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

Comparing EV/EV.pm (file contents):
Revision 1.18 by root, Thu Nov 1 08:10:03 2007 UTC vs.
Revision 1.58 by root, Wed Nov 28 19:22:16 2007 UTC

1=head1 NAME 1=head1 NAME
2 2
3EV - perl interface to libevent, monkey.org/~provos/libevent/ 3EV - perl interface to libev, a high performance full-featured event loop
4 4
5=head1 SYNOPSIS 5=head1 SYNOPSIS
6 6
7 use EV; 7 use EV;
8 8
9 # TIMER 9 # TIMERS
10 10
11 my $w = EV::timer 2, 0, sub { 11 my $w = EV::timer 2, 0, sub {
12 warn "is called after 2s"; 12 warn "is called after 2s";
13 }; 13 };
14 14
15 my $w = EV::timer 2, 1, sub { 15 my $w = EV::timer 2, 2, sub {
16 warn "is called roughly every 2s (repeat = 1)"; 16 warn "is called roughly every 2s (repeat = 2)";
17 }; 17 };
18 18
19 undef $w; # destroy event watcher again 19 undef $w; # destroy event watcher again
20 20
21 my $w = EV::timer_abs 0, 60, sub { 21 my $w = EV::periodic 0, 60, 0, sub {
22 warn "is called every minute, on the minute, exactly"; 22 warn "is called every minute, on the minute, exactly";
23 }; 23 };
24 24
25 # IO 25 # IO
26 26
27 my $w = EV::io \*STDIN, EV::READ | EV::PERSIST, sub { 27 my $w = EV::io *STDIN, EV::READ, sub {
28 my ($w, $revents) = @_; # all callbacks get the watcher object and event mask 28 my ($w, $revents) = @_; # all callbacks receive the watcher and event mask
29 if ($revents & EV::TIMEOUT) {
30 warn "nothing received on stdin for 10 seconds, retrying";
31 } else {
32 warn "stdin is readable, you entered: ", <STDIN>; 29 warn "stdin is readable, you entered: ", <STDIN>;
33 }
34 };
35 $w->timeout (10);
36
37 my $w = EV::timed_io \*STDIN, EV::READ, 30, sub {
38 my ($w, $revents) = @_;
39 if ($revents & EV::TIMEOUT) {
40 warn "nothing entered within 30 seconds, bye bye.\n";
41 $w->stop;
42 } else {
43 my $line = <STDIN>;
44 warn "you entered something, you again have 30 seconds.\n";
45 }
46 }; 30 };
47 31
48 # SIGNALS 32 # SIGNALS
49 33
50 my $w = EV::signal 'QUIT', sub { 34 my $w = EV::signal 'QUIT', sub {
51 warn "sigquit received\n"; 35 warn "sigquit received\n";
52 }; 36 };
53 37
38 # CHILD/PID STATUS CHANGES
39
54 my $w = EV::signal 3, sub { 40 my $w = EV::child 666, sub {
55 warn "sigquit received (this is GNU/Linux, right?)\n"; 41 my ($w, $revents) = @_;
42 my $status = $w->rstatus;
56 }; 43 };
57 44
58 # CHILD/PID STATUS CHANGES 45 # STAT CHANGES
59 46 my $w = EV::stat "/etc/passwd", 10, sub {
60 my $w = EV::child 666, sub {
61 my ($w, $revents, $status) = @_; 47 my ($w, $revents) = @_;
48 warn $w->path, " has changed somehow.\n";
62 }; 49 };
63 50
64 # MAINLOOP 51 # MAINLOOP
65 EV::dispatch; # loop as long as watchers are active 52 EV::loop; # loop until EV::unloop is called or all watchers stop
66 EV::loop; # the same thing
67 EV::loop EV::LOOP_ONESHOT; # block until some events could be handles 53 EV::loop EV::LOOP_ONESHOT; # block until at least one event could be handled
68 EV::loop EV::LOOP_NONBLOCK; # check and handle some events, but do not wait 54 EV::loop EV::LOOP_NONBLOCK; # try to handle same events, but do not block
69 55
70=head1 DESCRIPTION 56=head1 DESCRIPTION
71 57
72This module provides an interface to libev 58This module provides an interface to libev
73(L<http://software.schmorp.de/pkg/libev.html>). You probably should 59(L<http://software.schmorp.de/pkg/libev.html>). While the documentation
74acquaint yourself with its documentation and source code to be able to use 60below is comprehensive, one might also consult the documentation of libev
75this module fully. 61itself (L<http://cvs.schmorp.de/libev/ev.html>) for more subtle details on
62watcher semantics or some discussion on the available backends, or how to
63force a specific backend with C<LIBEV_FLAGS>.
76 64
77=cut 65=cut
78 66
79package EV; 67package EV;
80 68
81use strict; 69use strict;
82 70
83BEGIN { 71BEGIN {
84 our $VERSION = '0.03'; 72 our $VERSION = '1.5';
85 use XSLoader; 73 use XSLoader;
86 XSLoader::load "EV", $VERSION; 74 XSLoader::load "EV", $VERSION;
87} 75}
88 76
89@EV::Io::ISA = 77@EV::IO::ISA =
90@EV::Timer::ISA = 78@EV::Timer::ISA =
91@EV::Periodic::ISA = 79@EV::Periodic::ISA =
92@EV::Signal::ISA = 80@EV::Signal::ISA =
81@EV::Child::ISA =
82@EV::Stat::ISA =
93@EV::Idle::ISA = 83@EV::Idle::ISA =
94@EV::Prepare::ISA = 84@EV::Prepare::ISA =
95@EV::Check::ISA = 85@EV::Check::ISA =
96@EV::Child::ISA = "EV::Watcher"; 86@EV::Embed::ISA =
87@EV::Fork::ISA =
88 "EV::Watcher";
97 89
98=head1 BASIC INTERFACE 90=head1 BASIC INTERFACE
99 91
100=over 4 92=over 4
101
102=item $EV::NPRI
103
104How many priority levels are available.
105 93
106=item $EV::DIED 94=item $EV::DIED
107 95
108Must contain a reference to a function that is called when a callback 96Must contain a reference to a function that is called when a callback
109throws an exception (with $@ containing thr error). The default prints an 97throws an exception (with $@ containing thr error). The default prints an
110informative message and continues. 98informative message and continues.
111 99
112If this callback throws an exception it will be silently ignored. 100If this callback throws an exception it will be silently ignored.
113 101
102=item $time = EV::time
103
104Returns the current time in (fractional) seconds since the epoch.
105
114=item $time = EV::now 106=item $time = EV::now
115 107
116Returns the time in (fractional) seconds since the epoch. 108Returns the time the last event loop iteration has been started. This
117 109is the time that (relative) timers are based on, and refering to it is
118=item $version = EV::version 110usually faster then calling EV::time.
119 111
120=item $method = EV::method 112=item $method = EV::method
121 113
122Return version string and event polling method used. 114Returns an integer describing the backend used by libev (EV::METHOD_SELECT
115or EV::METHOD_EPOLL).
123 116
124=item EV::loop $flags # EV::LOOP_ONCE, EV::LOOP_ONESHOT 117=item EV::loop [$flags]
125 118
126=item EV::loopexit $after 119Begin checking for events and calling callbacks. It returns when a
120callback calls EV::unloop.
127 121
128Exit any active loop or dispatch after C<$after> seconds or immediately if 122The $flags argument can be one of the following:
129C<$after> is missing or zero.
130 123
131=item EV::dispatch 124 0 as above
125 EV::LOOP_ONESHOT block at most once (wait, but do not loop)
126 EV::LOOP_NONBLOCK do not block at all (fetch/handle events but do not wait)
132 127
133Same as C<EV::loop 0>. 128=item EV::unloop [$how]
134 129
135=item EV::event $callback 130When called with no arguments or an argument of EV::UNLOOP_ONE, makes the
131innermost call to EV::loop return.
136 132
137Creates a new event watcher waiting for nothing, calling the given callback. 133When called with an argument of EV::UNLOOP_ALL, all calls to EV::loop will return as
134fast as possible.
138 135
136=item EV::once $fh_or_undef, $events, $timeout, $cb->($revents)
137
138This function rolls together an I/O and a timer watcher for a single
139one-shot event without the need for managing a watcher object.
140
141If C<$fh_or_undef> is a filehandle or file descriptor, then C<$events>
142must be a bitset containing either C<EV::READ>, C<EV::WRITE> or C<EV::READ
143| EV::WRITE>, indicating the type of I/O event you want to wait for. If
144you do not want to wait for some I/O event, specify C<undef> for
145C<$fh_or_undef> and C<0> for C<$events>).
146
147If timeout is C<undef> or negative, then there will be no
148timeout. Otherwise a EV::timer with this value will be started.
149
150When an error occurs or either the timeout or I/O watcher triggers, then
151the callback will be called with the received event set (in general
152you can expect it to be a combination of C<EV:ERROR>, C<EV::READ>,
153C<EV::WRITE> and C<EV::TIMEOUT>).
154
155EV::once doesn't return anything: the watchers stay active till either
156of them triggers, then they will be stopped and freed, and the callback
157invoked.
158
159=back
160
161=head2 WATCHER OBJECTS
162
163A watcher is an object that gets created to record your interest in some
164event. For instance, if you want to wait for STDIN to become readable, you
165would create an EV::io watcher for that:
166
167 my $watcher = EV::io *STDIN, EV::READ, sub {
168 my ($watcher, $revents) = @_;
169 warn "yeah, STDIN should not be readable without blocking!\n"
170 };
171
172All watchers can be active (waiting for events) or inactive (paused). Only
173active watchers will have their callbacks invoked. All callbacks will be
174called with at least two arguments: the watcher and a bitmask of received
175events.
176
177Each watcher type has its associated bit in revents, so you can use the
178same callback for multiple watchers. The event mask is named after the
179type, i..e. EV::child sets EV::CHILD, EV::prepare sets EV::PREPARE,
180EV::periodic sets EV::PERIODIC and so on, with the exception of IO events
181(which can set both EV::READ and EV::WRITE bits), and EV::timer (which
182uses EV::TIMEOUT).
183
184In the rare case where one wants to create a watcher but not start it at
185the same time, each constructor has a variant with a trailing C<_ns> in
186its name, e.g. EV::io has a non-starting variant EV::io_ns and so on.
187
188Please note that a watcher will automatically be stopped when the watcher
189object is destroyed, so you I<need> to keep the watcher objects returned by
190the constructors.
191
192Also, all methods changing some aspect of a watcher (->set, ->priority,
193->fh and so on) automatically stop and start it again if it is active,
194which means pending events get lost.
195
196=head2 COMMON WATCHER METHODS
197
198This section lists methods common to all watchers.
199
200=over 4
201
202=item $w->start
203
204Starts a watcher if it isn't active already. Does nothing to an already
205active watcher. By default, all watchers start out in the active state
206(see the description of the C<_ns> variants if you need stopped watchers).
207
208=item $w->stop
209
210Stop a watcher if it is active. Also clear any pending events (events that
211have been received but that didn't yet result in a callback invocation),
212regardless of wether the watcher was active or not.
213
214=item $bool = $w->is_active
215
216Returns true if the watcher is active, false otherwise.
217
218=item $current_data = $w->data
219
220=item $old_data = $w->data ($new_data)
221
222Queries a freely usable data scalar on the watcher and optionally changes
223it. This is a way to associate custom data with a watcher:
224
225 my $w = EV::timer 60, 0, sub {
226 warn $_[0]->data;
227 };
228 $w->data ("print me!");
229
230=item $current_cb = $w->cb
231
232=item $old_cb = $w->cb ($new_cb)
233
234Queries the callback on the watcher and optionally changes it. You can do
235this at any time without the watcher restarting.
236
237=item $current_priority = $w->priority
238
239=item $old_priority = $w->priority ($new_priority)
240
241Queries the priority on the watcher and optionally changes it. Pending
242watchers with higher priority will be invoked first. The valid range of
243priorities lies between EV::MAXPRI (default 2) and EV::MINPRI (default
244-2). If the priority is outside this range it will automatically be
245normalised to the nearest valid priority.
246
247The default priority of any newly-created watcher is 0.
248
249Note that the priority semantics have not yet been fleshed out and are
250subject to almost certain change.
251
252=item $w->trigger ($revents)
253
254Call the callback *now* with the given event mask.
255
256=item $previous_state = $w->keepalive ($bool)
257
258Normally, C<EV::loop> will return when there are no active watchers
259(which is a "deadlock" because no progress can be made anymore). This is
260convinient because it allows you to start your watchers (and your jobs),
261call C<EV::loop> once and when it returns you know that all your jobs are
262finished (or they forgot to register some watchers for their task :).
263
264Sometimes, however, this gets in your way, for example when you the module
265that calls C<EV::loop> (usually the main program) is not the same module
266as a long-living watcher (for example a DNS client module written by
267somebody else even). Then you might want any outstanding requests to be
268handled, but you would not want to keep C<EV::loop> from returning just
269because you happen to have this long-running UDP port watcher.
270
271In this case you can clear the keepalive status, which means that even
272though your watcher is active, it won't keep C<EV::loop> from returning.
273
274The initial value for keepalive is true (enabled), and you cna change it
275any time.
276
277Example: Register an IO watcher for some UDP socket but do not keep the
278event loop from running just because of that watcher.
279
280 my $udp_socket = ...
281 my $udp_watcher = EV::io $udp_socket, EV::READ, sub { ... };
282 $udp_watcher->keepalive (0);
283
284=back
285
286
287=head2 WATCHER TYPES
288
289Each of the following subsections describes a single watcher type.
290
291=head3 IO WATCHERS - is this file descriptor readable or writable?
292
293=over 4
294
139=item my $w = EV::io $fileno_or_fh, $eventmask, $callback 295=item $w = EV::io $fileno_or_fh, $eventmask, $callback
140 296
141=item my $w = EV::io_ns $fileno_or_fh, $eventmask, $callback 297=item $w = EV::io_ns $fileno_or_fh, $eventmask, $callback
142 298
143As long as the returned watcher object is alive, call the C<$callback> 299As long as the returned watcher object is alive, call the C<$callback>
144when the events specified in C<$eventmask> happen. Initially, the timeout 300when at least one of events specified in C<$eventmask> occurs.
145is disabled.
146 301
147You can additionall set a timeout to occur on the watcher, but note that
148this timeout will not be reset when you get an I/O event in the EV::PERSIST
149case, and reaching a timeout will always stop the watcher even in the
150EV::PERSIST case.
151
152If you want a timeout to occur only after a specific time of inactivity, set
153a repeating timeout and do NOT use EV::PERSIST.
154
155Eventmask can be one or more of these constants ORed together: 302The $eventmask can be one or more of these constants ORed together:
156 303
157 EV::READ wait until read() wouldn't block anymore 304 EV::READ wait until read() wouldn't block anymore
158 EV::WRITE wait until write() wouldn't block anymore 305 EV::WRITE wait until write() wouldn't block anymore
159 EV::PERSIST stay active after a (non-timeout) event occured
160 306
161The C<io_ns> variant doesn't add/start the newly created watcher. 307The C<io_ns> variant doesn't start (activate) the newly created watcher.
162 308
163=item my $w = EV::timed_io $fileno_or_fh, $eventmask, $timeout, $callback 309=item $w->set ($fileno_or_fh, $eventmask)
164 310
165=item my $w = EV::timed_io_ns $fileno_or_fh, $eventmask, $timeout, $callback 311Reconfigures the watcher, see the constructor above for details. Can be
312called at any time.
166 313
167Same as C<io> and C<io_ns>, but also specifies a timeout (as if there was 314=item $current_fh = $w->fh
168a call to C<< $w->timeout ($timout, 1) >>. The persist flag is not allowed
169and will automatically be cleared. The watcher will be restarted after each event.
170 315
171If the timeout is zero or undef, no timeout will be set, and a normal 316=item $old_fh = $w->fh ($new_fh)
172watcher (with the persist flag set!) will be created.
173 317
174This has the effect of timing out after the specified period of inactivity 318Returns the previously set filehandle and optionally set a new one.
175has happened.
176 319
177Due to the design of libevent, this is also relatively inefficient, having 320=item $current_eventmask = $w->events
178one or two io watchers and a separate timeout watcher that you reset on
179activity (by calling its C<start> method) is usually more efficient.
180 321
322=item $old_eventmask = $w->events ($new_eventmask)
323
324Returns the previously set event mask and optionally set a new one.
325
326=back
327
328
329=head3 TIMER WATCHERS - relative and optionally repeating timeouts
330
331=over 4
332
181=item my $w = EV::timer $after, $repeat, $callback 333=item $w = EV::timer $after, $repeat, $callback
182 334
183=item my $w = EV::timer_ns $after, $repeat, $callback 335=item $w = EV::timer_ns $after, $repeat, $callback
184 336
185Calls the callback after C<$after> seconds. If C<$repeat> is true, the 337Calls the callback after C<$after> seconds (which may be fractional). If
186timer will be restarted after the callback returns. This means that the 338C<$repeat> is non-zero, the timer will be restarted (with the $repeat
187callback would be called roughly every C<$after> seconds, prolonged by the 339value as $after) after the callback returns.
188time the callback takes.
189 340
341This means that the callback would be called roughly after C<$after>
342seconds, and then every C<$repeat> seconds. The timer does his best not
343to drift, but it will not invoke the timer more often then once per event
344loop iteration, and might drift in other cases. If that isn't acceptable,
345look at EV::periodic, which can provide long-term stable timers.
346
347The timer is based on a monotonic clock, that is, if somebody is sitting
348in front of the machine while the timer is running and changes the system
349clock, the timer will nevertheless run (roughly) the same time.
350
190The C<timer_ns> variant doesn't add/start the newly created watcher. 351The C<timer_ns> variant doesn't start (activate) the newly created watcher.
191 352
192=item my $w = EV::timer_abs $at, $interval, $callback 353=item $w->set ($after, $repeat)
193 354
194=item my $w = EV::timer_abs_ns $at, $interval, $callback 355Reconfigures the watcher, see the constructor above for details. Can be called at
356any time.
195 357
196Similar to EV::timer, but the time is given as an absolute point in time 358=item $w->again
197(C<$at>), plus an optional C<$interval>.
198 359
199If the C<$interval> is zero, then the callback will be called at the time 360Similar to the C<start> method, but has special semantics for repeating timers:
200C<$at> if that is in the future, or as soon as possible if its in the
201past. It will not automatically repeat.
202 361
203If the C<$interval> is nonzero, then the watcher will always be scheduled 362If the timer is active and non-repeating, it will be stopped.
204to time out at the next C<$at + integer * $interval> time.
205 363
206This can be used to schedule a callback to run at very regular intervals, 364If the timer is active and repeating, reset the timeout to occur
207as long as the processing time is less then the interval (otherwise 365C<$repeat> seconds after now.
208obviously events will be skipped). 366
367If the timer is inactive and repeating, start it using the repeat value.
368
369Otherwise do nothing.
370
371This behaviour is useful when you have a timeout for some IO
372operation. You create a timer object with the same value for C<$after> and
373C<$repeat>, and then, in the read/write watcher, run the C<again> method
374on the timeout.
375
376=back
377
378
379=head3 PERIODIC WATCHERS - to cron or not to cron?
380
381=over 4
382
383=item $w = EV::periodic $at, $interval, $reschedule_cb, $callback
384
385=item $w = EV::periodic_ns $at, $interval, $reschedule_cb, $callback
386
387Similar to EV::timer, but is not based on relative timeouts but on
388absolute times. Apart from creating "simple" timers that trigger "at" the
389specified time, it can also be used for non-drifting absolute timers and
390more complex, cron-like, setups that are not adversely affected by time
391jumps (i.e. when the system clock is changed by explicit date -s or other
392means such as ntpd). It is also the most complex watcher type in EV.
393
394It has three distinct "modes":
395
396=over 4
397
398=item * absolute timer ($interval = $reschedule_cb = 0)
399
400This time simply fires at the wallclock time C<$at> and doesn't repeat. It
401will not adjust when a time jump occurs, that is, if it is to be run
402at January 1st 2011 then it will run when the system time reaches or
403surpasses this time.
404
405=item * non-repeating interval timer ($interval > 0, $reschedule_cb = 0)
406
407In this mode the watcher will always be scheduled to time out at the
408next C<$at + N * $interval> time (for some integer N) and then repeat,
409regardless of any time jumps.
410
411This can be used to create timers that do not drift with respect to system
412time:
413
414 my $hourly = EV::periodic 0, 3600, 0, sub { print "once/hour\n" };
415
416That doesn't mean there will always be 3600 seconds in between triggers,
417but only that the the clalback will be called when the system time shows a
418full hour (UTC).
209 419
210Another way to think about it (for the mathematically inclined) is that 420Another way to think about it (for the mathematically inclined) is that
211C<timer_abs> will try to tun the callback at the next possible time where 421EV::periodic will try to run the callback in this mode at the next
212C<$time = $at (mod $interval)>, regardless of any time jumps. 422possible time where C<$time = $at (mod $interval)>, regardless of any time
423jumps.
213 424
425=item * manual reschedule mode ($reschedule_cb = coderef)
426
427In this mode $interval and $at are both being ignored. Instead, each
428time the periodic watcher gets scheduled, the reschedule callback
429($reschedule_cb) will be called with the watcher as first, and the current
430time as second argument.
431
432I<This callback MUST NOT stop or destroy this or any other periodic
433watcher, ever>. If you need to stop it, return 1e30 and stop it
434afterwards.
435
436It must return the next time to trigger, based on the passed time value
437(that is, the lowest time value larger than to the second argument). It
438will usually be called just before the callback will be triggered, but
439might be called at other times, too.
440
441This can be used to create very complex timers, such as a timer that
442triggers on each midnight, local time (actually 24 hours after the last
443midnight, to keep the example simple. If you know a way to do it correctly
444in about the same space (without requiring elaborate modules), drop me a
445note :):
446
447 my $daily = EV::periodic 0, 0, sub {
448 my ($w, $now) = @_;
449
450 use Time::Local ();
451 my (undef, undef, undef, $d, $m, $y) = localtime $now;
452 86400 + Time::Local::timelocal 0, 0, 0, $d, $m, $y
453 }, sub {
454 print "it's midnight or likely shortly after, now\n";
455 };
456
457=back
458
214The C<timer_abs_ns> variant doesn't add/start the newly created watcher. 459The C<periodic_ns> variant doesn't start (activate) the newly created watcher.
215 460
461=item $w->set ($at, $interval, $reschedule_cb)
462
463Reconfigures the watcher, see the constructor above for details. Can be called at
464any time.
465
466=item $w->again
467
468Simply stops and starts the watcher again.
469
470=back
471
472
473=head3 SIGNAL WATCHERS - signal me when a signal gets signalled!
474
475=over 4
476
216=item my $w = EV::signal $signal, $callback 477=item $w = EV::signal $signal, $callback
217 478
218=item my $w = EV::signal_ns $signal, $callback 479=item $w = EV::signal_ns $signal, $callback
219 480
220Call the callback when $signal is received (the signal can be specified 481Call the callback when $signal is received (the signal can be specified by
221by number or by name, just as with kill or %SIG). Signal watchers are 482number or by name, just as with C<kill> or C<%SIG>).
222persistent no natter what.
223 483
224EV will grab the signal for the process (the kernel only allows one 484EV will grab the signal for the process (the kernel only allows one
225component to receive signals) when you start a signal watcher, and 485component to receive a signal at a time) when you start a signal watcher,
226removes it again when you stop it. Pelr does the same when you add/remove 486and removes it again when you stop it. Perl does the same when you
227callbacks to %SIG, so watch out. 487add/remove callbacks to C<%SIG>, so watch out.
228 488
229Unfortunately, only one handler can be registered per signal. Screw 489You can have as many signal watchers per signal as you want.
230libevent.
231 490
232The C<signal_ns> variant doesn't add/start the newly created watcher. 491The C<signal_ns> variant doesn't start (activate) the newly created watcher.
233 492
234=back 493=item $w->set ($signal)
235 494
236=head1 THE EV::Event CLASS 495Reconfigures the watcher, see the constructor above for details. Can be
496called at any time.
237 497
238All EV functions creating an event watcher (designated by C<my $w =>
239above) support the following methods on the returned watcher object:
240
241=over 4
242
243=item $w->add ($timeout)
244
245Stops and (re-)starts the event watcher, setting the optional timeout to
246the given value, or clearing the timeout if none is given.
247
248=item $w->start
249
250Stops and (re-)starts the event watcher without touching the timeout.
251
252=item $w->del
253
254=item $w->stop
255
256Stop the event watcher if it was started.
257
258=item $current_callback = $w->cb
259
260=item $old_callback = $w->cb ($new_callback)
261
262Return the previously set callback and optionally set a new one.
263
264=item $current_fh = $w->fh
265
266=item $old_fh = $w->fh ($new_fh)
267
268Returns the previously set filehandle and optionally set a new one (also
269clears the EV::SIGNAL flag when setting a filehandle).
270
271=item $current_signal = $w->signal 498=item $current_signum = $w->signal
272 499
273=item $old_signal = $w->signal ($new_signal) 500=item $old_signum = $w->signal ($new_signal)
274 501
502Returns the previously set signal (always as a number not name) and
503optionally set a new one.
504
505=back
506
507
508=head3 CHILD WATCHERS - watch out for process status changes
509
510=over 4
511
512=item $w = EV::child $pid, $callback
513
514=item $w = EV::child_ns $pid, $callback
515
516Call the callback when a status change for pid C<$pid> (or any pid if
517C<$pid> is 0) has been received. More precisely: when the process receives
518a C<SIGCHLD>, EV will fetch the outstanding exit/wait status for all
519changed/zombie children and call the callback.
520
521It is valid (and fully supported) to install a child watcher after a child
522has exited but before the event loop has started its next iteration (for
523example, first you C<fork>, then the new child process might exit, and
524only then do you install a child watcher in the parent for the new pid).
525
526You can access both exit (or tracing) status and pid by using the
527C<rstatus> and C<rpid> methods on the watcher object.
528
529You can have as many pid watchers per pid as you want, they will all be
530called.
531
532The C<child_ns> variant doesn't start (activate) the newly created watcher.
533
534=item $w->set ($pid)
535
536Reconfigures the watcher, see the constructor above for details. Can be called at
537any time.
538
539=item $current_pid = $w->pid
540
541=item $old_pid = $w->pid ($new_pid)
542
275Returns the previously set signal number and optionally set a new one (also sets 543Returns the previously set process id and optionally set a new one.
276the EV::SIGNAL flag when setting a signal).
277 544
278=item $current_eventmask = $w->events 545=item $exit_status = $w->rstatus
279 546
280=item $old_eventmask = $w->events ($new_eventmask) 547Return the exit/wait status (as returned by waitpid, see the waitpid entry
548in perlfunc).
281 549
550=item $pid = $w->rpid
551
552Return the pid of the awaited child (useful when you have installed a
553watcher for all pids).
554
555=back
556
557
558=head3 STAT WATCHERS - did the file attributes just change?
559
560=over 4
561
562=item $w = EV::stat $path, $interval, $callback
563
564=item $w = EV::stat_ns $path, $interval, $callback
565
566Call the callback when a file status change has been detected on
567C<$path>. The C<$path> does not need to exist, changing from "path exists"
568to "path does not exist" is a status change like any other.
569
570The C<$interval> is a recommended polling interval for systems where
571OS-supported change notifications don't exist or are not supported. If
572you use C<0> then an unspecified default is used (which is highly
573recommended!), which is to be expected to be around five seconds usually.
574
575This watcher type is not meant for massive numbers of stat watchers,
576as even with OS-supported change notifications, this can be
577resource-intensive.
578
579The C<stat_ns> variant doesn't start (activate) the newly created watcher.
580
581=item ... = $w->stat
582
583This call is very similar to the perl C<stat> built-in: It stats (using
584C<lstat>) the path specified in the watcher and sets perls stat cache (as
585well as EV's idea of the current stat values) to the values found.
586
587In scalar context, a boolean is return indicating success or failure of
588the stat. In list context, the same 13-value list as with stat is returned
589(except that the blksize and blocks fields are not reliable).
590
591In the case of an error, errno is set to C<ENOENT> (regardless of the
592actual error value) and the C<nlink> value is forced to zero (if the stat
593was successful then nlink is guaranteed to be non-zero).
594
595See also the next two entries for more info.
596
597=item ... = $w->attr
598
599Just like C<< $w->stat >>, but without the initial stat'ing: this returns
600the values most recently detected by EV. See the next entry for more info.
601
602=item ... = $w->prev
603
604Just like C<< $w->stat >>, but without the initial stat'ing: this returns
605the previous set of values, before the change.
606
607That is, when the watcher callback is invoked, C<< $w->prev >> will be set
608to the values found I<before> a change was detected, while C<< $w->attr >>
609returns the values found leading to the change detection. The difference (if any)
610between C<prev> and C<attr> is what triggered the callback.
611
612If you did something to the filesystem object and do not want to trigger
613yet another change, you can call C<stat> to update EV's idea of what the
614current attributes are.
615
616=item $w->set ($path, $interval)
617
618Reconfigures the watcher, see the constructor above for details. Can be
619called at any time.
620
621=item $current_path = $w->path
622
623=item $old_path = $w->path ($new_path)
624
282Returns the previously set event mask and optionally set a new one. 625Returns the previously set path and optionally set a new one.
283 626
284=item $w->timeout ($after, $repeat) 627=item $current_interval = $w->interval
285 628
286Resets the timeout (see C<EV::timer> for details). 629=item $old_interval = $w->interval ($new_interval)
287 630
288=item $w->timeout_abs ($at, $interval) 631Returns the previously set interval and optionally set a new one. Can be
632used to query the actual interval used.
289 633
290Resets the timeout (see C<EV::timer_abs> for details).
291
292=item $w->priority_set ($priority)
293
294Set the priority of the watcher to C<$priority> (0 <= $priority < $EV::NPRI).
295
296=back 634=back
635
636
637=head3 IDLE WATCHERS - when you've got nothing better to do...
638
639=over 4
640
641=item $w = EV::idle $callback
642
643=item $w = EV::idle_ns $callback
644
645Call the callback when there are no pending io, timer/periodic, signal or
646child events, i.e. when the process is idle.
647
648The process will not block as long as any idle watchers are active, and
649they will be called repeatedly until stopped.
650
651The C<idle_ns> variant doesn't start (activate) the newly created watcher.
652
653=back
654
655
656=head3 PREPARE WATCHERS - customise your event loop!
657
658=over 4
659
660=item $w = EV::prepare $callback
661
662=item $w = EV::prepare_ns $callback
663
664Call the callback just before the process would block. You can still
665create/modify any watchers at this point.
666
667See the EV::check watcher, below, for explanations and an example.
668
669The C<prepare_ns> variant doesn't start (activate) the newly created watcher.
670
671=back
672
673
674=head3 CHECK WATCHERS - customise your event loop even more!
675
676=over 4
677
678=item $w = EV::check $callback
679
680=item $w = EV::check_ns $callback
681
682Call the callback just after the process wakes up again (after it has
683gathered events), but before any other callbacks have been invoked.
684
685This is used to integrate other event-based software into the EV
686mainloop: You register a prepare callback and in there, you create io and
687timer watchers as required by the other software. Here is a real-world
688example of integrating Net::SNMP (with some details left out):
689
690 our @snmp_watcher;
691
692 our $snmp_prepare = EV::prepare sub {
693 # do nothing unless active
694 $dispatcher->{_event_queue_h}
695 or return;
696
697 # make the dispatcher handle any outstanding stuff
698 ... not shown
699
700 # create an IO watcher for each and every socket
701 @snmp_watcher = (
702 (map { EV::io $_, EV::READ, sub { } }
703 keys %{ $dispatcher->{_descriptors} }),
704
705 EV::timer +($event->[Net::SNMP::Dispatcher::_ACTIVE]
706 ? $event->[Net::SNMP::Dispatcher::_TIME] - EV::now : 0),
707 0, sub { },
708 );
709 };
710
711The callbacks are irrelevant (and are not even being called), the
712only purpose of those watchers is to wake up the process as soon as
713one of those events occurs (socket readable, or timer timed out). The
714corresponding EV::check watcher will then clean up:
715
716 our $snmp_check = EV::check sub {
717 # destroy all watchers
718 @snmp_watcher = ();
719
720 # make the dispatcher handle any new stuff
721 ... not shown
722 };
723
724The callbacks of the created watchers will not be called as the watchers
725are destroyed before this cna happen (remember EV::check gets called
726first).
727
728The C<check_ns> variant doesn't start (activate) the newly created watcher.
729
730=back
731
732
733=head3 FORK WATCHERS - the audacity to resume the event loop after a fork
734
735Fork watchers are called when a C<fork ()> was detected. The invocation
736is done before the event loop blocks next and before C<check> watchers
737are being called, and only in the child after the fork.
738
739=over 4
740
741=item $w = EV::fork $callback
742
743=item $w = EV::fork_ns $callback
744
745Call the callback before the event loop is resumed in the child process
746after a fork.
747
748The C<fork_ns> variant doesn't start (activate) the newly created watcher.
749
750=back
751
297 752
298=head1 THREADS 753=head1 THREADS
299 754
300Threads are not supported by this in any way. Perl pseudo-threads is evil 755Threads are not supported by this module in any way. Perl pseudo-threads
301and must die. 756is evil stuff and must die. As soon as Perl gains real threads I will work
757on thread support for it.
758
759=head1 FORK
760
761Most of the "improved" event delivering mechanisms of modern operating
762systems have quite a few problems with fork(2) (to put it bluntly: it is
763not supported and usually destructive). Libev makes it possible to work
764around this by having a function that recreates the kernel state after
765fork in the child.
766
767On non-win32 platforms, this module requires the pthread_atfork
768functionality to do this automatically for you. This function is quite
769buggy on most BSDs, though, so YMMV. The overhead for this is quite
770negligible, because everything the function currently does is set a flag
771that is checked only when the event loop gets used the next time, so when
772you do fork but not use EV, the overhead is minimal.
773
774On win32, there is no notion of fork so all this doesn't apply, of course.
302 775
303=cut 776=cut
304 777
305our $DIED = sub { 778our $DIED = sub {
306 warn "EV: error in callback (ignoring): $@"; 779 warn "EV: error in callback (ignoring): $@";
307}; 780};
308 781
309init; 782default_loop
310 783 or die 'EV: cannot initialise libev backend. bad $ENV{LIBEV_METHODS}?';
311push @AnyEvent::REGISTRY, [EV => "EV::AnyEvent"];
312 784
3131; 7851;
314 786
315=head1 SEE ALSO 787=head1 SEE ALSO
316 788
317 L<EV::DNS>, L<event(3)>, L<event.h>, L<evdns.h>. 789L<EV::DNS>.
318 L<EV::AnyEvent>.
319 790
320=head1 AUTHOR 791=head1 AUTHOR
321 792
322 Marc Lehmann <schmorp@schmorp.de> 793 Marc Lehmann <schmorp@schmorp.de>
323 http://home.schmorp.de/ 794 http://home.schmorp.de/

Diff Legend

Removed lines
+ Added lines
< Changed lines
> Changed lines