1 |
package AnyEvent::Handle; |
2 |
|
3 |
no warnings; |
4 |
use strict; |
5 |
|
6 |
use AnyEvent (); |
7 |
use AnyEvent::Util (); |
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 filehandles via AnyEvent |
16 |
|
17 |
This module is experimental. |
18 |
|
19 |
=cut |
20 |
|
21 |
our $VERSION = '0.04'; |
22 |
|
23 |
=head1 SYNOPSIS |
24 |
|
25 |
use AnyEvent; |
26 |
use AnyEvent::Handle; |
27 |
|
28 |
my $cv = AnyEvent->condvar; |
29 |
|
30 |
my $ae_fh = AnyEvent::Handle->new (fh => \*STDIN); |
31 |
|
32 |
#TODO |
33 |
|
34 |
# or use the constructor to pass the callback: |
35 |
|
36 |
my $ae_fh2 = |
37 |
AnyEvent::Handle->new ( |
38 |
fh => \*STDIN, |
39 |
on_eof => sub { |
40 |
$cv->broadcast; |
41 |
}, |
42 |
#TODO |
43 |
); |
44 |
|
45 |
$cv->wait; |
46 |
|
47 |
=head1 DESCRIPTION |
48 |
|
49 |
This module is a helper module to make it easier to do event-based I/O on |
50 |
filehandles. For utility functions for doing non-blocking connects and accepts |
51 |
on sockets see L<AnyEvent::Util>. |
52 |
|
53 |
In the following, when the documentation refers to of "bytes" then this |
54 |
means characters. As sysread and syswrite are used for all I/O, their |
55 |
treatment of characters applies to this module as well. |
56 |
|
57 |
All callbacks will be invoked with the handle object as their first |
58 |
argument. |
59 |
|
60 |
=head1 METHODS |
61 |
|
62 |
=over 4 |
63 |
|
64 |
=item B<new (%args)> |
65 |
|
66 |
The constructor supports these arguments (all as key => value pairs). |
67 |
|
68 |
=over 4 |
69 |
|
70 |
=item fh => $filehandle [MANDATORY] |
71 |
|
72 |
The filehandle this L<AnyEvent::Handle> object will operate on. |
73 |
|
74 |
NOTE: The filehandle will be set to non-blocking (using |
75 |
AnyEvent::Util::fh_nonblocking). |
76 |
|
77 |
=item on_eof => $cb->($self) |
78 |
|
79 |
Set the callback to be called on EOF. |
80 |
|
81 |
While not mandatory, it is highly recommended to set an eof callback, |
82 |
otherwise you might end up with a closed socket while you are still |
83 |
waiting for data. |
84 |
|
85 |
=item on_error => $cb->($self) |
86 |
|
87 |
This is the fatal error callback, that is called when, well, a fatal error |
88 |
ocurs, such as not being able to resolve the hostname, failure to connect |
89 |
or a read error. |
90 |
|
91 |
The object will not be in a usable state when this callback has been |
92 |
called. |
93 |
|
94 |
On callback entrance, the value of C<$!> contains the operating system |
95 |
error (or C<ENOSPC> or C<EPIPE>). |
96 |
|
97 |
While not mandatory, it is I<highly> recommended to set this callback, as |
98 |
you will not be notified of errors otherwise. The default simply calls |
99 |
die. |
100 |
|
101 |
=item on_read => $cb->($self) |
102 |
|
103 |
This sets the default read callback, which is called when data arrives |
104 |
and no read request is in the queue. |
105 |
|
106 |
To access (and remove data from) the read buffer, use the C<< ->rbuf >> |
107 |
method or acces sthe C<$self->{rbuf}> member directly. |
108 |
|
109 |
When an EOF condition is detected then AnyEvent::Handle will first try to |
110 |
feed all the remaining data to the queued callbacks and C<on_read> before |
111 |
calling the C<on_eof> callback. If no progress can be made, then a fatal |
112 |
error will be raised (with C<$!> set to C<EPIPE>). |
113 |
|
114 |
=item on_drain => $cb->() |
115 |
|
116 |
This sets the callback that is called when the write buffer becomes empty |
117 |
(or when the callback is set and the buffer is empty already). |
118 |
|
119 |
To append to the write buffer, use the C<< ->push_write >> method. |
120 |
|
121 |
=item rbuf_max => <bytes> |
122 |
|
123 |
If defined, then a fatal error will be raised (with C<$!> set to C<ENOSPC>) |
124 |
when the read buffer ever (strictly) exceeds this size. This is useful to |
125 |
avoid denial-of-service attacks. |
126 |
|
127 |
For example, a server accepting connections from untrusted sources should |
128 |
be configured to accept only so-and-so much data that it cannot act on |
129 |
(for example, when expecting a line, an attacker could send an unlimited |
130 |
amount of data without a callback ever being called as long as the line |
131 |
isn't finished). |
132 |
|
133 |
=item read_size => <bytes> |
134 |
|
135 |
The default read block size (the amount of bytes this module will try to read |
136 |
on each [loop iteration). Default: C<4096>. |
137 |
|
138 |
=item low_water_mark => <bytes> |
139 |
|
140 |
Sets the amount of bytes (default: C<0>) that make up an "empty" write |
141 |
buffer: If the write reaches this size or gets even samller it is |
142 |
considered empty. |
143 |
|
144 |
=back |
145 |
|
146 |
=cut |
147 |
|
148 |
sub new { |
149 |
my $class = shift; |
150 |
|
151 |
my $self = bless { @_ }, $class; |
152 |
|
153 |
$self->{fh} or Carp::croak "mandatory argument fh is missing"; |
154 |
|
155 |
AnyEvent::Util::fh_nonblocking $self->{fh}, 1; |
156 |
|
157 |
$self->on_eof (delete $self->{on_eof} ) if $self->{on_eof}; |
158 |
$self->on_error (delete $self->{on_error}) if $self->{on_error}; |
159 |
$self->on_drain (delete $self->{on_drain}) if $self->{on_drain}; |
160 |
$self->on_read (delete $self->{on_read} ) if $self->{on_read}; |
161 |
|
162 |
$self->start_read; |
163 |
|
164 |
$self |
165 |
} |
166 |
|
167 |
sub _shutdown { |
168 |
my ($self) = @_; |
169 |
|
170 |
delete $self->{rw}; |
171 |
delete $self->{ww}; |
172 |
delete $self->{fh}; |
173 |
} |
174 |
|
175 |
sub error { |
176 |
my ($self) = @_; |
177 |
|
178 |
{ |
179 |
local $!; |
180 |
$self->_shutdown; |
181 |
} |
182 |
|
183 |
if ($self->{on_error}) { |
184 |
$self->{on_error}($self); |
185 |
} else { |
186 |
die "AnyEvent::Handle uncaught fatal error: $!"; |
187 |
} |
188 |
} |
189 |
|
190 |
=item $fh = $handle->fh |
191 |
|
192 |
This method returns the filehandle of the L<AnyEvent::Handle> object. |
193 |
|
194 |
=cut |
195 |
|
196 |
sub fh { $_[0]->{fh} } |
197 |
|
198 |
=item $handle->on_error ($cb) |
199 |
|
200 |
Replace the current C<on_error> callback (see the C<on_error> constructor argument). |
201 |
|
202 |
=cut |
203 |
|
204 |
sub on_error { |
205 |
$_[0]{on_error} = $_[1]; |
206 |
} |
207 |
|
208 |
=item $handle->on_eof ($cb) |
209 |
|
210 |
Replace the current C<on_eof> callback (see the C<on_eof> constructor argument). |
211 |
|
212 |
=cut |
213 |
|
214 |
sub on_eof { |
215 |
$_[0]{on_eof} = $_[1]; |
216 |
} |
217 |
|
218 |
############################################################################# |
219 |
|
220 |
=back |
221 |
|
222 |
=head2 WRITE QUEUE |
223 |
|
224 |
AnyEvent::Handle manages two queues per handle, one for writing and one |
225 |
for reading. |
226 |
|
227 |
The write queue is very simple: you can add data to its end, and |
228 |
AnyEvent::Handle will automatically try to get rid of it for you. |
229 |
|
230 |
When data could be writtena nd the write buffer is shorter then the low |
231 |
water mark, the C<on_drain> callback will be invoked. |
232 |
|
233 |
=over 4 |
234 |
|
235 |
=item $handle->on_drain ($cb) |
236 |
|
237 |
Sets the C<on_drain> callback or clears it (see the description of |
238 |
C<on_drain> in the constructor). |
239 |
|
240 |
=cut |
241 |
|
242 |
sub on_drain { |
243 |
my ($self, $cb) = @_; |
244 |
|
245 |
$self->{on_drain} = $cb; |
246 |
|
247 |
$cb->($self) |
248 |
if $cb && $self->{low_water_mark} >= length $self->{wbuf}; |
249 |
} |
250 |
|
251 |
=item $handle->push_write ($data) |
252 |
|
253 |
Queues the given scalar to be written. You can push as much data as you |
254 |
want (only limited by the available memory), as C<AnyEvent::Handle> |
255 |
buffers it independently of the kernel. |
256 |
|
257 |
=cut |
258 |
|
259 |
sub _drain_wbuf { |
260 |
my ($self) = @_; |
261 |
|
262 |
unless ($self->{ww}) { |
263 |
Scalar::Util::weaken $self; |
264 |
my $cb = sub { |
265 |
my $len = syswrite $self->{fh}, $self->{wbuf}; |
266 |
|
267 |
if ($len > 0) { |
268 |
substr $self->{wbuf}, 0, $len, ""; |
269 |
|
270 |
$self->{on_drain}($self) |
271 |
if $self->{low_water_mark} >= length $self->{wbuf} |
272 |
&& $self->{on_drain}; |
273 |
|
274 |
delete $self->{ww} unless length $self->{wbuf}; |
275 |
} elsif ($! != EAGAIN && $! != EINTR) { |
276 |
$self->error; |
277 |
} |
278 |
}; |
279 |
|
280 |
$self->{ww} = AnyEvent->io (fh => $self->{fh}, poll => "w", cb => $cb); |
281 |
|
282 |
$cb->($self); |
283 |
}; |
284 |
} |
285 |
|
286 |
sub push_write { |
287 |
my $self = shift; |
288 |
|
289 |
if ($self->{filter_w}) { |
290 |
$self->{filter_w}->(\$_[0]); |
291 |
} else { |
292 |
$self->{wbuf} .= $_[0]; |
293 |
$self->_drain_wbuf; |
294 |
} |
295 |
} |
296 |
|
297 |
############################################################################# |
298 |
|
299 |
=back |
300 |
|
301 |
=head2 READ QUEUE |
302 |
|
303 |
AnyEvent::Handle manages two queues per handle, one for writing and one |
304 |
for reading. |
305 |
|
306 |
The read queue is more complex than the write queue. It can be used in two |
307 |
ways, the "simple" way, using only C<on_read> and the "complex" way, using |
308 |
a queue. |
309 |
|
310 |
In the simple case, you just install an C<on_read> callback and whenever |
311 |
new data arrives, it will be called. You can then remove some data (if |
312 |
enough is there) from the read buffer (C<< $handle->rbuf >>) if you want |
313 |
or not. |
314 |
|
315 |
In the more complex case, you want to queue multiple callbacks. In this |
316 |
case, AnyEvent::Handle will call the first queued callback each time new |
317 |
data arrives and removes it when it has done its job (see C<push_read>, |
318 |
below). |
319 |
|
320 |
This way you can, for example, push three line-reads, followed by reading |
321 |
a chunk of data, and AnyEvent::Handle will execute them in order. |
322 |
|
323 |
Example 1: EPP protocol parser. EPP sends 4 byte length info, followed by |
324 |
the specified number of bytes which give an XML datagram. |
325 |
|
326 |
# in the default state, expect some header bytes |
327 |
$handle->on_read (sub { |
328 |
# some data is here, now queue the length-header-read (4 octets) |
329 |
shift->unshift_read_chunk (4, sub { |
330 |
# header arrived, decode |
331 |
my $len = unpack "N", $_[1]; |
332 |
|
333 |
# now read the payload |
334 |
shift->unshift_read_chunk ($len, sub { |
335 |
my $xml = $_[1]; |
336 |
# handle xml |
337 |
}); |
338 |
}); |
339 |
}); |
340 |
|
341 |
Example 2: Implement a client for a protocol that replies either with |
342 |
"OK" and another line or "ERROR" for one request, and 64 bytes for the |
343 |
second request. Due tot he availability of a full queue, we can just |
344 |
pipeline sending both requests and manipulate the queue as necessary in |
345 |
the callbacks: |
346 |
|
347 |
# request one |
348 |
$handle->push_write ("request 1\015\012"); |
349 |
|
350 |
# we expect "ERROR" or "OK" as response, so push a line read |
351 |
$handle->push_read_line (sub { |
352 |
# if we got an "OK", we have to _prepend_ another line, |
353 |
# so it will be read before the second request reads its 64 bytes |
354 |
# which are already in the queue when this callback is called |
355 |
# we don't do this in case we got an error |
356 |
if ($_[1] eq "OK") { |
357 |
$_[0]->unshift_read_line (sub { |
358 |
my $response = $_[1]; |
359 |
... |
360 |
}); |
361 |
} |
362 |
}); |
363 |
|
364 |
# request two |
365 |
$handle->push_write ("request 2\015\012"); |
366 |
|
367 |
# simply read 64 bytes, always |
368 |
$handle->push_read_chunk (64, sub { |
369 |
my $response = $_[1]; |
370 |
... |
371 |
}); |
372 |
|
373 |
=over 4 |
374 |
|
375 |
=cut |
376 |
|
377 |
sub _drain_rbuf { |
378 |
my ($self) = @_; |
379 |
|
380 |
if ( |
381 |
defined $self->{rbuf_max} |
382 |
&& $self->{rbuf_max} < length $self->{rbuf} |
383 |
) { |
384 |
$! = &Errno::ENOSPC; return $self->error; |
385 |
} |
386 |
|
387 |
return if $self->{in_drain}; |
388 |
local $self->{in_drain} = 1; |
389 |
|
390 |
while (my $len = length $self->{rbuf}) { |
391 |
no strict 'refs'; |
392 |
if (my $cb = shift @{ $self->{queue} }) { |
393 |
if (!$cb->($self)) { |
394 |
if ($self->{eof}) { |
395 |
# no progress can be made (not enough data and no data forthcoming) |
396 |
$! = &Errno::EPIPE; return $self->error; |
397 |
} |
398 |
|
399 |
unshift @{ $self->{queue} }, $cb; |
400 |
return; |
401 |
} |
402 |
} elsif ($self->{on_read}) { |
403 |
$self->{on_read}($self); |
404 |
|
405 |
if ( |
406 |
$self->{eof} # if no further data will arrive |
407 |
&& $len == length $self->{rbuf} # and no data has been consumed |
408 |
&& !@{ $self->{queue} } # and the queue is still empty |
409 |
&& $self->{on_read} # and we still want to read data |
410 |
) { |
411 |
# then no progress can be made |
412 |
$! = &Errno::EPIPE; return $self->error; |
413 |
} |
414 |
} else { |
415 |
# read side becomes idle |
416 |
delete $self->{rw}; |
417 |
return; |
418 |
} |
419 |
} |
420 |
|
421 |
if ($self->{eof}) { |
422 |
$self->_shutdown; |
423 |
$self->{on_eof}($self) |
424 |
if $self->{on_eof}; |
425 |
} |
426 |
} |
427 |
|
428 |
=item $handle->on_read ($cb) |
429 |
|
430 |
This replaces the currently set C<on_read> callback, or clears it (when |
431 |
the new callback is C<undef>). See the description of C<on_read> in the |
432 |
constructor. |
433 |
|
434 |
=cut |
435 |
|
436 |
sub on_read { |
437 |
my ($self, $cb) = @_; |
438 |
|
439 |
$self->{on_read} = $cb; |
440 |
} |
441 |
|
442 |
=item $handle->rbuf |
443 |
|
444 |
Returns the read buffer (as a modifiable lvalue). |
445 |
|
446 |
You can access the read buffer directly as the C<< ->{rbuf} >> member, if |
447 |
you want. |
448 |
|
449 |
NOTE: The read buffer should only be used or modified if the C<on_read>, |
450 |
C<push_read> or C<unshift_read> methods are used. The other read methods |
451 |
automatically manage the read buffer. |
452 |
|
453 |
=cut |
454 |
|
455 |
sub rbuf : lvalue { |
456 |
$_[0]{rbuf} |
457 |
} |
458 |
|
459 |
=item $handle->push_read ($cb) |
460 |
|
461 |
=item $handle->unshift_read ($cb) |
462 |
|
463 |
Append the given callback to the end of the queue (C<push_read>) or |
464 |
prepend it (C<unshift_read>). |
465 |
|
466 |
The callback is called each time some additional read data arrives. |
467 |
|
468 |
It must check wether enough data is in the read buffer already. |
469 |
|
470 |
If not enough data is available, it must return the empty list or a false |
471 |
value, in which case it will be called repeatedly until enough data is |
472 |
available (or an error condition is detected). |
473 |
|
474 |
If enough data was available, then the callback must remove all data it is |
475 |
interested in (which can be none at all) and return a true value. After returning |
476 |
true, it will be removed from the queue. |
477 |
|
478 |
=cut |
479 |
|
480 |
sub push_read { |
481 |
my ($self, $cb) = @_; |
482 |
|
483 |
push @{ $self->{queue} }, $cb; |
484 |
$self->_drain_rbuf; |
485 |
} |
486 |
|
487 |
sub unshift_read { |
488 |
my ($self, $cb) = @_; |
489 |
|
490 |
push @{ $self->{queue} }, $cb; |
491 |
$self->_drain_rbuf; |
492 |
} |
493 |
|
494 |
=item $handle->push_read_chunk ($len, $cb->($self, $data)) |
495 |
|
496 |
=item $handle->unshift_read_chunk ($len, $cb->($self, $data)) |
497 |
|
498 |
Append the given callback to the end of the queue (C<push_read_chunk>) or |
499 |
prepend it (C<unshift_read_chunk>). |
500 |
|
501 |
The callback will be called only once C<$len> bytes have been read, and |
502 |
these C<$len> bytes will be passed to the callback. |
503 |
|
504 |
=cut |
505 |
|
506 |
sub _read_chunk($$) { |
507 |
my ($self, $len, $cb) = @_; |
508 |
|
509 |
sub { |
510 |
$len <= length $_[0]{rbuf} or return; |
511 |
$cb->($_[0], substr $_[0]{rbuf}, 0, $len, ""); |
512 |
1 |
513 |
} |
514 |
} |
515 |
|
516 |
sub push_read_chunk { |
517 |
$_[0]->push_read (&_read_chunk); |
518 |
} |
519 |
|
520 |
|
521 |
sub unshift_read_chunk { |
522 |
$_[0]->unshift_read (&_read_chunk); |
523 |
} |
524 |
|
525 |
=item $handle->push_read_line ([$eol, ]$cb->($self, $line, $eol)) |
526 |
|
527 |
=item $handle->unshift_read_line ([$eol, ]$cb->($self, $line, $eol)) |
528 |
|
529 |
Append the given callback to the end of the queue (C<push_read_line>) or |
530 |
prepend it (C<unshift_read_line>). |
531 |
|
532 |
The callback will be called only once a full line (including the end of |
533 |
line marker, C<$eol>) has been read. This line (excluding the end of line |
534 |
marker) will be passed to the callback as second argument (C<$line>), and |
535 |
the end of line marker as the third argument (C<$eol>). |
536 |
|
537 |
The end of line marker, C<$eol>, can be either a string, in which case it |
538 |
will be interpreted as a fixed record end marker, or it can be a regex |
539 |
object (e.g. created by C<qr>), in which case it is interpreted as a |
540 |
regular expression. |
541 |
|
542 |
The end of line marker argument C<$eol> is optional, if it is missing (NOT |
543 |
undef), then C<qr|\015?\012|> is used (which is good for most internet |
544 |
protocols). |
545 |
|
546 |
Partial lines at the end of the stream will never be returned, as they are |
547 |
not marked by the end of line marker. |
548 |
|
549 |
=cut |
550 |
|
551 |
sub _read_line($$) { |
552 |
my $self = shift; |
553 |
my $cb = pop; |
554 |
my $eol = @_ ? shift : qr|(\015?\012)|; |
555 |
my $pos; |
556 |
|
557 |
$eol = quotemeta $eol unless ref $eol; |
558 |
$eol = qr|^(.*?)($eol)|s; |
559 |
|
560 |
sub { |
561 |
$_[0]{rbuf} =~ s/$eol// or return; |
562 |
|
563 |
$cb->($_[0], $1, $2); |
564 |
1 |
565 |
} |
566 |
} |
567 |
|
568 |
sub push_read_line { |
569 |
$_[0]->push_read (&_read_line); |
570 |
} |
571 |
|
572 |
sub unshift_read_line { |
573 |
$_[0]->unshift_read (&_read_line); |
574 |
} |
575 |
|
576 |
=item $handle->stop_read |
577 |
|
578 |
=item $handle->start_read |
579 |
|
580 |
In rare cases you actually do not want to read anything form the |
581 |
socket. In this case you can call C<stop_read>. Neither C<on_read> no |
582 |
any queued callbacks will be executed then. To start readign again, call |
583 |
C<start_read>. |
584 |
|
585 |
=cut |
586 |
|
587 |
sub stop_read { |
588 |
my ($self) = @_; |
589 |
|
590 |
delete $self->{rw}; |
591 |
} |
592 |
|
593 |
sub start_read { |
594 |
my ($self) = @_; |
595 |
|
596 |
unless ($self->{rw} || $self->{eof}) { |
597 |
Scalar::Util::weaken $self; |
598 |
|
599 |
$self->{rw} = AnyEvent->io (fh => $self->{fh}, poll => "r", cb => sub { |
600 |
my $rbuf = $self->{filter_r} ? \my $buf : \$self->{rbuf}; |
601 |
my $len = sysread $self->{fh}, $$rbuf, $self->{read_size} || 8192, length $$rbuf; |
602 |
|
603 |
if ($len > 0) { |
604 |
$self->{filter_r} |
605 |
? $self->{filter_r}->($rbuf) |
606 |
: $self->_drain_rbuf; |
607 |
|
608 |
} elsif (defined $len) { |
609 |
delete $self->{rw}; |
610 |
$self->{eof} = 1; |
611 |
$self->_drain_rbuf; |
612 |
|
613 |
} elsif ($! != EAGAIN && $! != EINTR) { |
614 |
return $self->error; |
615 |
} |
616 |
}); |
617 |
} |
618 |
} |
619 |
|
620 |
=back |
621 |
|
622 |
=head1 AUTHOR |
623 |
|
624 |
Robin Redeker C<< <elmex at ta-sa.org> >>, Marc Lehmann <schmorp@schmorp.de>. |
625 |
|
626 |
=cut |
627 |
|
628 |
1; # End of AnyEvent::Handle |