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

Comparing EV/EV.pm (file contents):
Revision 1.26 by root, Fri Nov 2 23:22:17 2007 UTC vs.
Revision 1.72 by root, Thu Dec 20 07:12:57 2007 UTC

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::periodic 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, 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 warn "stdin is readable, you entered: ", <STDIN>; 29 warn "stdin is readable, you entered: ", <STDIN>;
30 }; 30 };
31 31
32 # SIGNALS 32 # SIGNALS
33 33
34 my $w = EV::signal 'QUIT', sub { 34 my $w = EV::signal 'QUIT', sub {
35 warn "sigquit received\n"; 35 warn "sigquit received\n";
36 }; 36 };
37 37
38 # CHILD/PID STATUS CHANGES
39
38 my $w = EV::signal 3, sub { 40 my $w = EV::child 666, sub {
39 warn "sigquit received (this is GNU/Linux, right?)\n"; 41 my ($w, $revents) = @_;
42 my $status = $w->rstatus;
40 }; 43 };
41 44
42 # CHILD/PID STATUS CHANGES 45 # STAT CHANGES
43 46 my $w = EV::stat "/etc/passwd", 10, sub {
44 my $w = EV::child 666, sub {
45 my ($w, $revents, $status) = @_; 47 my ($w, $revents) = @_;
48 warn $w->path, " has changed somehow.\n";
46 }; 49 };
47 50
48 # MAINLOOP 51 # MAINLOOP
49 EV::loop; # loop until EV::loop_done is called 52 EV::loop; # loop until EV::unloop is called or all watchers stop
50 EV::loop EV::LOOP_ONESHOT; # block until at least one event could be handled 53 EV::loop EV::LOOP_ONESHOT; # block until at least one event could be handled
51 EV::loop EV::LOOP_NONBLOCK; # try to handle same events, but do not block 54 EV::loop EV::LOOP_NONBLOCK; # try to handle same events, but do not block
52 55
53=head1 DESCRIPTION 56=head1 DESCRIPTION
54 57
55This module provides an interface to libev 58This module provides an interface to libev
56(L<http://software.schmorp.de/pkg/libev.html>). 59(L<http://software.schmorp.de/pkg/libev.html>). While the documentation
60below is comprehensive, one might also consult the documentation of libev
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>, or just about in any case
64because it has much more detailed information.
57 65
58=cut 66=cut
59 67
60package EV; 68package EV;
61 69
62use strict; 70use strict;
63 71
64BEGIN { 72BEGIN {
65 our $VERSION = '0.5'; 73 our $VERSION = '2.0';
66 use XSLoader; 74 use XSLoader;
67 XSLoader::load "EV", $VERSION; 75 XSLoader::load "EV", $VERSION;
68} 76}
69 77
70@EV::Io::ISA = 78@EV::IO::ISA =
71@EV::Timer::ISA = 79@EV::Timer::ISA =
72@EV::Periodic::ISA = 80@EV::Periodic::ISA =
73@EV::Signal::ISA = 81@EV::Signal::ISA =
82@EV::Child::ISA =
83@EV::Stat::ISA =
74@EV::Idle::ISA = 84@EV::Idle::ISA =
75@EV::Prepare::ISA = 85@EV::Prepare::ISA =
76@EV::Check::ISA = 86@EV::Check::ISA =
77@EV::Child::ISA = "EV::Watcher"; 87@EV::Embed::ISA =
88@EV::Fork::ISA =
89 "EV::Watcher";
90
91@EV::Loop::Default::ISA = "EV::Loop";
78 92
79=head1 BASIC INTERFACE 93=head1 BASIC INTERFACE
80 94
81=over 4 95=over 4
82 96
83=item $EV::DIED 97=item $EV::DIED
84 98
85Must contain a reference to a function that is called when a callback 99Must contain a reference to a function that is called when a callback
86throws an exception (with $@ containing thr error). The default prints an 100throws an exception (with $@ containing the error). The default prints an
87informative message and continues. 101informative message and continues.
88 102
89If this callback throws an exception it will be silently ignored. 103If this callback throws an exception it will be silently ignored.
90 104
91=item $time = EV::time 105=item $time = EV::time
96 110
97Returns the time the last event loop iteration has been started. This 111Returns the time the last event loop iteration has been started. This
98is the time that (relative) timers are based on, and refering to it is 112is the time that (relative) timers are based on, and refering to it is
99usually faster then calling EV::time. 113usually faster then calling EV::time.
100 114
101=item $method = EV::ev_method 115=item $method = EV::method
102 116
103Returns an integer describing the backend used by libev (EV::METHOD_SELECT 117Returns an integer describing the backend used by libev (EV::METHOD_SELECT
104or EV::METHOD_EPOLL). 118or EV::METHOD_EPOLL).
105 119
106=item EV::loop [$flags] 120=item EV::loop [$flags]
107 121
108Begin checking for events and calling callbacks. It returns when a 122Begin checking for events and calling callbacks. It returns when a
109callback calls EV::loop_done. 123callback calls EV::unloop.
110 124
111The $flags argument can be one of the following: 125The $flags argument can be one of the following:
112 126
113 0 as above 127 0 as above
114 EV::LOOP_ONESHOT block at most once (wait, but do not loop) 128 EV::LOOP_ONESHOT block at most once (wait, but do not loop)
115 EV::LOOP_NONBLOCK do not block at all (fetch/handle events but do not wait) 129 EV::LOOP_NONBLOCK do not block at all (fetch/handle events but do not wait)
116 130
117=item EV::loop_done [$how] 131=item EV::unloop [$how]
118 132
119When called with no arguments or an argument of 1, makes the innermost 133When called with no arguments or an argument of EV::UNLOOP_ONE, makes the
120call to EV::loop return. 134innermost call to EV::loop return.
121 135
122When called with an agrument of 2, all calls to EV::loop will return as 136When called with an argument of EV::UNLOOP_ALL, all calls to EV::loop will return as
123fast as possible. 137fast as possible.
124 138
125=back 139=item $count = EV::loop_count
126 140
141Return the number of times the event loop has polled for new
142events. Sometiems useful as a generation counter.
143
144=item EV::once $fh_or_undef, $events, $timeout, $cb->($revents)
145
146This function rolls together an I/O and a timer watcher for a single
147one-shot event without the need for managing a watcher object.
148
149If C<$fh_or_undef> is a filehandle or file descriptor, then C<$events>
150must be a bitset containing either C<EV::READ>, C<EV::WRITE> or C<EV::READ
151| EV::WRITE>, indicating the type of I/O event you want to wait for. If
152you do not want to wait for some I/O event, specify C<undef> for
153C<$fh_or_undef> and C<0> for C<$events>).
154
155If timeout is C<undef> or negative, then there will be no
156timeout. Otherwise a EV::timer with this value will be started.
157
158When an error occurs or either the timeout or I/O watcher triggers, then
159the callback will be called with the received event set (in general
160you can expect it to be a combination of C<EV:ERROR>, C<EV::READ>,
161C<EV::WRITE> and C<EV::TIMEOUT>).
162
163EV::once doesn't return anything: the watchers stay active till either
164of them triggers, then they will be stopped and freed, and the callback
165invoked.
166
167=item EV::feed_fd_event ($fd, $revents)
168
169Feed an event on a file descriptor into EV. EV will react to this call as
170if the readyness notifications specified by C<$revents> (a combination of
171C<EV::READ> and C<EV::WRITE>) happened on the file descriptor C<$fd>.
172
173=item EV::feed_signal_event ($signal)
174
175Feed a signal event into EV. EV will react to this call as if the signal
176specified by C<$signal> had occured.
177
178=back
179
180
127=head2 WATCHER 181=head2 WATCHER OBJECTS
128 182
129A watcher is an object that gets created to record your interest in some 183A watcher is an object that gets created to record your interest in some
130event. For instance, if you want to wait for STDIN to become readable, you 184event. For instance, if you want to wait for STDIN to become readable, you
131would create an EV::io watcher for that: 185would create an EV::io watcher for that:
132 186
141events. 195events.
142 196
143Each watcher type has its associated bit in revents, so you can use the 197Each watcher type has its associated bit in revents, so you can use the
144same callback for multiple watchers. The event mask is named after the 198same callback for multiple watchers. The event mask is named after the
145type, i..e. EV::child sets EV::CHILD, EV::prepare sets EV::PREPARE, 199type, i..e. EV::child sets EV::CHILD, EV::prepare sets EV::PREPARE,
146EV::periodic sets EV::PERIODIC and so on, with the exception of IO events 200EV::periodic sets EV::PERIODIC and so on, with the exception of I/O events
147(which can set both EV::READ and EV::WRITE bits), and EV::timer (which 201(which can set both EV::READ and EV::WRITE bits), and EV::timer (which
148uses EV::TIMEOUT). 202uses EV::TIMEOUT).
149 203
150In the rare case where one wants to create a watcher but not start it at 204In the rare case where one wants to create a watcher but not start it at
151the same time, each constructor has a variant with a trailing C<_ns> in 205the same time, each constructor has a variant with a trailing C<_ns> in
157 211
158Also, all methods changing some aspect of a watcher (->set, ->priority, 212Also, all methods changing some aspect of a watcher (->set, ->priority,
159->fh and so on) automatically stop and start it again if it is active, 213->fh and so on) automatically stop and start it again if it is active,
160which means pending events get lost. 214which means pending events get lost.
161 215
162=head2 WATCHER TYPES 216=head2 COMMON WATCHER METHODS
163 217
164Now lets move to the existing watcher types and asociated methods. 218This section lists methods common to all watchers.
165
166The following methods are available for all watchers. Then followes a
167description of each watcher constructor (EV::io, EV::timer, EV::periodic,
168EV::signal, EV::child, EV::idle, EV::prepare and EV::check), followed by
169any type-specific methods (if any).
170 219
171=over 4 220=over 4
172 221
173=item $w->start 222=item $w->start
174 223
178 227
179=item $w->stop 228=item $w->stop
180 229
181Stop a watcher if it is active. Also clear any pending events (events that 230Stop a watcher if it is active. Also clear any pending events (events that
182have been received but that didn't yet result in a callback invocation), 231have been received but that didn't yet result in a callback invocation),
183regardless of wether the watcher was active or not. 232regardless of whether the watcher was active or not.
184 233
185=item $bool = $w->is_active 234=item $bool = $w->is_active
186 235
187Returns true if the watcher is active, false otherwise. 236Returns true if the watcher is active, false otherwise.
237
238=item $current_data = $w->data
239
240=item $old_data = $w->data ($new_data)
241
242Queries a freely usable data scalar on the watcher and optionally changes
243it. This is a way to associate custom data with a watcher:
244
245 my $w = EV::timer 60, 0, sub {
246 warn $_[0]->data;
247 };
248 $w->data ("print me!");
188 249
189=item $current_cb = $w->cb 250=item $current_cb = $w->cb
190 251
191=item $old_cb = $w->cb ($new_cb) 252=item $old_cb = $w->cb ($new_cb)
192 253
201watchers with higher priority will be invoked first. The valid range of 262watchers with higher priority will be invoked first. The valid range of
202priorities lies between EV::MAXPRI (default 2) and EV::MINPRI (default 263priorities lies between EV::MAXPRI (default 2) and EV::MINPRI (default
203-2). If the priority is outside this range it will automatically be 264-2). If the priority is outside this range it will automatically be
204normalised to the nearest valid priority. 265normalised to the nearest valid priority.
205 266
206The default priority of any newly-created weatcher is 0. 267The default priority of any newly-created watcher is 0.
207 268
269Note that the priority semantics have not yet been fleshed out and are
270subject to almost certain change.
271
208=item $w->trigger ($revents) 272=item $w->invoke ($revents)
209 273
210Call the callback *now* with the given event mask. 274Call the callback *now* with the given event mask.
211 275
276=item $w->feed_event ($revents)
277
278Feed some events on this watcher into EV. EV will react to this call as if
279the watcher had received the given C<$revents> mask.
280
281=item $revents = $w->clear_pending
282
283If the watcher is pending, this function returns clears its pending status
284and returns its C<$revents> bitset (as if its callback was invoked). If the
285watcher isn't pending it does nothing and returns C<0>.
286
287=item $previous_state = $w->keepalive ($bool)
288
289Normally, C<EV::loop> will return when there are no active watchers
290(which is a "deadlock" because no progress can be made anymore). This is
291convinient because it allows you to start your watchers (and your jobs),
292call C<EV::loop> once and when it returns you know that all your jobs are
293finished (or they forgot to register some watchers for their task :).
294
295Sometimes, however, this gets in your way, for example when you the module
296that calls C<EV::loop> (usually the main program) is not the same module
297as a long-living watcher (for example a DNS client module written by
298somebody else even). Then you might want any outstanding requests to be
299handled, but you would not want to keep C<EV::loop> from returning just
300because you happen to have this long-running UDP port watcher.
301
302In this case you can clear the keepalive status, which means that even
303though your watcher is active, it won't keep C<EV::loop> from returning.
304
305The initial value for keepalive is true (enabled), and you cna change it
306any time.
307
308Example: Register an I/O watcher for some UDP socket but do not keep the
309event loop from running just because of that watcher.
310
311 my $udp_socket = ...
312 my $udp_watcher = EV::io $udp_socket, EV::READ, sub { ... };
313 $udp_watcher->keepalive (0);
314
315=back
316
317
318=head2 WATCHER TYPES
319
320Each of the following subsections describes a single watcher type.
321
322=head3 I/O WATCHERS - is this file descriptor readable or writable?
323
324=over 4
212 325
213=item $w = EV::io $fileno_or_fh, $eventmask, $callback 326=item $w = EV::io $fileno_or_fh, $eventmask, $callback
214 327
215=item $w = EV::io_ns $fileno_or_fh, $eventmask, $callback 328=item $w = EV::io_ns $fileno_or_fh, $eventmask, $callback
216 329
217As long as the returned watcher object is alive, call the C<$callback> 330As long as the returned watcher object is alive, call the C<$callback>
218when the events specified in C<$eventmask>. 331when at least one of events specified in C<$eventmask> occurs.
219 332
220The $eventmask can be one or more of these constants ORed together: 333The $eventmask can be one or more of these constants ORed together:
221 334
222 EV::READ wait until read() wouldn't block anymore 335 EV::READ wait until read() wouldn't block anymore
223 EV::WRITE wait until write() wouldn't block anymore 336 EV::WRITE wait until write() wouldn't block anymore
239 352
240=item $old_eventmask = $w->events ($new_eventmask) 353=item $old_eventmask = $w->events ($new_eventmask)
241 354
242Returns the previously set event mask and optionally set a new one. 355Returns the previously set event mask and optionally set a new one.
243 356
357=back
358
359
360=head3 TIMER WATCHERS - relative and optionally repeating timeouts
361
362=over 4
244 363
245=item $w = EV::timer $after, $repeat, $callback 364=item $w = EV::timer $after, $repeat, $callback
246 365
247=item $w = EV::timer_ns $after, $repeat, $callback 366=item $w = EV::timer_ns $after, $repeat, $callback
248 367
249Calls the callback after C<$after> seconds. If C<$repeat> is non-zero, 368Calls the callback after C<$after> seconds (which may be fractional). If
250the timer will be restarted (with the $repeat value as $after) after the 369C<$repeat> is non-zero, the timer will be restarted (with the $repeat
251callback returns. 370value as $after) after the callback returns.
252 371
253This means that the callback would be called roughly after C<$after> 372This means that the callback would be called roughly after C<$after>
254seconds, and then every C<$repeat> seconds. "Roughly" because the time of 373seconds, and then every C<$repeat> seconds. The timer does his best not
255callback processing is not taken into account, so the timer will slowly 374to drift, but it will not invoke the timer more often then once per event
256drift. If that isn't acceptable, look at EV::periodic. 375loop iteration, and might drift in other cases. If that isn't acceptable,
376look at EV::periodic, which can provide long-term stable timers.
257 377
258The timer is based on a monotonic clock, that is if somebody is sitting 378The timer is based on a monotonic clock, that is, if somebody is sitting
259in front of the machine while the timer is running and changes the system 379in front of the machine while the timer is running and changes the system
260clock, the timer will nevertheless run (roughly) the same time. 380clock, the timer will nevertheless run (roughly) the same time.
261 381
262The C<timer_ns> variant doesn't start (activate) the newly created watcher. 382The C<timer_ns> variant doesn't start (activate) the newly created watcher.
263 383
264=item $w->set ($after, $repeat) 384=item $w->set ($after, $repeat)
265 385
266Reconfigures the watcher, see the constructor above for details. Can be at 386Reconfigures the watcher, see the constructor above for details. Can be called at
267any time. 387any time.
268 388
269=item $w->again 389=item $w->again
270 390
271Similar to the C<start> method, but has special semantics for repeating timers: 391Similar to the C<start> method, but has special semantics for repeating timers:
392
393If the timer is active and non-repeating, it will be stopped.
272 394
273If the timer is active and repeating, reset the timeout to occur 395If the timer is active and repeating, reset the timeout to occur
274C<$repeat> seconds after now. 396C<$repeat> seconds after now.
275 397
276If the timer is active and non-repeating, it will be stopped.
277
278If the timer is in active and repeating, start it. 398If the timer is inactive and repeating, start it using the repeat value.
279 399
280Otherwise do nothing. 400Otherwise do nothing.
281 401
282This behaviour is useful when you have a timeout for some IO 402This behaviour is useful when you have a timeout for some IO
283operation. You create a timer object with the same value for C<$after> and 403operation. You create a timer object with the same value for C<$after> and
284C<$repeat>, and then, in the read/write watcher, run the C<again> method 404C<$repeat>, and then, in the read/write watcher, run the C<again> method
285on the timeout. 405on the timeout.
286 406
407=back
287 408
409
410=head3 PERIODIC WATCHERS - to cron or not to cron?
411
412=over 4
413
288=item $w = EV::periodic $at, $interval, $callback 414=item $w = EV::periodic $at, $interval, $reschedule_cb, $callback
289 415
290=item $w = EV::periodic_ns $at, $interval, $callback 416=item $w = EV::periodic_ns $at, $interval, $reschedule_cb, $callback
291 417
292Similar to EV::timer, but the time is given as an absolute point in time 418Similar to EV::timer, but is not based on relative timeouts but on
293(C<$at>), plus an optional C<$interval>. 419absolute times. Apart from creating "simple" timers that trigger "at" the
420specified time, it can also be used for non-drifting absolute timers and
421more complex, cron-like, setups that are not adversely affected by time
422jumps (i.e. when the system clock is changed by explicit date -s or other
423means such as ntpd). It is also the most complex watcher type in EV.
294 424
295If the C<$interval> is zero, then the callback will be called at the time 425It has three distinct "modes":
296C<$at> if that is in the future, or as soon as possible if it is in the
297past. It will not automatically repeat.
298 426
299If the C<$interval> is nonzero, then the watcher will always be scheduled 427=over 4
300to time out at the next C<$at + N * $interval> time.
301 428
302This can be used to schedule a callback to run at very regular intervals, 429=item * absolute timer ($interval = $reschedule_cb = 0)
303as long as the processing time is less then the interval (otherwise 430
304obviously events will be skipped). 431This time simply fires at the wallclock time C<$at> and doesn't repeat. It
432will not adjust when a time jump occurs, that is, if it is to be run
433at January 1st 2011 then it will run when the system time reaches or
434surpasses this time.
435
436=item * non-repeating interval timer ($interval > 0, $reschedule_cb = 0)
437
438In this mode the watcher will always be scheduled to time out at the
439next C<$at + N * $interval> time (for some integer N) and then repeat,
440regardless of any time jumps.
441
442This can be used to create timers that do not drift with respect to system
443time:
444
445 my $hourly = EV::periodic 0, 3600, 0, sub { print "once/hour\n" };
446
447That doesn't mean there will always be 3600 seconds in between triggers,
448but only that the the clalback will be called when the system time shows a
449full hour (UTC).
305 450
306Another way to think about it (for the mathematically inclined) is that 451Another way to think about it (for the mathematically inclined) is that
307EV::periodic will try to run the callback at the next possible time where 452EV::periodic will try to run the callback in this mode at the next
308C<$time = $at (mod $interval)>, regardless of any time jumps. 453possible time where C<$time = $at (mod $interval)>, regardless of any time
454jumps.
309 455
310This periodic timer is based on "wallclock time", that is, if the clock 456=item * manual reschedule mode ($reschedule_cb = coderef)
311changes (C<ntp>, C<date -s> etc.), then the timer will nevertheless run at 457
312the specified time. This means it will never drift (it might jitter, but 458In this mode $interval and $at are both being ignored. Instead, each
313it will not drift). 459time the periodic watcher gets scheduled, the reschedule callback
460($reschedule_cb) will be called with the watcher as first, and the current
461time as second argument.
462
463I<This callback MUST NOT stop or destroy this or any other periodic
464watcher, ever>. If you need to stop it, return 1e30 and stop it
465afterwards.
466
467It must return the next time to trigger, based on the passed time value
468(that is, the lowest time value larger than to the second argument). It
469will usually be called just before the callback will be triggered, but
470might be called at other times, too.
471
472This can be used to create very complex timers, such as a timer that
473triggers on each midnight, local time (actually 24 hours after the last
474midnight, to keep the example simple. If you know a way to do it correctly
475in about the same space (without requiring elaborate modules), drop me a
476note :):
477
478 my $daily = EV::periodic 0, 0, sub {
479 my ($w, $now) = @_;
480
481 use Time::Local ();
482 my (undef, undef, undef, $d, $m, $y) = localtime $now;
483 86400 + Time::Local::timelocal 0, 0, 0, $d, $m, $y
484 }, sub {
485 print "it's midnight or likely shortly after, now\n";
486 };
487
488=back
314 489
315The C<periodic_ns> variant doesn't start (activate) the newly created watcher. 490The C<periodic_ns> variant doesn't start (activate) the newly created watcher.
316 491
317=item $w->set ($at, $interval) 492=item $w->set ($at, $interval, $reschedule_cb)
318 493
319Reconfigures the watcher, see the constructor above for details. Can be at 494Reconfigures the watcher, see the constructor above for details. Can be called at
320any time. 495any time.
321 496
497=item $w->again
498
499Simply stops and starts the watcher again.
500
501=item $time = $w->at
502
503Return the time that the watcher is expected to trigger next.
504
505=back
506
507
508=head3 SIGNAL WATCHERS - signal me when a signal gets signalled!
509
510=over 4
322 511
323=item $w = EV::signal $signal, $callback 512=item $w = EV::signal $signal, $callback
324 513
325=item $w = EV::signal_ns $signal, $callback 514=item $w = EV::signal_ns $signal, $callback
326 515
327Call the callback when $signal is received (the signal can be specified 516Call the callback when $signal is received (the signal can be specified by
328by number or by name, just as with kill or %SIG). 517number or by name, just as with C<kill> or C<%SIG>).
329 518
330EV will grab the signal for the process (the kernel only allows one 519EV will grab the signal for the process (the kernel only allows one
331component to receive a signal at a time) when you start a signal watcher, 520component to receive a signal at a time) when you start a signal watcher,
332and removes it again when you stop it. Perl does the same when you 521and removes it again when you stop it. Perl does the same when you
333add/remove callbacks to %SIG, so watch out. 522add/remove callbacks to C<%SIG>, so watch out.
334 523
335You can have as many signal watchers per signal as you want. 524You can have as many signal watchers per signal as you want.
336 525
337The C<signal_ns> variant doesn't start (activate) the newly created watcher. 526The C<signal_ns> variant doesn't start (activate) the newly created watcher.
338 527
339=item $w->set ($signal) 528=item $w->set ($signal)
340 529
341Reconfigures the watcher, see the constructor above for details. Can be at 530Reconfigures the watcher, see the constructor above for details. Can be
342any time. 531called at any time.
343 532
344=item $current_signum = $w->signal 533=item $current_signum = $w->signal
345 534
346=item $old_signum = $w->signal ($new_signal) 535=item $old_signum = $w->signal ($new_signal)
347 536
348Returns the previously set signal (always as a number not name) and 537Returns the previously set signal (always as a number not name) and
349optionally set a new one. 538optionally set a new one.
350 539
540=back
541
542
543=head3 CHILD WATCHERS - watch out for process status changes
544
545=over 4
351 546
352=item $w = EV::child $pid, $callback 547=item $w = EV::child $pid, $callback
353 548
354=item $w = EV::child_ns $pid, $callback 549=item $w = EV::child_ns $pid, $callback
355 550
356Call the callback when a status change for pid C<$pid> (or any pid 551Call the callback when a status change for pid C<$pid> (or any pid if
357if C<$pid> is 0) has been received. More precisely: when the process 552C<$pid> is 0) has been received. More precisely: when the process receives
358receives a SIGCHLD, EV will fetch the outstanding exit/wait status for all 553a C<SIGCHLD>, EV will fetch the outstanding exit/wait status for all
359changed/zombie children and call the callback. 554changed/zombie children and call the callback.
360 555
361Unlike all other callbacks, this callback will be called with an 556It is valid (and fully supported) to install a child watcher after a child
362additional third argument which is the exit status. See the C<waitpid> 557has exited but before the event loop has started its next iteration (for
363function for details. 558example, first you C<fork>, then the new child process might exit, and
559only then do you install a child watcher in the parent for the new pid).
364 560
561You can access both exit (or tracing) status and pid by using the
562C<rstatus> and C<rpid> methods on the watcher object.
563
365You can have as many pid watchers per pid as you want. 564You can have as many pid watchers per pid as you want, they will all be
565called.
366 566
367The C<child_ns> variant doesn't start (activate) the newly created watcher. 567The C<child_ns> variant doesn't start (activate) the newly created watcher.
368 568
369=item $w->set ($pid) 569=item $w->set ($pid)
370 570
371Reconfigures the watcher, see the constructor above for details. Can be at 571Reconfigures the watcher, see the constructor above for details. Can be called at
372any time. 572any time.
373 573
374=item $current_pid = $w->pid 574=item $current_pid = $w->pid
375 575
376=item $old_pid = $w->pid ($new_pid) 576=item $old_pid = $w->pid ($new_pid)
377 577
378Returns the previously set process id and optionally set a new one. 578Returns the previously set process id and optionally set a new one.
379 579
580=item $exit_status = $w->rstatus
581
582Return the exit/wait status (as returned by waitpid, see the waitpid entry
583in perlfunc).
584
585=item $pid = $w->rpid
586
587Return the pid of the awaited child (useful when you have installed a
588watcher for all pids).
589
590=back
591
592
593=head3 STAT WATCHERS - did the file attributes just change?
594
595=over 4
596
597=item $w = EV::stat $path, $interval, $callback
598
599=item $w = EV::stat_ns $path, $interval, $callback
600
601Call the callback when a file status change has been detected on
602C<$path>. The C<$path> does not need to exist, changing from "path exists"
603to "path does not exist" is a status change like any other.
604
605The C<$interval> is a recommended polling interval for systems where
606OS-supported change notifications don't exist or are not supported. If
607you use C<0> then an unspecified default is used (which is highly
608recommended!), which is to be expected to be around five seconds usually.
609
610This watcher type is not meant for massive numbers of stat watchers,
611as even with OS-supported change notifications, this can be
612resource-intensive.
613
614The C<stat_ns> variant doesn't start (activate) the newly created watcher.
615
616=item ... = $w->stat
617
618This call is very similar to the perl C<stat> built-in: It stats (using
619C<lstat>) the path specified in the watcher and sets perls stat cache (as
620well as EV's idea of the current stat values) to the values found.
621
622In scalar context, a boolean is return indicating success or failure of
623the stat. In list context, the same 13-value list as with stat is returned
624(except that the blksize and blocks fields are not reliable).
625
626In the case of an error, errno is set to C<ENOENT> (regardless of the
627actual error value) and the C<nlink> value is forced to zero (if the stat
628was successful then nlink is guaranteed to be non-zero).
629
630See also the next two entries for more info.
631
632=item ... = $w->attr
633
634Just like C<< $w->stat >>, but without the initial stat'ing: this returns
635the values most recently detected by EV. See the next entry for more info.
636
637=item ... = $w->prev
638
639Just like C<< $w->stat >>, but without the initial stat'ing: this returns
640the previous set of values, before the change.
641
642That is, when the watcher callback is invoked, C<< $w->prev >> will be set
643to the values found I<before> a change was detected, while C<< $w->attr >>
644returns the values found leading to the change detection. The difference (if any)
645between C<prev> and C<attr> is what triggered the callback.
646
647If you did something to the filesystem object and do not want to trigger
648yet another change, you can call C<stat> to update EV's idea of what the
649current attributes are.
650
651=item $w->set ($path, $interval)
652
653Reconfigures the watcher, see the constructor above for details. Can be
654called at any time.
655
656=item $current_path = $w->path
657
658=item $old_path = $w->path ($new_path)
659
660Returns the previously set path and optionally set a new one.
661
662=item $current_interval = $w->interval
663
664=item $old_interval = $w->interval ($new_interval)
665
666Returns the previously set interval and optionally set a new one. Can be
667used to query the actual interval used.
668
669=back
670
671
672=head3 IDLE WATCHERS - when you've got nothing better to do...
673
674=over 4
380 675
381=item $w = EV::idle $callback 676=item $w = EV::idle $callback
382 677
383=item $w = EV::idle_ns $callback 678=item $w = EV::idle_ns $callback
384 679
385Call the callback when there are no pending io, timer/periodic, signal or 680Call the callback when there are no other pending watchers of the same or
386child events, i.e. when the process is idle. 681higher priority (excluding check, prepare and other idle watchers of the
682same or lower priority, of course). They are called idle watchers because
683when the watcher is the highest priority pending event in the process, the
684process is considered to be idle at that priority.
685
686If you want a watcher that is only ever called when I<no> other events are
687outstanding you have to set the priority to C<EV::MINPRI>.
387 688
388The process will not block as long as any idle watchers are active, and 689The process will not block as long as any idle watchers are active, and
389they will be called repeatedly until stopped. 690they will be called repeatedly until stopped.
390 691
692For example, if you have idle watchers at priority C<0> and C<1>, and
693an I/O watcher at priority C<0>, then the idle watcher at priority C<1>
694and the I/O watcher will always run when ready. Only when the idle watcher
695at priority C<1> is stopped and the I/O watcher at priority C<0> is not
696pending with the C<0>-priority idle watcher be invoked.
697
391The C<idle_ns> variant doesn't start (activate) the newly created watcher. 698The C<idle_ns> variant doesn't start (activate) the newly created watcher.
392 699
700=back
701
702
703=head3 PREPARE WATCHERS - customise your event loop!
704
705=over 4
393 706
394=item $w = EV::prepare $callback 707=item $w = EV::prepare $callback
395 708
396=item $w = EV::prepare_ns $callback 709=item $w = EV::prepare_ns $callback
397 710
400 713
401See the EV::check watcher, below, for explanations and an example. 714See the EV::check watcher, below, for explanations and an example.
402 715
403The C<prepare_ns> variant doesn't start (activate) the newly created watcher. 716The C<prepare_ns> variant doesn't start (activate) the newly created watcher.
404 717
718=back
719
720
721=head3 CHECK WATCHERS - customise your event loop even more!
722
723=over 4
405 724
406=item $w = EV::check $callback 725=item $w = EV::check $callback
407 726
408=item $w = EV::check_ns $callback 727=item $w = EV::check_ns $callback
409 728
421 # do nothing unless active 740 # do nothing unless active
422 $dispatcher->{_event_queue_h} 741 $dispatcher->{_event_queue_h}
423 or return; 742 or return;
424 743
425 # make the dispatcher handle any outstanding stuff 744 # make the dispatcher handle any outstanding stuff
745 ... not shown
426 746
427 # create an IO watcher for each and every socket 747 # create an I/O watcher for each and every socket
428 @snmp_watcher = ( 748 @snmp_watcher = (
429 (map { EV::io $_, EV::READ, sub { } } 749 (map { EV::io $_, EV::READ, sub { } }
430 keys %{ $dispatcher->{_descriptors} }), 750 keys %{ $dispatcher->{_descriptors} }),
751
752 EV::timer +($event->[Net::SNMP::Dispatcher::_ACTIVE]
753 ? $event->[Net::SNMP::Dispatcher::_TIME] - EV::now : 0),
754 0, sub { },
431 ); 755 );
432
433 # if there are any timeouts, also create a timer
434 push @snmp_watcher, EV::timer $event->[Net::SNMP::Dispatcher::_TIME] - EV::now, 0, sub { }
435 if $event->[Net::SNMP::Dispatcher::_ACTIVE];
436 }; 756 };
437 757
438The callbacks are irrelevant, the only purpose of those watchers is 758The callbacks are irrelevant (and are not even being called), the
439to wake up the process as soon as one of those events occurs (socket 759only purpose of those watchers is to wake up the process as soon as
440readable, or timer timed out). The corresponding EV::check watcher will then 760one of those events occurs (socket readable, or timer timed out). The
441clean up: 761corresponding EV::check watcher will then clean up:
442 762
443 our $snmp_check = EV::check sub { 763 our $snmp_check = EV::check sub {
444 # destroy all watchers 764 # destroy all watchers
445 @snmp_watcher = (); 765 @snmp_watcher = ();
446 766
447 # make the dispatcher handle any new stuff 767 # make the dispatcher handle any new stuff
768 ... not shown
448 }; 769 };
449 770
450The callbacks of the created watchers will not be called as the watchers 771The callbacks of the created watchers will not be called as the watchers
451are destroyed before this cna happen (remember EV::check gets called 772are destroyed before this cna happen (remember EV::check gets called
452first). 773first).
453 774
454The C<check_ns> variant doesn't start (activate) the newly created watcher. 775The C<check_ns> variant doesn't start (activate) the newly created watcher.
455 776
456=back 777=back
457 778
779
780=head3 FORK WATCHERS - the audacity to resume the event loop after a fork
781
782Fork watchers are called when a C<fork ()> was detected. The invocation
783is done before the event loop blocks next and before C<check> watchers
784are being called, and only in the child after the fork.
785
786=over 4
787
788=item $w = EV::fork $callback
789
790=item $w = EV::fork_ns $callback
791
792Call the callback before the event loop is resumed in the child process
793after a fork.
794
795The C<fork_ns> variant doesn't start (activate) the newly created watcher.
796
797=back
798
799
800=head1 PERL SIGNALS
801
802While Perl signal handling (C<%SIG>) is not affected by EV, the behaviour
803with EV is as the same as any other C library: Perl-signals will only be
804handled when Perl runs, which means your signal handler might be invoked
805only the next time an event callback is invoked.
806
807The solution is to use EV signal watchers (see C<EV::signal>), which will
808ensure proper operations with regards to other event watchers.
809
810If you cannot do this for whatever reason, you can also force a watcher
811to be called on every event loop iteration by installing a C<EV::check>
812watcher:
813
814 my $async_check = EV::check sub { };
815
816This ensures that perl shortly gets into control for a short time, and
817also ensures slower overall operation.
818
458=head1 THREADS 819=head1 THREADS
459 820
460Threads are not supported by this in any way. Perl pseudo-threads is evil 821Threads are not supported by this module in any way. Perl pseudo-threads
461stuff and must die. 822is evil stuff and must die. As soon as Perl gains real threads I will work
823on thread support for it.
824
825=head1 FORK
826
827Most of the "improved" event delivering mechanisms of modern operating
828systems have quite a few problems with fork(2) (to put it bluntly: it is
829not supported and usually destructive). Libev makes it possible to work
830around this by having a function that recreates the kernel state after
831fork in the child.
832
833On non-win32 platforms, this module requires the pthread_atfork
834functionality to do this automatically for you. This function is quite
835buggy on most BSDs, though, so YMMV. The overhead for this is quite
836negligible, because everything the function currently does is set a flag
837that is checked only when the event loop gets used the next time, so when
838you do fork but not use EV, the overhead is minimal.
839
840On win32, there is no notion of fork so all this doesn't apply, of course.
462 841
463=cut 842=cut
464 843
465our $DIED = sub { 844our $DIED = sub {
466 warn "EV: error in callback (ignoring): $@"; 845 warn "EV: error in callback (ignoring): $@";
467}; 846};
468 847
469init 848default_loop
470 or die 'EV: cannot initialise libev backend. bad $ENV{LIBEV_METHODS}?'; 849 or die 'EV: cannot initialise libev backend. bad $ENV{LIBEV_FLAGS}?';
471
472push @AnyEvent::REGISTRY, [EV => "EV::AnyEvent"];
473 850
4741; 8511;
475 852
476=head1 SEE ALSO 853=head1 SEE ALSO
477 854
478 L<EV::DNS>, L<EV::AnyEvent>. 855L<EV::ADNS> (asynchronous dns), L<Glib::EV> (makes Glib/Gtk2 use EV as
856event loop), L<Coro::EV> (efficient coroutines with EV).
479 857
480=head1 AUTHOR 858=head1 AUTHOR
481 859
482 Marc Lehmann <schmorp@schmorp.de> 860 Marc Lehmann <schmorp@schmorp.de>
483 http://home.schmorp.de/ 861 http://home.schmorp.de/

Diff Legend

Removed lines
+ Added lines
< Changed lines
> Changed lines