstring (http://cr.yp.to/proto/netstrings.txt, this is not an endorsement).
708
709Throws an error with C<$!> set to EBADMSG on format violations.
710
711=cut
712
713register_read_type netstring => sub {
714 my ($self, $cb) = @_;
715
716 sub {
717 unless ($_[0]{rbuf} =~ s/^(0|[1-9][0-9]*)://) {
718 if ($_[0]{rbuf} =~ /[^0-9]/) {
719 $! = &Errno::EBADMSG;
720 $self->error;
721 }
722 return;
723 }
724
725 my $len = $1;
726
727 $self->unshift_read (chunk => $len, sub {
728 my $string = $_[1];
729 $_[0]->unshift_read (chunk => 1, sub {
730 if ($_[1] eq ",") {
731 $cb->($_[0], $string);
732 } else {
733 $! = &Errno::EBADMSG;
734 $self->error;
735 }
736 });
737 });
738
739 1
740 }
741};
742
743=item regex => $accept[, $reject[, $skip], $cb->($data) 993=item regex => $accept[, $reject[, $skip], $cb->($handle, $data)
744 994
745Makes a regex match against the regex object C<$accept> and returns 995Makes a regex match against the regex object C<$accept> and returns
746everything up to and including the match. 996everything up to and including the match.
747 997
748Example: read a single line terminated by '\n'. 998Example: read a single line terminated by '\n'.
796 return 1; 1046 return 1;
797 } 1047 }
798 1048
799 # reject 1049 # reject
800 if ($reject && $$rbuf =~ $reject) { 1050 if ($reject && $$rbuf =~ $reject) {
801 $! = &Errno::EBADMSG; 1051 $self->_error (&Errno::EBADMSG);
802 $self->error;
803 } 1052 }
804 1053
805 # skip 1054 # skip
806 if ($skip && $$rbuf =~ $skip) { 1055 if ($skip && $$rbuf =~ $skip) {
807 $data .= substr $$rbuf, 0, $+[0], ""; 1056 $data .= substr $$rbuf, 0, $+[0], "";
809 1058
810 () 1059 ()
811 } 1060 }
812}; 1061};
813 1062
1063=item netstring => $cb->($handle, $string)
1064
1065A netstring (http://cr.yp.to/proto/netstrings.txt, this is not an endorsement).
1066
1067Throws an error with C<$!> set to EBADMSG on format violations.
1068
1069=cut
1070
1071register_read_type netstring => sub {
1072 my ($self, $cb) = @_;
1073
1074 sub {
1075 unless ($_[0]{rbuf} =~ s/^(0|[1-9][0-9]*)://) {
1076 if ($_[0]{rbuf} =~ /[^0-9]/) {
1077 $self->_error (&Errno::EBADMSG);
1078 }
1079 return;
1080 }
1081
1082 my $len = $1;
1083
1084 $self->unshift_read (chunk => $len, sub {
1085 my $string = $_[1];
1086 $_[0]->unshift_read (chunk => 1, sub {
1087 if ($_[1] eq ",") {
1088 $cb->($_[0], $string);
1089 } else {
1090 $self->_error (&Errno::EBADMSG);
1091 }
1092 });
1093 });
1094
1095 1
1096 }
1097};
1098
1099=item packstring => $format, $cb->($handle, $string)
1100
1101An octet string prefixed with an encoded length. The encoding C<$format>
1102uses the same format as a Perl C<pack> format, but must specify a single
1103integer only (only one of C<cCsSlLqQiInNvVjJw> is allowed, plus an
1104optional C<!>, C<< < >> or C<< > >> modifier).
1105
1106DNS over TCP uses a prefix of C<n>, EPP uses a prefix of C<N>.
1107
1108Example: read a block of data prefixed by its length in BER-encoded
1109format (very efficient).
1110
1111 $handle->push_read (packstring => "w", sub {
1112 my ($handle, $data) = @_;
1113 });
1114
1115=cut
1116
1117register_read_type packstring => sub {
1118 my ($self, $cb, $format) = @_;
1119
1120 sub {
1121 # when we can use 5.10 we can use ".", but for 5.8 we use the re-pack method
1122 defined (my $len = eval { unpack $format, $_[0]{rbuf} })
1123 or return;
1124
1125 $format = length pack $format, $len;
1126
1127 # bypass unshift if we already have the remaining chunk
1128 if ($format + $len <= length $_[0]{rbuf}) {
1129 my $data = substr $_[0]{rbuf}, $format, $len;
1130 substr $_[0]{rbuf}, 0, $format + $len, "";
1131 $cb->($_[0], $data);
1132 } else {
1133 # remove prefix
1134 substr $_[0]{rbuf}, 0, $format, "";
1135
1136 # read remaining chunk
1137 $_[0]->unshift_read (chunk => $len, $cb);
1138 }
1139
1140 1
1141 }
1142};
1143
1144=item json => $cb->($handle, $hash_or_arrayref)
1145
1146Reads a JSON object or array, decodes it and passes it to the callback.
1147
1148If a C<json> object was passed to the constructor, then that will be used
1149for the final decode, otherwise it will create a JSON coder expecting UTF-8.
1150
1151This read type uses the incremental parser available with JSON version
11522.09 (and JSON::XS version 2.2) and above. You have to provide a
1153dependency on your own: this module will load the JSON module, but
1154AnyEvent does not depend on it itself.
1155
1156Since JSON texts are fully self-delimiting, the C<json> read and write
1157types are an ideal simple RPC protocol: just exchange JSON datagrams. See
1158the C<json> write type description, above, for an actual example.
1159
1160=cut
1161
1162register_read_type json => sub {
1163 my ($self, $cb) = @_;
1164
1165 require JSON;
1166
1167 my $data;
1168 my $rbuf = \$self->{rbuf};
1169
1170 my $json = $self->{json} ||= JSON->new->utf8;
1171
1172 sub {
1173 my $ref = $json->incr_parse ($self->{rbuf});
1174
1175 if ($ref) {
1176 $self->{rbuf} = $json->incr_text;
1177 $json->incr_text = "";
1178 $cb->($self, $ref);
1179
1180 1
1181 } else {
1182 $self->{rbuf} = "";
1183 ()
1184 }
1185 }
1186};
1187
1188=item storable => $cb->($handle, $ref)
1189
1190Deserialises a L<Storable> frozen representation as written by the
1191C<storable> write type (BER-encoded length prefix followed by nfreeze'd
1192data).
1193
1194Raises C<EBADMSG> error if the data could not be decoded.
1195
1196=cut
1197
1198register_read_type storable => sub {
1199 my ($self, $cb) = @_;
1200
1201 require Storable;
1202
1203 sub {
1204 # when we can use 5.10 we can use ".", but for 5.8 we use the re-pack method
1205 defined (my $len = eval { unpack "w", $_[0]{rbuf} })
1206 or return;
1207
1208 my $format = length pack "w", $len;
1209
1210 # bypass unshift if we already have the remaining chunk
1211 if ($format + $len <= length $_[0]{rbuf}) {
1212 my $data = substr $_[0]{rbuf}, $format, $len;
1213 substr $_[0]{rbuf}, 0, $format + $len, "";
1214 $cb->($_[0], Storable::thaw ($data));
1215 } else {
1216 # remove prefix
1217 substr $_[0]{rbuf}, 0, $format, "";
1218
1219 # read remaining chunk
1220 $_[0]->unshift_read (chunk => $len, sub {
1221 if (my $ref = eval { Storable::thaw ($_[1]) }) {
1222 $cb->($_[0], $ref);
1223 } else {
1224 $self->_error (&Errno::EBADMSG);
1225 }
1226 });
1227 }
1228
1229 1
1230 }
1231};
1232
814=back 1233=back
815 1234
816=item AnyEvent::Handle::register_read_type type => $coderef->($self, $cb, @args) 1235=item AnyEvent::Handle::register_read_type type => $coderef->($handle, $cb, @args)
817 1236
818This function (not method) lets you add your own types to C<push_read>. 1237This function (not method) lets you add your own types to C<push_read>.
819 1238
820Whenever the given C<type> is used, C<push_read> will invoke the code 1239Whenever the given C<type> is used, C<push_read> will invoke the code
821reference with the handle object, the callback and the remaining 1240reference with the handle object, the callback and the remaining
823 1242
824The code reference is supposed to return a callback (usually a closure) 1243The code reference is supposed to return a callback (usually a closure)
825that works as a plain read callback (see C<< ->push_read ($cb) >>). 1244that works as a plain read callback (see C<< ->push_read ($cb) >>).
826 1245
827It should invoke the passed callback when it is done reading (remember to 1246It should invoke the passed callback when it is done reading (remember to
828pass C<$self> as first argument as all other callbacks do that). 1247pass C<$handle> as first argument as all other callbacks do that).
829 1248
830Note that this is a function, and all types registered this way will be 1249Note that this is a function, and all types registered this way will be
831global, so try to use unique names. 1250global, so try to use unique names.
832 1251
833For examples, see the source of this module (F<perldoc -m AnyEvent::Handle>, 1252For examples, see the source of this module (F<perldoc -m AnyEvent::Handle>,
836=item $handle->stop_read 1255=item $handle->stop_read
837 1256
838=item $handle->start_read 1257=item $handle->start_read
839 1258
840In rare cases you actually do not want to read anything from the 1259In rare cases you actually do not want to read anything from the
841socket. In this case you can call C<stop_read>. Neither C<on_read> no 1260socket. In this case you can call C<stop_read>. Neither C<on_read> nor
842any queued callbacks will be executed then. To start reading again, call 1261any queued callbacks will be executed then. To start reading again, call
843C<start_read>. 1262C<start_read>.
844 1263
1264Note that AnyEvent::Handle will automatically C<start_read> for you when
1265you change the C<on_read> callback or push/unshift a read callback, and it
1266will automatically C<stop_read> for you when neither C<on_read> is set nor
1267there are any read requests in the queue.
1268
1269These methods will have no effect when in TLS mode (as TLS doesn't support
1270half-duplex connections).
1271
845=cut 1272=cut
846 1273
847sub stop_read { 1274sub stop_read {
848 my ($self) = @_; 1275 my ($self) = @_;
849 1276
850 delete $self->{rw}; 1277 delete $self->{_rw} unless $self->{tls};
851} 1278}
852 1279
853sub start_read { 1280sub start_read {
854 my ($self) = @_; 1281 my ($self) = @_;
855 1282
856 unless ($self->{rw} || $self->{eof}) { 1283 unless ($self->{_rw} || $self->{_eof}) {
857 Scalar::Util::weaken $self; 1284 Scalar::Util::weaken $self;
858 1285
859 $self->{rw} = AnyEvent->io (fh => $self->{fh}, poll => "r", cb => sub { 1286 $self->{_rw} = AnyEvent->io (fh => $self->{fh}, poll => "r", cb => sub {
860 my $rbuf = $self->{filter_r} ? \my $buf : \$self->{rbuf}; 1287 my $rbuf = \($self->{tls} ? my $buf : $self->{rbuf});
861 my $len = sysread $self->{fh}, $$rbuf, $self->{read_size} || 8192, length $$rbuf; 1288 my $len = sysread $self->{fh}, $$rbuf, $self->{read_size} || 8192, length $$rbuf;
862 1289
863 if ($len > 0) { 1290 if ($len > 0) {
864 $self->{filter_r} 1291 $self->{_activity} = AnyEvent->now;
865 ? $self->{filter_r}->($self, $rbuf) 1292
866 : $self->_drain_rbuf; 1293 if ($self->{tls}) {
1294 Net::SSLeay::BIO_write ($self->{_rbio}, $$rbuf);
1295 &_dotls ($self);
1296 } else {
1297 $self->_drain_rbuf unless $self->{_in_drain};
1298 }
867 1299
868 } elsif (defined $len) { 1300 } elsif (defined $len) {
869 delete $self->{rw}; 1301 delete $self->{_rw};
870 $self->{eof} = 1; 1302 $self->{_eof} = 1;
871 $self->_drain_rbuf; 1303 $self->_drain_rbuf unless $self->{_in_drain};
872 1304
873 } elsif ($! != EAGAIN && $! != EINTR && $! != &AnyEvent::Util::WSAWOULDBLOCK) { 1305 } elsif ($! != EAGAIN && $! != EINTR && $! != WSAEWOULDBLOCK) {
874 return $self->error; 1306 return $self->_error ($!, 1);
875 } 1307 }
876 }); 1308 });
877 } 1309 }
878} 1310}
879 1311
880sub _dotls { 1312sub _dotls {
881 my ($self) = @_; 1313 my ($self) = @_;
882 1314
1315 my $buf;
1316
883 if (length $self->{tls_wbuf}) { 1317 if (length $self->{_tls_wbuf}) {
884 while ((my $len = Net::SSLeay::write ($self->{tls}, $self->{tls_wbuf})) > 0) { 1318 while ((my $len = Net::SSLeay::write ($self->{tls}, $self->{_tls_wbuf})) > 0) {
885 substr $self->{tls_wbuf}, 0, $len, ""; 1319 substr $self->{_tls_wbuf}, 0, $len, "";
886 } 1320 }
887 } 1321 }
888 1322
1323 while (defined ($buf = Net::SSLeay::read ($self->{tls}))) {
1324 unless (length $buf) {
1325 # let's treat SSL-eof as we treat normal EOF
1326 delete $self->{_rw};
1327 $self->{_eof} = 1;
1328 &_freetls;
1329 }
1330
1331 $self->{rbuf} .= $buf;
1332 $self->_drain_rbuf unless $self->{_in_drain};
1333 $self->{tls} or return; # tls session might have gone away in callback
1334 }
1335
1336 my $err = Net::SSLeay::get_error ($self->{tls}, -1);
1337
1338 if ($err!= Net::SSLeay::ERROR_WANT_READ ()) {
1339 if ($err == Net::SSLeay::ERROR_SYSCALL ()) {
1340 return $self->_error ($!, 1);
1341 } elsif ($err == Net::SSLeay::ERROR_SSL ()) {
1342 return $self->_error (&Errno::EIO, 1);
1343 }
1344
1345 # all others are fine for our purposes
1346 }
1347
889 if (defined (my $buf = Net::SSLeay::BIO_read ($self->{tls_wbio}))) { 1348 if (length ($buf = Net::SSLeay::BIO_read ($self->{_wbio}))) {
890 $self->{wbuf} .= $buf; 1349 $self->{wbuf} .= $buf;
891 $self->_drain_wbuf; 1350 $self->_drain_wbuf;
892 }
893
894 while (defined (my $buf = Net::SSLeay::read ($self->{tls}))) {
895 $self->{rbuf} .= $buf;
896 $self->_drain_rbuf;
897 }
898
899 my $err = Net::SSLeay::get_error ($self->{tls}, -1);
900
901 if ($err!= Net::SSLeay::ERROR_WANT_READ ()) {
902 if ($err == Net::SSLeay::ERROR_SYSCALL ()) {
903 $self->error;
904 } elsif ($err == Net::SSLeay::ERROR_SSL ()) {
905 $! = &Errno::EIO;
906 $self->error;
907 }
908
909 # all others are fine for our purposes
910 } 1351 }
911} 1352}
912 1353
913=item $handle->starttls ($tls[, $tls_ctx]) 1354=item $handle->starttls ($tls[, $tls_ctx])
914 1355
920C<"connect">, C<"accept"> or an existing Net::SSLeay object). 1361C<"connect">, C<"accept"> or an existing Net::SSLeay object).
921 1362
922The second argument is the optional C<Net::SSLeay::CTX> object that is 1363The second argument is the optional C<Net::SSLeay::CTX> object that is
923used when AnyEvent::Handle has to create its own TLS connection object. 1364used when AnyEvent::Handle has to create its own TLS connection object.
924 1365
925=cut 1366The TLS connection object will end up in C<< $handle->{tls} >> after this
1367call and can be used or changed to your liking. Note that the handshake
1368might have already started when this function returns.
926 1369
927# TODO: maybe document... 1370If it an error to start a TLS handshake more than once per
1371AnyEvent::Handle object (this is due to bugs in OpenSSL).
1372
1373=cut
1374
928sub starttls { 1375sub starttls {
929 my ($self, $ssl, $ctx) = @_; 1376 my ($self, $ssl, $ctx) = @_;
930 1377
931 $self->stoptls; 1378 require Net::SSLeay;
932 1379
1380 Carp::croak "it is an error to call starttls more than once on an Anyevent::Handle object"
1381 if $self->{tls};
1382
933 if ($ssl eq "accept") { 1383 if ($ssl eq "accept") {
934 $ssl = Net::SSLeay::new ($ctx || TLS_CTX ()); 1384 $ssl = Net::SSLeay::new ($ctx || TLS_CTX ());
935 Net::SSLeay::set_accept_state ($ssl); 1385 Net::SSLeay::set_accept_state ($ssl);
936 } elsif ($ssl eq "connect") { 1386 } elsif ($ssl eq "connect") {
937 $ssl = Net::SSLeay::new ($ctx || TLS_CTX ()); 1387 $ssl = Net::SSLeay::new ($ctx || TLS_CTX ());
943 # basically, this is deep magic (because SSL_read should have the same issues) 1393 # basically, this is deep magic (because SSL_read should have the same issues)
944 # but the openssl maintainers basically said: "trust us, it just works". 1394 # but the openssl maintainers basically said: "trust us, it just works".
945 # (unfortunately, we have to hardcode constants because the abysmally misdesigned 1395 # (unfortunately, we have to hardcode constants because the abysmally misdesigned
946 # and mismaintained ssleay-module doesn't even offer them). 1396 # and mismaintained ssleay-module doesn't even offer them).
947 # http://www.mail-archive.com/openssl-dev@openssl.org/msg22420.html 1397 # http://www.mail-archive.com/openssl-dev@openssl.org/msg22420.html
1398 #
1399 # in short: this is a mess.
1400 #
1401 # note that we do not try to keep the length constant between writes as we are required to do.
1402 # we assume that most (but not all) of this insanity only applies to non-blocking cases,
1403 # and we drive openssl fully in blocking mode here. Or maybe we don't - openssl seems to
1404 # have identity issues in that area.
948 Net::SSLeay::CTX_set_mode ($self->{tls}, 1405 Net::SSLeay::CTX_set_mode ($self->{tls},
949 (eval { local $SIG{__DIE__}; Net::SSLeay::MODE_ENABLE_PARTIAL_WRITE () } || 1) 1406 (eval { local $SIG{__DIE__}; Net::SSLeay::MODE_ENABLE_PARTIAL_WRITE () } || 1)
950 | (eval { local $SIG{__DIE__}; Net::SSLeay::MODE_ACCEPT_MOVING_WRITE_BUFFER () } || 2)); 1407 | (eval { local $SIG{__DIE__}; Net::SSLeay::MODE_ACCEPT_MOVING_WRITE_BUFFER () } || 2));
951 1408
952 $self->{tls_rbio} = Net::SSLeay::BIO_new (Net::SSLeay::BIO_s_mem ()); 1409 $self->{_rbio} = Net::SSLeay::BIO_new (Net::SSLeay::BIO_s_mem ());
953 $self->{tls_wbio} = Net::SSLeay::BIO_new (Net::SSLeay::BIO_s_mem ()); 1410 $self->{_wbio} = Net::SSLeay::BIO_new (Net::SSLeay::BIO_s_mem ());
954 1411
955 Net::SSLeay::set_bio ($ssl, $self->{tls_rbio}, $self->{tls_wbio}); 1412 Net::SSLeay::set_bio ($ssl, $self->{_rbio}, $self->{_wbio});
956 1413
957 $self->{filter_w} = sub { 1414 &_dotls; # need to trigger the initial handshake
958 $_[0]{tls_wbuf} .= ${$_[1]}; 1415 $self->start_read; # make sure we actually do read
959 &_dotls;
960 };
961 $self->{filter_r} = sub {
962 Net::SSLeay::BIO_write ($_[0]{tls_rbio}, ${$_[1]});
963 &_dotls;
964 };
965} 1416}
966 1417
967=item $handle->stoptls 1418=item $handle->stoptls
968 1419
969Destroys the SSL connection, if any. Partial read or write data will be 1420Shuts down the SSL connection - this makes a proper EOF handshake by
970lost. 1421sending a close notify to the other side, but since OpenSSL doesn't
1422support non-blocking shut downs, it is not possible to re-use the stream
1423afterwards.
971 1424
972=cut 1425=cut
973 1426
974sub stoptls { 1427sub stoptls {
975 my ($self) = @_; 1428 my ($self) = @_;
976 1429
1430 if ($self->{tls}) {
1431 Net::SSLeay::shutdown ($self->{tls});
1432
1433 &_dotls;
1434
1435 # we don't give a shit. no, we do, but we can't. no...
1436 # we, we... have to use openssl :/
1437 &_freetls;
1438 }
1439}
1440
1441sub _freetls {
1442 my ($self) = @_;
1443
1444 return unless $self->{tls};
1445
977 Net::SSLeay::free (delete $self->{tls}) if $self->{tls}; 1446 Net::SSLeay::free (delete $self->{tls});
978 delete $self->{tls_rbio}; 1447
979 delete $self->{tls_wbio}; 1448 delete @$self{qw(_rbio _wbio _tls_wbuf)};
980 delete $self->{tls_wbuf};
981 delete $self->{filter_r};
982 delete $self->{filter_w};
983} 1449}
984 1450
985sub DESTROY { 1451sub DESTROY {
986 my $self = shift; 1452 my $self = shift;
987 1453
988 $self->stoptls; 1454 &_freetls;
1455
1456 my $linger = exists $self->{linger} ? $self->{linger} : 3600;
1457
1458 if ($linger && length $self->{wbuf}) {
1459 my $fh = delete $self->{fh};
1460 my $wbuf = delete $self->{wbuf};
1461
1462 my @linger;
1463
1464 push @linger, AnyEvent->io (fh => $fh, poll => "w", cb => sub {
1465 my $len = syswrite $fh, $wbuf, length $wbuf;
1466
1467 if ($len > 0) {
1468 substr $wbuf, 0, $len, "";
1469 } else {
1470 @linger = (); # end
1471 }
1472 });
1473 push @linger, AnyEvent->timer (after => $linger, cb => sub {
1474 @linger = ();
1475 });
1476 }
989} 1477}
990 1478
991=item AnyEvent::Handle::TLS_CTX 1479=item AnyEvent::Handle::TLS_CTX
992 1480
993This function creates and returns the Net::SSLeay::CTX object used by 1481This function creates and returns the Net::SSLeay::CTX object used by
1023 } 1511 }
1024} 1512}
1025 1513
1026=back 1514=back
1027 1515
1516
1517=head1 NONFREQUENTLY ASKED QUESTIONS
1518
1519=over 4
1520
1521=item How do I read data until the other side closes the connection?
1522
1523If you just want to read your data into a perl scalar, the easiest way to achieve this is
1524by setting an C<on_read> callback that does nothing, clearing the C<on_eof> callback
1525and in the C<on_error> callback, the data will be in C<$_[0]{rbuf}>:
1526
1527 $handle->on_read (sub { });
1528 $handle->on_eof (undef);
1529 $handle->on_error (sub {
1530 my $data = delete $_[0]{rbuf};
1531 undef $handle;
1532 });
1533
1534The reason to use C<on_error> is that TCP connections, due to latencies
1535and packets loss, might get closed quite violently with an error, when in
1536fact, all data has been received.
1537
1538It is usually better to use acknowledgements when transfering data,
1539to make sure the other side hasn't just died and you got the data
1540intact. This is also one reason why so many internet protocols have an
1541explicit QUIT command.
1542
1543
1544=item I don't want to destroy the handle too early - how do I wait until all data has been sent?
1545
1546After writing your last bits of data, set the C<on_drain> callback
1547and destroy the handle in there - with the default setting of
1548C<low_water_mark> this will be called precisely when all data has been
1549written to the socket:
1550
1551 $handle->push_write (...);
1552 $handle->on_drain (sub {
1553 warn "all data submitted to the kernel\n";
1554 undef $handle;
1555 });
1556
1557=back
1558
1559
1560=head1 SUBCLASSING AnyEvent::Handle
1561
1562In many cases, you might want to subclass AnyEvent::Handle.
1563
1564To make this easier, a given version of AnyEvent::Handle uses these
1565conventions:
1566
1567=over 4
1568
1569=item * all constructor arguments become object members.
1570
1571At least initially, when you pass a C<tls>-argument to the constructor it
1572will end up in C<< $handle->{tls} >>. Those members might be changed or
1573mutated later on (for example C<tls> will hold the TLS connection object).
1574
1575=item * other object member names are prefixed with an C<_>.
1576
1577All object members not explicitly documented (internal use) are prefixed
1578with an underscore character, so the remaining non-C<_>-namespace is free
1579for use for subclasses.
1580
1581=item * all members not documented here and not prefixed with an underscore
1582are free to use in subclasses.
1583
1584Of course, new versions of AnyEvent::Handle may introduce more "public"
1585member variables, but thats just life, at least it is documented.
1586
1587=back
1588
1028=head1 AUTHOR 1589=head1 AUTHOR
1029 1590
1030Robin Redeker C<< <elmex at ta-sa.org> >>, Marc Lehmann <schmorp@schmorp.de>. 1591Robin Redeker C<< <elmex at ta-sa.org> >>, Marc Lehmann <schmorp@schmorp.de>.
1031 1592
1032=cut 1593=cut

Diff Legend

Removed lines
+ Added lines
< Changed lines
> Changed lines