1 |
package AnyEvent::Handle; |
2 |
|
3 |
no warnings; |
4 |
use strict; |
5 |
|
6 |
use AnyEvent (); |
7 |
use AnyEvent::Util qw(WSAWOULDBLOCK); |
8 |
use Scalar::Util (); |
9 |
use Carp (); |
10 |
use Fcntl (); |
11 |
use Errno qw/EAGAIN EINTR/; |
12 |
|
13 |
=head1 NAME |
14 |
|
15 |
AnyEvent::Handle - non-blocking I/O on file handles via AnyEvent |
16 |
|
17 |
=cut |
18 |
|
19 |
our $VERSION = '0.04'; |
20 |
|
21 |
=head1 SYNOPSIS |
22 |
|
23 |
use AnyEvent; |
24 |
use AnyEvent::Handle; |
25 |
|
26 |
my $cv = AnyEvent->condvar; |
27 |
|
28 |
my $handle = |
29 |
AnyEvent::Handle->new ( |
30 |
fh => \*STDIN, |
31 |
on_eof => sub { |
32 |
$cv->broadcast; |
33 |
}, |
34 |
); |
35 |
|
36 |
# send some request line |
37 |
$handle->push_write ("getinfo\015\012"); |
38 |
|
39 |
# read the response line |
40 |
$handle->push_read (line => sub { |
41 |
my ($handle, $line) = @_; |
42 |
warn "read line <$line>\n"; |
43 |
$cv->send; |
44 |
}); |
45 |
|
46 |
$cv->recv; |
47 |
|
48 |
=head1 DESCRIPTION |
49 |
|
50 |
This module is a helper module to make it easier to do event-based I/O on |
51 |
filehandles. For utility functions for doing non-blocking connects and accepts |
52 |
on sockets see L<AnyEvent::Util>. |
53 |
|
54 |
In the following, when the documentation refers to of "bytes" then this |
55 |
means characters. As sysread and syswrite are used for all I/O, their |
56 |
treatment of characters applies to this module as well. |
57 |
|
58 |
All callbacks will be invoked with the handle object as their first |
59 |
argument. |
60 |
|
61 |
=head1 METHODS |
62 |
|
63 |
=over 4 |
64 |
|
65 |
=item B<new (%args)> |
66 |
|
67 |
The constructor supports these arguments (all as key => value pairs). |
68 |
|
69 |
=over 4 |
70 |
|
71 |
=item fh => $filehandle [MANDATORY] |
72 |
|
73 |
The filehandle this L<AnyEvent::Handle> object will operate on. |
74 |
|
75 |
NOTE: The filehandle will be set to non-blocking (using |
76 |
AnyEvent::Util::fh_nonblocking). |
77 |
|
78 |
=item on_eof => $cb->($self) |
79 |
|
80 |
Set the callback to be called on EOF. |
81 |
|
82 |
While not mandatory, it is highly recommended to set an eof callback, |
83 |
otherwise you might end up with a closed socket while you are still |
84 |
waiting for data. |
85 |
|
86 |
=item on_error => $cb->($self) |
87 |
|
88 |
This is the fatal error callback, that is called when, well, a fatal error |
89 |
occurs, such as not being able to resolve the hostname, failure to connect |
90 |
or a read error. |
91 |
|
92 |
The object will not be in a usable state when this callback has been |
93 |
called. |
94 |
|
95 |
On callback entrance, the value of C<$!> contains the operating system |
96 |
error (or C<ENOSPC>, C<EPIPE> or C<EBADMSG>). |
97 |
|
98 |
The callback should throw an exception. If it returns, then |
99 |
AnyEvent::Handle will C<croak> for you. |
100 |
|
101 |
While not mandatory, it is I<highly> recommended to set this callback, as |
102 |
you will not be notified of errors otherwise. The default simply calls |
103 |
die. |
104 |
|
105 |
=item on_read => $cb->($self) |
106 |
|
107 |
This sets the default read callback, which is called when data arrives |
108 |
and no read request is in the queue. |
109 |
|
110 |
To access (and remove data from) the read buffer, use the C<< ->rbuf >> |
111 |
method or access the C<$self->{rbuf}> member directly. |
112 |
|
113 |
When an EOF condition is detected then AnyEvent::Handle will first try to |
114 |
feed all the remaining data to the queued callbacks and C<on_read> before |
115 |
calling the C<on_eof> callback. If no progress can be made, then a fatal |
116 |
error will be raised (with C<$!> set to C<EPIPE>). |
117 |
|
118 |
=item on_drain => $cb->() |
119 |
|
120 |
This sets the callback that is called when the write buffer becomes empty |
121 |
(or when the callback is set and the buffer is empty already). |
122 |
|
123 |
To append to the write buffer, use the C<< ->push_write >> method. |
124 |
|
125 |
=item rbuf_max => <bytes> |
126 |
|
127 |
If defined, then a fatal error will be raised (with C<$!> set to C<ENOSPC>) |
128 |
when the read buffer ever (strictly) exceeds this size. This is useful to |
129 |
avoid denial-of-service attacks. |
130 |
|
131 |
For example, a server accepting connections from untrusted sources should |
132 |
be 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 |
134 |
amount of data without a callback ever being called as long as the line |
135 |
isn't finished). |
136 |
|
137 |
=item read_size => <bytes> |
138 |
|
139 |
The default read block size (the amount of bytes this module will try to read |
140 |
on each [loop iteration). Default: C<4096>. |
141 |
|
142 |
=item low_water_mark => <bytes> |
143 |
|
144 |
Sets the amount of bytes (default: C<0>) that make up an "empty" write |
145 |
buffer: If the write reaches this size or gets even samller it is |
146 |
considered empty. |
147 |
|
148 |
=item tls => "accept" | "connect" | Net::SSLeay::SSL object |
149 |
|
150 |
When this parameter is given, it enables TLS (SSL) mode, that means it |
151 |
will start making tls handshake and will transparently encrypt/decrypt |
152 |
data. |
153 |
|
154 |
TLS mode requires Net::SSLeay to be installed (it will be loaded |
155 |
automatically when you try to create a TLS handle). |
156 |
|
157 |
For the TLS server side, use C<accept>, and for the TLS client side of a |
158 |
connection, use C<connect> mode. |
159 |
|
160 |
You can also provide your own TLS connection object, but you have |
161 |
to make sure that you call either C<Net::SSLeay::set_connect_state> |
162 |
or C<Net::SSLeay::set_accept_state> on it before you pass it to |
163 |
AnyEvent::Handle. |
164 |
|
165 |
See the C<starttls> method if you need to start TLs negotiation later. |
166 |
|
167 |
=item tls_ctx => $ssl_ctx |
168 |
|
169 |
Use the given Net::SSLeay::CTX object to create the new TLS connection |
170 |
(unless a connection object was specified directly). If this parameter is |
171 |
missing, then AnyEvent::Handle will use C<AnyEvent::Handle::TLS_CTX>. |
172 |
|
173 |
=item filter_r => $cb |
174 |
|
175 |
=item filter_w => $cb |
176 |
|
177 |
These exist, but are undocumented at this time. |
178 |
|
179 |
=back |
180 |
|
181 |
=cut |
182 |
|
183 |
sub new { |
184 |
my $class = shift; |
185 |
|
186 |
my $self = bless { @_ }, $class; |
187 |
|
188 |
$self->{fh} or Carp::croak "mandatory argument fh is missing"; |
189 |
|
190 |
AnyEvent::Util::fh_nonblocking $self->{fh}, 1; |
191 |
|
192 |
if ($self->{tls}) { |
193 |
require Net::SSLeay; |
194 |
$self->starttls (delete $self->{tls}, delete $self->{tls_ctx}); |
195 |
} |
196 |
|
197 |
$self->on_eof (delete $self->{on_eof} ) if $self->{on_eof}; |
198 |
$self->on_error (delete $self->{on_error}) if $self->{on_error}; |
199 |
$self->on_drain (delete $self->{on_drain}) if $self->{on_drain}; |
200 |
$self->on_read (delete $self->{on_read} ) if $self->{on_read}; |
201 |
|
202 |
$self->start_read; |
203 |
|
204 |
$self |
205 |
} |
206 |
|
207 |
sub _shutdown { |
208 |
my ($self) = @_; |
209 |
|
210 |
delete $self->{_rw}; |
211 |
delete $self->{_ww}; |
212 |
delete $self->{fh}; |
213 |
} |
214 |
|
215 |
sub error { |
216 |
my ($self) = @_; |
217 |
|
218 |
{ |
219 |
local $!; |
220 |
$self->_shutdown; |
221 |
} |
222 |
|
223 |
$self->{on_error}($self) |
224 |
if $self->{on_error}; |
225 |
|
226 |
Carp::croak "AnyEvent::Handle uncaught fatal error: $!"; |
227 |
} |
228 |
|
229 |
=item $fh = $handle->fh |
230 |
|
231 |
This method returns the file handle of the L<AnyEvent::Handle> object. |
232 |
|
233 |
=cut |
234 |
|
235 |
sub fh { $_[0]{fh} } |
236 |
|
237 |
=item $handle->on_error ($cb) |
238 |
|
239 |
Replace the current C<on_error> callback (see the C<on_error> constructor argument). |
240 |
|
241 |
=cut |
242 |
|
243 |
sub on_error { |
244 |
$_[0]{on_error} = $_[1]; |
245 |
} |
246 |
|
247 |
=item $handle->on_eof ($cb) |
248 |
|
249 |
Replace the current C<on_eof> callback (see the C<on_eof> constructor argument). |
250 |
|
251 |
=cut |
252 |
|
253 |
sub on_eof { |
254 |
$_[0]{on_eof} = $_[1]; |
255 |
} |
256 |
|
257 |
############################################################################# |
258 |
|
259 |
=back |
260 |
|
261 |
=head2 WRITE QUEUE |
262 |
|
263 |
AnyEvent::Handle manages two queues per handle, one for writing and one |
264 |
for reading. |
265 |
|
266 |
The write queue is very simple: you can add data to its end, and |
267 |
AnyEvent::Handle will automatically try to get rid of it for you. |
268 |
|
269 |
When data could be written and the write buffer is shorter then the low |
270 |
water mark, the C<on_drain> callback will be invoked. |
271 |
|
272 |
=over 4 |
273 |
|
274 |
=item $handle->on_drain ($cb) |
275 |
|
276 |
Sets the C<on_drain> callback or clears it (see the description of |
277 |
C<on_drain> in the constructor). |
278 |
|
279 |
=cut |
280 |
|
281 |
sub on_drain { |
282 |
my ($self, $cb) = @_; |
283 |
|
284 |
$self->{on_drain} = $cb; |
285 |
|
286 |
$cb->($self) |
287 |
if $cb && $self->{low_water_mark} >= length $self->{wbuf}; |
288 |
} |
289 |
|
290 |
=item $handle->push_write ($data) |
291 |
|
292 |
Queues the given scalar to be written. You can push as much data as you |
293 |
want (only limited by the available memory), as C<AnyEvent::Handle> |
294 |
buffers it independently of the kernel. |
295 |
|
296 |
=cut |
297 |
|
298 |
sub _drain_wbuf { |
299 |
my ($self) = @_; |
300 |
|
301 |
if (!$self->{_ww} && length $self->{wbuf}) { |
302 |
|
303 |
Scalar::Util::weaken $self; |
304 |
|
305 |
my $cb = sub { |
306 |
my $len = syswrite $self->{fh}, $self->{wbuf}; |
307 |
|
308 |
if ($len >= 0) { |
309 |
substr $self->{wbuf}, 0, $len, ""; |
310 |
|
311 |
$self->{on_drain}($self) |
312 |
if $self->{low_water_mark} >= length $self->{wbuf} |
313 |
&& $self->{on_drain}; |
314 |
|
315 |
delete $self->{_ww} unless length $self->{wbuf}; |
316 |
} elsif ($! != EAGAIN && $! != EINTR && $! != WSAWOULDBLOCK) { |
317 |
$self->error; |
318 |
} |
319 |
}; |
320 |
|
321 |
# try to write data immediately |
322 |
$cb->(); |
323 |
|
324 |
# if still data left in wbuf, we need to poll |
325 |
$self->{_ww} = AnyEvent->io (fh => $self->{fh}, poll => "w", cb => $cb) |
326 |
if length $self->{wbuf}; |
327 |
}; |
328 |
} |
329 |
|
330 |
our %WH; |
331 |
|
332 |
sub register_write_type($$) { |
333 |
$WH{$_[0]} = $_[1]; |
334 |
} |
335 |
|
336 |
sub push_write { |
337 |
my $self = shift; |
338 |
|
339 |
if (@_ > 1) { |
340 |
my $type = shift; |
341 |
|
342 |
@_ = ($WH{$type} or Carp::croak "unsupported type passed to AnyEvent::Handle::push_write") |
343 |
->($self, @_); |
344 |
} |
345 |
|
346 |
if ($self->{filter_w}) { |
347 |
$self->{filter_w}->($self, \$_[0]); |
348 |
} else { |
349 |
$self->{wbuf} .= $_[0]; |
350 |
$self->_drain_wbuf; |
351 |
} |
352 |
} |
353 |
|
354 |
=item $handle->push_write (type => @args) |
355 |
|
356 |
=item $handle->unshift_write (type => @args) |
357 |
|
358 |
Instead of formatting your data yourself, you can also let this module do |
359 |
the job by specifying a type and type-specific arguments. |
360 |
|
361 |
Predefined types are (if you have ideas for additional types, feel free to |
362 |
drop by and tell us): |
363 |
|
364 |
=over 4 |
365 |
|
366 |
=item netstring => $string |
367 |
|
368 |
Formats the given value as netstring |
369 |
(http://cr.yp.to/proto/netstrings.txt, this is not a recommendation to use them). |
370 |
|
371 |
=back |
372 |
|
373 |
=cut |
374 |
|
375 |
register_write_type netstring => sub { |
376 |
my ($self, $string) = @_; |
377 |
|
378 |
sprintf "%d:%s,", (length $string), $string |
379 |
}; |
380 |
|
381 |
=item json => $array_or_hashref |
382 |
|
383 |
=item AnyEvent::Handle::register_write_type type => $coderef->($self, @args) |
384 |
|
385 |
This function (not method) lets you add your own types to C<push_write>. |
386 |
Whenever the given C<type> is used, C<push_write> will invoke the code |
387 |
reference with the handle object and the remaining arguments. |
388 |
|
389 |
The code reference is supposed to return a single octet string that will |
390 |
be appended to the write buffer. |
391 |
|
392 |
Note that this is a function, and all types registered this way will be |
393 |
global, so try to use unique names. |
394 |
|
395 |
=cut |
396 |
|
397 |
############################################################################# |
398 |
|
399 |
=back |
400 |
|
401 |
=head2 READ QUEUE |
402 |
|
403 |
AnyEvent::Handle manages two queues per handle, one for writing and one |
404 |
for reading. |
405 |
|
406 |
The read queue is more complex than the write queue. It can be used in two |
407 |
ways, the "simple" way, using only C<on_read> and the "complex" way, using |
408 |
a queue. |
409 |
|
410 |
In the simple case, you just install an C<on_read> callback and whenever |
411 |
new data arrives, it will be called. You can then remove some data (if |
412 |
enough is there) from the read buffer (C<< $handle->rbuf >>) if you want |
413 |
or not. |
414 |
|
415 |
In the more complex case, you want to queue multiple callbacks. In this |
416 |
case, AnyEvent::Handle will call the first queued callback each time new |
417 |
data arrives and removes it when it has done its job (see C<push_read>, |
418 |
below). |
419 |
|
420 |
This way you can, for example, push three line-reads, followed by reading |
421 |
a chunk of data, and AnyEvent::Handle will execute them in order. |
422 |
|
423 |
Example 1: EPP protocol parser. EPP sends 4 byte length info, followed by |
424 |
the specified number of bytes which give an XML datagram. |
425 |
|
426 |
# in the default state, expect some header bytes |
427 |
$handle->on_read (sub { |
428 |
# some data is here, now queue the length-header-read (4 octets) |
429 |
shift->unshift_read_chunk (4, sub { |
430 |
# header arrived, decode |
431 |
my $len = unpack "N", $_[1]; |
432 |
|
433 |
# now read the payload |
434 |
shift->unshift_read_chunk ($len, sub { |
435 |
my $xml = $_[1]; |
436 |
# handle xml |
437 |
}); |
438 |
}); |
439 |
}); |
440 |
|
441 |
Example 2: Implement a client for a protocol that replies either with |
442 |
"OK" and another line or "ERROR" for one request, and 64 bytes for the |
443 |
second request. Due tot he availability of a full queue, we can just |
444 |
pipeline sending both requests and manipulate the queue as necessary in |
445 |
the callbacks: |
446 |
|
447 |
# request one |
448 |
$handle->push_write ("request 1\015\012"); |
449 |
|
450 |
# we expect "ERROR" or "OK" as response, so push a line read |
451 |
$handle->push_read_line (sub { |
452 |
# if we got an "OK", we have to _prepend_ another line, |
453 |
# so it will be read before the second request reads its 64 bytes |
454 |
# which are already in the queue when this callback is called |
455 |
# we don't do this in case we got an error |
456 |
if ($_[1] eq "OK") { |
457 |
$_[0]->unshift_read_line (sub { |
458 |
my $response = $_[1]; |
459 |
... |
460 |
}); |
461 |
} |
462 |
}); |
463 |
|
464 |
# request two |
465 |
$handle->push_write ("request 2\015\012"); |
466 |
|
467 |
# simply read 64 bytes, always |
468 |
$handle->push_read_chunk (64, sub { |
469 |
my $response = $_[1]; |
470 |
... |
471 |
}); |
472 |
|
473 |
=over 4 |
474 |
|
475 |
=cut |
476 |
|
477 |
sub _drain_rbuf { |
478 |
my ($self) = @_; |
479 |
|
480 |
if ( |
481 |
defined $self->{rbuf_max} |
482 |
&& $self->{rbuf_max} < length $self->{rbuf} |
483 |
) { |
484 |
$! = &Errno::ENOSPC; |
485 |
$self->error; |
486 |
} |
487 |
|
488 |
return if $self->{in_drain}; |
489 |
local $self->{in_drain} = 1; |
490 |
|
491 |
while (my $len = length $self->{rbuf}) { |
492 |
no strict 'refs'; |
493 |
if (my $cb = shift @{ $self->{_queue} }) { |
494 |
unless ($cb->($self)) { |
495 |
if ($self->{_eof}) { |
496 |
# no progress can be made (not enough data and no data forthcoming) |
497 |
$! = &Errno::EPIPE; |
498 |
$self->error; |
499 |
} |
500 |
|
501 |
unshift @{ $self->{_queue} }, $cb; |
502 |
return; |
503 |
} |
504 |
} elsif ($self->{on_read}) { |
505 |
$self->{on_read}($self); |
506 |
|
507 |
if ( |
508 |
$self->{_eof} # if no further data will arrive |
509 |
&& $len == length $self->{rbuf} # and no data has been consumed |
510 |
&& !@{ $self->{_queue} } # and the queue is still empty |
511 |
&& $self->{on_read} # and we still want to read data |
512 |
) { |
513 |
# then no progress can be made |
514 |
$! = &Errno::EPIPE; |
515 |
$self->error; |
516 |
} |
517 |
} else { |
518 |
# read side becomes idle |
519 |
delete $self->{_rw}; |
520 |
return; |
521 |
} |
522 |
} |
523 |
|
524 |
if ($self->{_eof}) { |
525 |
$self->_shutdown; |
526 |
$self->{on_eof}($self) |
527 |
if $self->{on_eof}; |
528 |
} |
529 |
} |
530 |
|
531 |
=item $handle->on_read ($cb) |
532 |
|
533 |
This replaces the currently set C<on_read> callback, or clears it (when |
534 |
the new callback is C<undef>). See the description of C<on_read> in the |
535 |
constructor. |
536 |
|
537 |
=cut |
538 |
|
539 |
sub on_read { |
540 |
my ($self, $cb) = @_; |
541 |
|
542 |
$self->{on_read} = $cb; |
543 |
} |
544 |
|
545 |
=item $handle->rbuf |
546 |
|
547 |
Returns the read buffer (as a modifiable lvalue). |
548 |
|
549 |
You can access the read buffer directly as the C<< ->{rbuf} >> member, if |
550 |
you want. |
551 |
|
552 |
NOTE: The read buffer should only be used or modified if the C<on_read>, |
553 |
C<push_read> or C<unshift_read> methods are used. The other read methods |
554 |
automatically manage the read buffer. |
555 |
|
556 |
=cut |
557 |
|
558 |
sub rbuf : lvalue { |
559 |
$_[0]{rbuf} |
560 |
} |
561 |
|
562 |
=item $handle->push_read ($cb) |
563 |
|
564 |
=item $handle->unshift_read ($cb) |
565 |
|
566 |
Append the given callback to the end of the queue (C<push_read>) or |
567 |
prepend it (C<unshift_read>). |
568 |
|
569 |
The callback is called each time some additional read data arrives. |
570 |
|
571 |
It must check whether enough data is in the read buffer already. |
572 |
|
573 |
If not enough data is available, it must return the empty list or a false |
574 |
value, in which case it will be called repeatedly until enough data is |
575 |
available (or an error condition is detected). |
576 |
|
577 |
If enough data was available, then the callback must remove all data it is |
578 |
interested in (which can be none at all) and return a true value. After returning |
579 |
true, it will be removed from the queue. |
580 |
|
581 |
=cut |
582 |
|
583 |
our %RH; |
584 |
|
585 |
sub register_read_type($$) { |
586 |
$RH{$_[0]} = $_[1]; |
587 |
} |
588 |
|
589 |
sub push_read { |
590 |
my $self = shift; |
591 |
my $cb = pop; |
592 |
|
593 |
if (@_) { |
594 |
my $type = shift; |
595 |
|
596 |
$cb = ($RH{$type} or Carp::croak "unsupported type passed to AnyEvent::Handle::push_read") |
597 |
->($self, $cb, @_); |
598 |
} |
599 |
|
600 |
push @{ $self->{_queue} }, $cb; |
601 |
$self->_drain_rbuf; |
602 |
} |
603 |
|
604 |
sub unshift_read { |
605 |
my $self = shift; |
606 |
my $cb = pop; |
607 |
|
608 |
if (@_) { |
609 |
my $type = shift; |
610 |
|
611 |
$cb = ($RH{$type} or Carp::croak "unsupported type passed to AnyEvent::Handle::unshift_read") |
612 |
->($self, $cb, @_); |
613 |
} |
614 |
|
615 |
|
616 |
unshift @{ $self->{_queue} }, $cb; |
617 |
$self->_drain_rbuf; |
618 |
} |
619 |
|
620 |
=item $handle->push_read (type => @args, $cb) |
621 |
|
622 |
=item $handle->unshift_read (type => @args, $cb) |
623 |
|
624 |
Instead of providing a callback that parses the data itself you can chose |
625 |
between a number of predefined parsing formats, for chunks of data, lines |
626 |
etc. |
627 |
|
628 |
Predefined types are (if you have ideas for additional types, feel free to |
629 |
drop by and tell us): |
630 |
|
631 |
=over 4 |
632 |
|
633 |
=item chunk => $octets, $cb->($self, $data) |
634 |
|
635 |
Invoke the callback only once C<$octets> bytes have been read. Pass the |
636 |
data read to the callback. The callback will never be called with less |
637 |
data. |
638 |
|
639 |
Example: read 2 bytes. |
640 |
|
641 |
$handle->push_read (chunk => 2, sub { |
642 |
warn "yay ", unpack "H*", $_[1]; |
643 |
}); |
644 |
|
645 |
=cut |
646 |
|
647 |
register_read_type chunk => sub { |
648 |
my ($self, $cb, $len) = @_; |
649 |
|
650 |
sub { |
651 |
$len <= length $_[0]{rbuf} or return; |
652 |
$cb->($_[0], substr $_[0]{rbuf}, 0, $len, ""); |
653 |
1 |
654 |
} |
655 |
}; |
656 |
|
657 |
# compatibility with older API |
658 |
sub push_read_chunk { |
659 |
$_[0]->push_read (chunk => $_[1], $_[2]); |
660 |
} |
661 |
|
662 |
sub unshift_read_chunk { |
663 |
$_[0]->unshift_read (chunk => $_[1], $_[2]); |
664 |
} |
665 |
|
666 |
=item line => [$eol, ]$cb->($self, $line, $eol) |
667 |
|
668 |
The callback will be called only once a full line (including the end of |
669 |
line marker, C<$eol>) has been read. This line (excluding the end of line |
670 |
marker) will be passed to the callback as second argument (C<$line>), and |
671 |
the end of line marker as the third argument (C<$eol>). |
672 |
|
673 |
The end of line marker, C<$eol>, can be either a string, in which case it |
674 |
will be interpreted as a fixed record end marker, or it can be a regex |
675 |
object (e.g. created by C<qr>), in which case it is interpreted as a |
676 |
regular expression. |
677 |
|
678 |
The end of line marker argument C<$eol> is optional, if it is missing (NOT |
679 |
undef), then C<qr|\015?\012|> is used (which is good for most internet |
680 |
protocols). |
681 |
|
682 |
Partial lines at the end of the stream will never be returned, as they are |
683 |
not marked by the end of line marker. |
684 |
|
685 |
=cut |
686 |
|
687 |
register_read_type line => sub { |
688 |
my ($self, $cb, $eol) = @_; |
689 |
|
690 |
$eol = qr|(\015?\012)| if @_ < 3; |
691 |
$eol = quotemeta $eol unless ref $eol; |
692 |
$eol = qr|^(.*?)($eol)|s; |
693 |
|
694 |
sub { |
695 |
$_[0]{rbuf} =~ s/$eol// or return; |
696 |
|
697 |
$cb->($_[0], $1, $2); |
698 |
1 |
699 |
} |
700 |
}; |
701 |
|
702 |
# compatibility with older API |
703 |
sub push_read_line { |
704 |
my $self = shift; |
705 |
$self->push_read (line => @_); |
706 |
} |
707 |
|
708 |
sub unshift_read_line { |
709 |
my $self = shift; |
710 |
$self->unshift_read (line => @_); |
711 |
} |
712 |
|
713 |
=item netstring => $cb->($string) |
714 |
|
715 |
A netstring (http://cr.yp.to/proto/netstrings.txt, this is not an endorsement). |
716 |
|
717 |
Throws an error with C<$!> set to EBADMSG on format violations. |
718 |
|
719 |
=cut |
720 |
|
721 |
register_read_type netstring => sub { |
722 |
my ($self, $cb) = @_; |
723 |
|
724 |
sub { |
725 |
unless ($_[0]{rbuf} =~ s/^(0|[1-9][0-9]*)://) { |
726 |
if ($_[0]{rbuf} =~ /[^0-9]/) { |
727 |
$! = &Errno::EBADMSG; |
728 |
$self->error; |
729 |
} |
730 |
return; |
731 |
} |
732 |
|
733 |
my $len = $1; |
734 |
|
735 |
$self->unshift_read (chunk => $len, sub { |
736 |
my $string = $_[1]; |
737 |
$_[0]->unshift_read (chunk => 1, sub { |
738 |
if ($_[1] eq ",") { |
739 |
$cb->($_[0], $string); |
740 |
} else { |
741 |
$! = &Errno::EBADMSG; |
742 |
$self->error; |
743 |
} |
744 |
}); |
745 |
}); |
746 |
|
747 |
1 |
748 |
} |
749 |
}; |
750 |
|
751 |
=item regex => $accept[, $reject[, $skip], $cb->($data) |
752 |
|
753 |
Makes a regex match against the regex object C<$accept> and returns |
754 |
everything up to and including the match. |
755 |
|
756 |
Example: read a single line terminated by '\n'. |
757 |
|
758 |
$handle->push_read (regex => qr<\n>, sub { ... }); |
759 |
|
760 |
If C<$reject> is given and not undef, then it determines when the data is |
761 |
to be rejected: it is matched against the data when the C<$accept> regex |
762 |
does not match and generates an C<EBADMSG> error when it matches. This is |
763 |
useful to quickly reject wrong data (to avoid waiting for a timeout or a |
764 |
receive buffer overflow). |
765 |
|
766 |
Example: expect a single decimal number followed by whitespace, reject |
767 |
anything else (not the use of an anchor). |
768 |
|
769 |
$handle->push_read (regex => qr<^[0-9]+\s>, qr<[^0-9]>, sub { ... }); |
770 |
|
771 |
If C<$skip> is given and not C<undef>, then it will be matched against |
772 |
the receive buffer when neither C<$accept> nor C<$reject> match, |
773 |
and everything preceding and including the match will be accepted |
774 |
unconditionally. This is useful to skip large amounts of data that you |
775 |
know cannot be matched, so that the C<$accept> or C<$reject> regex do not |
776 |
have to start matching from the beginning. This is purely an optimisation |
777 |
and is usually worth only when you expect more than a few kilobytes. |
778 |
|
779 |
Example: expect a http header, which ends at C<\015\012\015\012>. Since we |
780 |
expect the header to be very large (it isn't in practise, but...), we use |
781 |
a skip regex to skip initial portions. The skip regex is tricky in that |
782 |
it only accepts something not ending in either \015 or \012, as these are |
783 |
required for the accept regex. |
784 |
|
785 |
$handle->push_read (regex => |
786 |
qr<\015\012\015\012>, |
787 |
undef, # no reject |
788 |
qr<^.*[^\015\012]>, |
789 |
sub { ... }); |
790 |
|
791 |
=cut |
792 |
|
793 |
register_read_type regex => sub { |
794 |
my ($self, $cb, $accept, $reject, $skip) = @_; |
795 |
|
796 |
my $data; |
797 |
my $rbuf = \$self->{rbuf}; |
798 |
|
799 |
sub { |
800 |
# accept |
801 |
if ($$rbuf =~ $accept) { |
802 |
$data .= substr $$rbuf, 0, $+[0], ""; |
803 |
$cb->($self, $data); |
804 |
return 1; |
805 |
} |
806 |
|
807 |
# reject |
808 |
if ($reject && $$rbuf =~ $reject) { |
809 |
$! = &Errno::EBADMSG; |
810 |
$self->error; |
811 |
} |
812 |
|
813 |
# skip |
814 |
if ($skip && $$rbuf =~ $skip) { |
815 |
$data .= substr $$rbuf, 0, $+[0], ""; |
816 |
} |
817 |
|
818 |
() |
819 |
} |
820 |
}; |
821 |
|
822 |
=back |
823 |
|
824 |
=item AnyEvent::Handle::register_read_type type => $coderef->($self, $cb, @args) |
825 |
|
826 |
This function (not method) lets you add your own types to C<push_read>. |
827 |
|
828 |
Whenever the given C<type> is used, C<push_read> will invoke the code |
829 |
reference with the handle object, the callback and the remaining |
830 |
arguments. |
831 |
|
832 |
The code reference is supposed to return a callback (usually a closure) |
833 |
that works as a plain read callback (see C<< ->push_read ($cb) >>). |
834 |
|
835 |
It should invoke the passed callback when it is done reading (remember to |
836 |
pass C<$self> as first argument as all other callbacks do that). |
837 |
|
838 |
Note that this is a function, and all types registered this way will be |
839 |
global, so try to use unique names. |
840 |
|
841 |
For examples, see the source of this module (F<perldoc -m AnyEvent::Handle>, |
842 |
search for C<register_read_type>)). |
843 |
|
844 |
=item $handle->stop_read |
845 |
|
846 |
=item $handle->start_read |
847 |
|
848 |
In rare cases you actually do not want to read anything from the |
849 |
socket. In this case you can call C<stop_read>. Neither C<on_read> no |
850 |
any queued callbacks will be executed then. To start reading again, call |
851 |
C<start_read>. |
852 |
|
853 |
=cut |
854 |
|
855 |
sub stop_read { |
856 |
my ($self) = @_; |
857 |
|
858 |
delete $self->{_rw}; |
859 |
} |
860 |
|
861 |
sub start_read { |
862 |
my ($self) = @_; |
863 |
|
864 |
unless ($self->{_rw} || $self->{_eof}) { |
865 |
Scalar::Util::weaken $self; |
866 |
|
867 |
$self->{_rw} = AnyEvent->io (fh => $self->{fh}, poll => "r", cb => sub { |
868 |
my $rbuf = $self->{filter_r} ? \my $buf : \$self->{rbuf}; |
869 |
my $len = sysread $self->{fh}, $$rbuf, $self->{read_size} || 8192, length $$rbuf; |
870 |
|
871 |
if ($len > 0) { |
872 |
$self->{filter_r} |
873 |
? $self->{filter_r}->($self, $rbuf) |
874 |
: $self->_drain_rbuf; |
875 |
|
876 |
} elsif (defined $len) { |
877 |
delete $self->{_rw}; |
878 |
$self->{_eof} = 1; |
879 |
$self->_drain_rbuf; |
880 |
|
881 |
} elsif ($! != EAGAIN && $! != EINTR && $! != &AnyEvent::Util::WSAWOULDBLOCK) { |
882 |
return $self->error; |
883 |
} |
884 |
}); |
885 |
} |
886 |
} |
887 |
|
888 |
sub _dotls { |
889 |
my ($self) = @_; |
890 |
|
891 |
if (length $self->{_tls_wbuf}) { |
892 |
while ((my $len = Net::SSLeay::write ($self->{tls}, $self->{_tls_wbuf})) > 0) { |
893 |
substr $self->{_tls_wbuf}, 0, $len, ""; |
894 |
} |
895 |
} |
896 |
|
897 |
if (defined (my $buf = Net::SSLeay::BIO_read ($self->{_wbio}))) { |
898 |
$self->{wbuf} .= $buf; |
899 |
$self->_drain_wbuf; |
900 |
} |
901 |
|
902 |
while (defined (my $buf = Net::SSLeay::read ($self->{tls}))) { |
903 |
$self->{rbuf} .= $buf; |
904 |
$self->_drain_rbuf; |
905 |
} |
906 |
|
907 |
my $err = Net::SSLeay::get_error ($self->{tls}, -1); |
908 |
|
909 |
if ($err!= Net::SSLeay::ERROR_WANT_READ ()) { |
910 |
if ($err == Net::SSLeay::ERROR_SYSCALL ()) { |
911 |
$self->error; |
912 |
} elsif ($err == Net::SSLeay::ERROR_SSL ()) { |
913 |
$! = &Errno::EIO; |
914 |
$self->error; |
915 |
} |
916 |
|
917 |
# all others are fine for our purposes |
918 |
} |
919 |
} |
920 |
|
921 |
=item $handle->starttls ($tls[, $tls_ctx]) |
922 |
|
923 |
Instead of starting TLS negotiation immediately when the AnyEvent::Handle |
924 |
object is created, you can also do that at a later time by calling |
925 |
C<starttls>. |
926 |
|
927 |
The first argument is the same as the C<tls> constructor argument (either |
928 |
C<"connect">, C<"accept"> or an existing Net::SSLeay object). |
929 |
|
930 |
The second argument is the optional C<Net::SSLeay::CTX> object that is |
931 |
used when AnyEvent::Handle has to create its own TLS connection object. |
932 |
|
933 |
The TLS connection object will end up in C<< $handle->{tls} >> after this |
934 |
call and can be used or changed to your liking. Note that the handshake |
935 |
might have already started when this function returns. |
936 |
|
937 |
=cut |
938 |
|
939 |
# TODO: maybe document... |
940 |
sub starttls { |
941 |
my ($self, $ssl, $ctx) = @_; |
942 |
|
943 |
$self->stoptls; |
944 |
|
945 |
if ($ssl eq "accept") { |
946 |
$ssl = Net::SSLeay::new ($ctx || TLS_CTX ()); |
947 |
Net::SSLeay::set_accept_state ($ssl); |
948 |
} elsif ($ssl eq "connect") { |
949 |
$ssl = Net::SSLeay::new ($ctx || TLS_CTX ()); |
950 |
Net::SSLeay::set_connect_state ($ssl); |
951 |
} |
952 |
|
953 |
$self->{tls} = $ssl; |
954 |
|
955 |
# basically, this is deep magic (because SSL_read should have the same issues) |
956 |
# but the openssl maintainers basically said: "trust us, it just works". |
957 |
# (unfortunately, we have to hardcode constants because the abysmally misdesigned |
958 |
# and mismaintained ssleay-module doesn't even offer them). |
959 |
# http://www.mail-archive.com/openssl-dev@openssl.org/msg22420.html |
960 |
Net::SSLeay::CTX_set_mode ($self->{tls}, |
961 |
(eval { local $SIG{__DIE__}; Net::SSLeay::MODE_ENABLE_PARTIAL_WRITE () } || 1) |
962 |
| (eval { local $SIG{__DIE__}; Net::SSLeay::MODE_ACCEPT_MOVING_WRITE_BUFFER () } || 2)); |
963 |
|
964 |
$self->{_rbio} = Net::SSLeay::BIO_new (Net::SSLeay::BIO_s_mem ()); |
965 |
$self->{_wbio} = Net::SSLeay::BIO_new (Net::SSLeay::BIO_s_mem ()); |
966 |
|
967 |
Net::SSLeay::set_bio ($ssl, $self->{_rbio}, $self->{_wbio}); |
968 |
|
969 |
$self->{filter_w} = sub { |
970 |
$_[0]{_tls_wbuf} .= ${$_[1]}; |
971 |
&_dotls; |
972 |
}; |
973 |
$self->{filter_r} = sub { |
974 |
Net::SSLeay::BIO_write ($_[0]{_rbio}, ${$_[1]}); |
975 |
&_dotls; |
976 |
}; |
977 |
} |
978 |
|
979 |
=item $handle->stoptls |
980 |
|
981 |
Destroys the SSL connection, if any. Partial read or write data will be |
982 |
lost. |
983 |
|
984 |
=cut |
985 |
|
986 |
sub stoptls { |
987 |
my ($self) = @_; |
988 |
|
989 |
Net::SSLeay::free (delete $self->{tls}) if $self->{tls}; |
990 |
|
991 |
delete $self->{_rbio}; |
992 |
delete $self->{_wbio}; |
993 |
delete $self->{_tls_wbuf}; |
994 |
delete $self->{filter_r}; |
995 |
delete $self->{filter_w}; |
996 |
} |
997 |
|
998 |
sub DESTROY { |
999 |
my $self = shift; |
1000 |
|
1001 |
$self->stoptls; |
1002 |
} |
1003 |
|
1004 |
=item AnyEvent::Handle::TLS_CTX |
1005 |
|
1006 |
This function creates and returns the Net::SSLeay::CTX object used by |
1007 |
default for TLS mode. |
1008 |
|
1009 |
The context is created like this: |
1010 |
|
1011 |
Net::SSLeay::load_error_strings; |
1012 |
Net::SSLeay::SSLeay_add_ssl_algorithms; |
1013 |
Net::SSLeay::randomize; |
1014 |
|
1015 |
my $CTX = Net::SSLeay::CTX_new; |
1016 |
|
1017 |
Net::SSLeay::CTX_set_options $CTX, Net::SSLeay::OP_ALL |
1018 |
|
1019 |
=cut |
1020 |
|
1021 |
our $TLS_CTX; |
1022 |
|
1023 |
sub TLS_CTX() { |
1024 |
$TLS_CTX || do { |
1025 |
require Net::SSLeay; |
1026 |
|
1027 |
Net::SSLeay::load_error_strings (); |
1028 |
Net::SSLeay::SSLeay_add_ssl_algorithms (); |
1029 |
Net::SSLeay::randomize (); |
1030 |
|
1031 |
$TLS_CTX = Net::SSLeay::CTX_new (); |
1032 |
|
1033 |
Net::SSLeay::CTX_set_options ($TLS_CTX, Net::SSLeay::OP_ALL ()); |
1034 |
|
1035 |
$TLS_CTX |
1036 |
} |
1037 |
} |
1038 |
|
1039 |
=back |
1040 |
|
1041 |
=head1 SUBCLASSING AnyEvent::Handle |
1042 |
|
1043 |
In many cases, you might want to subclass AnyEvent::Handle. |
1044 |
|
1045 |
To make this easier, a given version of AnyEvent::Handle uses these |
1046 |
conventions: |
1047 |
|
1048 |
=over 4 |
1049 |
|
1050 |
=item * all constructor arguments become object members. |
1051 |
|
1052 |
At least initially, when you pass a C<tls>-argument to the constructor it |
1053 |
will end up in C<< $handle->{tls} >>. Those members might be changes or |
1054 |
mutated later on (for example C<tls> will hold the TLS connection object). |
1055 |
|
1056 |
=item * other object member names are prefixed with an C<_>. |
1057 |
|
1058 |
All object members not explicitly documented (internal use) are prefixed |
1059 |
with an underscore character, so the remaining non-C<_>-namespace is free |
1060 |
for use for subclasses. |
1061 |
|
1062 |
=item * all members not documented here and not prefixed with an underscore |
1063 |
are free to use in subclasses. |
1064 |
|
1065 |
Of course, new versions of AnyEvent::Handle may introduce more "public" |
1066 |
member variables, but thats just life, at least it is documented. |
1067 |
|
1068 |
=back |
1069 |
|
1070 |
=head1 AUTHOR |
1071 |
|
1072 |
Robin Redeker C<< <elmex at ta-sa.org> >>, Marc Lehmann <schmorp@schmorp.de>. |
1073 |
|
1074 |
=cut |
1075 |
|
1076 |
1; # End of AnyEvent::Handle |