[ViewVC] Diff of: cvs/EV/EV.pm

Comparing EV/EV.pm (file contents):
Revision 1.27 by root, Sat Nov 3 09:19:58 2007 UTC vs.
Revision 1.126 by root, Tue Mar 16 17:11:48 2010 UTC

…		…
2		2
3	EV - perl interface to libev, a high performance full-featured event loop	3	EV - 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
		9	# TIMERS
		10
		11	my $w = EV::timer 2, 0, sub {
		12	warn "is called after 2s";
		13	};
		14
		15	my $w = EV::timer 2, 2, sub {
		16	warn "is called roughly every 2s (repeat = 2)";
		17	};
		18
		19	undef $w; # destroy event watcher again
		20
		21	my $w = EV::periodic 0, 60, 0, sub {
		22	warn "is called every minute, on the minute, exactly";
		23	};
		24
		25	# IO
		26
		27	my $w = EV::io *STDIN, EV::READ, sub {
		28	my ($w, $revents) = @_; # all callbacks receive the watcher and event mask
		29	warn "stdin is readable, you entered: ", <STDIN>;
		30	};
		31
		32	# SIGNALS
		33
		34	my $w = EV::signal 'QUIT', sub {
		35	warn "sigquit received\n";
		36	};
		37
		38	# CHILD/PID STATUS CHANGES
8		39
9	# TIMERS	40	my $w = EV::child 666, 0, sub {
		41	my ($w, $revents) = @_;
		42	my $status = $w->rstatus;
		43	};
10		44
11	my $w = EV::timer 2, 0, sub {	45	# STAT CHANGES
12	warn "is called after 2s";	46	my $w = EV::stat "/etc/passwd", 10, sub {
13	};
14
15	my $w = EV::timer 2, 1, sub {
16	warn "is called roughly every 2s (repeat = 1)";
17	};
18
19	undef $w; # destroy event watcher again
20
21	my $w = EV::periodic 0, 60, sub {
22	warn "is called every minute, on the minute, exactly";
23	};
24
25	# IO
26
27	my $w = EV::io *STDIN, EV::READ, sub {
28	my ($w, $revents) = @_; # all callbacks get the watcher object and event mask
29	warn "stdin is readable, you entered: ", <STDIN>;
30	};
31
32	# SIGNALS
33
34	my $w = EV::signal 'QUIT', sub {
35	warn "sigquit received\n";
36	};
37
38	my $w = EV::signal 3, sub {
39	warn "sigquit received (this is GNU/Linux, right?)\n";
40	};
41
42	# CHILD/PID STATUS CHANGES
43
44	my $w = EV::child 666, sub {
45	my ($w, $revents) = @_;	47	my ($w, $revents) = @_;
46	# my $pid = $w->rpid;	48	warn $w->path, " has changed somehow.\n";
47	my $status = $w->rstatus;
48	};	49	};
49		50
50	# MAINLOOP	51	# MAINLOOP
51	EV::loop; # loop until EV::loop_done is called	52	EV::loop; # loop until EV::unloop is called or all watchers stop
52	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
53	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
54		55
55	=head1 DESCRIPTION	56	=head1 DESCRIPTION
56		57
57	This module provides an interface to libev	58	This module provides an interface to libev
58	(L<http://software.schmorp.de/pkg/libev.html>).	59	(L<http://software.schmorp.de/pkg/libev.html>). While the documentation
		60	below is comprehensive, one might also consult the documentation of
		61	libev itself (L<http://pod.tst.eu/http://cvs.schmorp.de/libev/ev.pod> or
		62	F<perldoc EV::libev>) for more subtle details on watcher semantics or some
		63	discussion on the available backends, or how to force a specific backend
		64	with C<LIBEV_FLAGS>, or just about in any case because it has much more
		65	detailed information.
		66
		67	This module is very fast and scalable. It is actually so fast that you
		68	can use it through the L<AnyEvent> module, stay portable to other event
		69	loops (if you don't rely on any watcher types not available through it)
		70	and still be faster than with any other event loop currently supported in
		71	Perl.
		72
		73	=head2 MODULE EXPORTS
		74
		75	This module does not export any symbols.
59		76
60	=cut	77	=cut
61		78
62	package EV;	79	package EV;
63		80
64	use strict;	81	use common::sense;
65		82
66	BEGIN {	83	BEGIN {
67	our $VERSION = '0.5';	84	our $VERSION = '4.00';
68	use XSLoader;	85	use XSLoader;
69	XSLoader::load "EV", $VERSION;	86	XSLoader::load "EV", $VERSION;
70	}	87	}
71		88
72	@EV::Io::ISA =	89	@EV::IO::ISA =
73	@EV::Timer::ISA =	90	@EV::Timer::ISA =
74	@EV::Periodic::ISA =	91	@EV::Periodic::ISA =
75	@EV::Signal::ISA =	92	@EV::Signal::ISA =
		93	@EV::Child::ISA =
		94	@EV::Stat::ISA =
76	@EV::Idle::ISA =	95	@EV::Idle::ISA =
77	@EV::Prepare::ISA =	96	@EV::Prepare::ISA =
78	@EV::Check::ISA =	97	@EV::Check::ISA =
79	@EV::Child::ISA = "EV::Watcher";	98	@EV::Embed::ISA =
		99	@EV::Fork::ISA =
		100	@EV::Async::ISA =
		101	"EV::Watcher";
		102
		103	@EV::Loop::Default::ISA = "EV::Loop";
		104
		105	=head1 EVENT LOOPS
		106
		107	EV supports multiple event loops: There is a single "default event loop"
		108	that can handle everything including signals and child watchers, and any
		109	number of "dynamic event loops" that can use different backends (with
		110	various limitations), but no child and signal watchers.
		111
		112	You do not have to do anything to create the default event loop: When
		113	the module is loaded a suitable backend is selected on the premise of
		114	selecting a working backend (which for example rules out kqueue on most
		115	BSDs). Modules should, unless they have "special needs" always use the
		116	default loop as this is fastest (perl-wise), best supported by other
		117	modules (e.g. AnyEvent or Coro) and most portable event loop.
		118
		119	For specific programs you can create additional event loops dynamically.
		120
		121	If you want to take advantage of kqueue (which often works properly for
		122	sockets only) even though the default loop doesn't enable it, you can
		123	I<embed> a kqueue loop into the default loop: running the default loop
		124	will then also service the kqueue loop to some extent. See the example in
		125	the section about embed watchers for an example on how to achieve that.
		126
		127	=over 4
		128
		129	=item $loop = new EV::Loop [$flags]
		130
		131	Create a new event loop as per the specified flags. Please refer to
		132	the C<ev_loop_new ()> function description in the libev documentation
		133	(L<http://pod.tst.eu/http://cvs.schmorp.de/libev/ev.pod#GLOBAL_FUNCTIONS>,
		134	or locally-installed as F<EV::libev> manpage) for more info.
		135
		136	The loop will automatically be destroyed when it is no longer referenced
		137	by any watcher and the loop object goes out of scope.
		138
		139	If you are not embedding the loop, then Using C<EV::FLAG_FORKCHECK>
		140	is recommended, as only the default event loop is protected by this
		141	module. If you I<are> embedding this loop in the default loop, this is not
		142	necessary, as C<EV::embed> automatically does the right thing on fork.
		143
		144	=item $loop->loop_fork
		145
		146	Must be called after a fork in the child, before entering or continuing
		147	the event loop. An alternative is to use C<EV::FLAG_FORKCHECK> which calls
		148	this function automatically, at some performance loss (refer to the libev
		149	documentation).
		150
		151	=item $loop->loop_verify
		152
		153	Calls C<ev_verify> to make internal consistency checks (for debugging
		154	libev) and abort the program if any data structures were found to be
		155	corrupted.
		156
		157	=item $loop = EV::default_loop [$flags]
		158
		159	Return the default loop (which is a singleton object). Since this module
		160	already creates the default loop with default flags, specifying flags here
		161	will not have any effect unless you destroy the default loop first, which
		162	isn't supported. So in short: don't do it, and if you break it, you get to
		163	keep the pieces.
		164
		165	=back
		166
80		167
81	=head1 BASIC INTERFACE	168	=head1 BASIC INTERFACE
82		169
83	=over 4	170	=over 4
84		171
85	=item $EV::DIED	172	=item $EV::DIED
86		173
87	Must contain a reference to a function that is called when a callback	174	Must contain a reference to a function that is called when a callback
88	throws an exception (with $@ containing thr error). The default prints an	175	throws an exception (with $@ containing the error). The default prints an
89	informative message and continues.	176	informative message and continues.
90		177
91	If this callback throws an exception it will be silently ignored.	178	If this callback throws an exception it will be silently ignored.
92		179
		180	=item $flags = EV::supported_backends
		181
		182	=item $flags = EV::recommended_backends
		183
		184	=item $flags = EV::embeddable_backends
		185
		186	Returns the set (see C<EV::BACKEND_*> flags) of backends supported by this
		187	instance of EV, the set of recommended backends (supposed to be good) for
		188	this platform and the set of embeddable backends (see EMBED WATCHERS).
		189
		190	=item EV::sleep $seconds
		191
		192	Block the process for the given number of (fractional) seconds.
		193
93	=item $time = EV::time	194	=item $time = EV::time
94		195
95	Returns the current time in (fractional) seconds since the epoch.	196	Returns the current time in (fractional) seconds since the epoch.
96		197
97	=item $time = EV::now	198	=item $time = EV::now
98		199
		200	=item $time = $loop->now
		201
99	Returns the time the last event loop iteration has been started. This	202	Returns the time the last event loop iteration has been started. This
100	is the time that (relative) timers are based on, and refering to it is	203	is the time that (relative) timers are based on, and referring to it is
101	usually faster then calling EV::time.	204	usually faster then calling EV::time.
102		205
103	=item $method = EV::ev_method	206	=item EV::now_update
104		207
		208	=item $loop->now_update
		209
		210	Establishes the current time by querying the kernel, updating the time
		211	returned by C<EV::now> in the progress. This is a costly operation and
		212	is usually done automatically within C<EV::loop>.
		213
		214	This function is rarely useful, but when some event callback runs for a
		215	very long time without entering the event loop, updating libev's idea of
		216	the current time is a good idea.
		217
		218	=item EV::suspend
		219
		220	=item $loop->suspend
		221
		222	=item EV::resume
		223
		224	=item $loop->resume
		225
		226	These two functions suspend and resume a loop, for use when the loop is
		227	not used for a while and timeouts should not be processed.
		228
		229	A typical use case would be an interactive program such as a game: When
		230	the user presses C<^Z> to suspend the game and resumes it an hour later it
		231	would be best to handle timeouts as if no time had actually passed while
		232	the program was suspended. This can be achieved by calling C<suspend>
		233	in your C<SIGTSTP> handler, sending yourself a C<SIGSTOP> and calling
		234	C<resume> directly afterwards to resume timer processing.
		235
		236	Effectively, all C<timer> watchers will be delayed by the time spend
		237	between C<suspend> and C<resume>, and all C<periodic> watchers
		238	will be rescheduled (that is, they will lose any events that would have
		239	occured while suspended).
		240
		241	After calling C<suspend> you B<must not> call I<any> function on the given
		242	loop other than C<resume>, and you B<must not> call C<resume>
		243	without a previous call to C<suspend>.
		244
		245	Calling C<suspend>/C<resume> has the side effect of updating the event
		246	loop time (see C<now_update>).
		247
		248	=item $backend = EV::backend
		249
		250	=item $backend = $loop->backend
		251
105	Returns an integer describing the backend used by libev (EV::METHOD_SELECT	252	Returns an integer describing the backend used by libev (EV::BACKEND_SELECT
106	or EV::METHOD_EPOLL).	253	or EV::BACKEND_EPOLL).
107		254
108	=item EV::loop [$flags]	255	=item EV::loop [$flags]
109		256
		257	=item $loop->loop ([$flags])
		258
110	Begin checking for events and calling callbacks. It returns when a	259	Begin checking for events and calling callbacks. It returns when a
111	callback calls EV::loop_done.	260	callback calls EV::unloop.
112		261
113	The $flags argument can be one of the following:	262	The $flags argument can be one of the following:
114		263
115	0 as above	264	0 as above
116	EV::LOOP_ONESHOT block at most once (wait, but do not loop)	265	EV::LOOP_ONESHOT block at most once (wait, but do not loop)
117	EV::LOOP_NONBLOCK do not block at all (fetch/handle events but do not wait)	266	EV::LOOP_NONBLOCK do not block at all (fetch/handle events but do not wait)
118		267
119	=item EV::loop_done [$how]	268	=item EV::unloop [$how]
120		269
		270	=item $loop->unloop ([$how])
		271
121	When called with no arguments or an argument of 1, makes the innermost	272	When called with no arguments or an argument of EV::UNLOOP_ONE, makes the
122	call to EV::loop return.	273	innermost call to EV::loop return.
123		274
124	When called with an agrument of 2, all calls to EV::loop will return as	275	When called with an argument of EV::UNLOOP_ALL, all calls to EV::loop will return as
125	fast as possible.	276	fast as possible.
126		277
127	=back	278	=item $count = EV::loop_count
128		279
129	=head2 WATCHER	280	=item $count = $loop->loop_count
		281
		282	Return the number of times the event loop has polled for new
		283	events. Sometimes useful as a generation counter.
		284
		285	=item EV::once $fh_or_undef, $events, $timeout, $cb->($revents)
		286
		287	=item $loop->once ($fh_or_undef, $events, $timeout, $cb->($revents))
		288
		289	This function rolls together an I/O and a timer watcher for a single
		290	one-shot event without the need for managing a watcher object.
		291
		292	If C<$fh_or_undef> is a filehandle or file descriptor, then C<$events>
		293	must be a bitset containing either C<EV::READ>, C<EV::WRITE> or C<EV::READ
		294	\| EV::WRITE>, indicating the type of I/O event you want to wait for. If
		295	you do not want to wait for some I/O event, specify C<undef> for
		296	C<$fh_or_undef> and C<0> for C<$events>).
		297
		298	If timeout is C<undef> or negative, then there will be no
		299	timeout. Otherwise a EV::timer with this value will be started.
		300
		301	When an error occurs or either the timeout or I/O watcher triggers, then
		302	the callback will be called with the received event set (in general
		303	you can expect it to be a combination of C<EV::ERROR>, C<EV::READ>,
		304	C<EV::WRITE> and C<EV::TIMER>).
		305
		306	EV::once doesn't return anything: the watchers stay active till either
		307	of them triggers, then they will be stopped and freed, and the callback
		308	invoked.
		309
		310	=item EV::feed_fd_event ($fd, $revents)
		311
		312	=item $loop->feed_fd_event ($fd, $revents)
		313
		314	Feed an event on a file descriptor into EV. EV will react to this call as
		315	if the readyness notifications specified by C<$revents> (a combination of
		316	C<EV::READ> and C<EV::WRITE>) happened on the file descriptor C<$fd>.
		317
		318	=item EV::feed_signal_event ($signal)
		319
		320	Feed a signal event into EV. EV will react to this call as if the signal
		321	specified by C<$signal> had occured.
		322
		323	=item EV::set_io_collect_interval $time
		324
		325	=item $loop->set_io_collect_interval ($time)
		326
		327	=item EV::set_timeout_collect_interval $time
		328
		329	=item $loop->set_timeout_collect_interval ($time)
		330
		331	These advanced functions set the minimum block interval when polling for I/O events and the minimum
		332	wait interval for timer events. See the libev documentation at
		333	L<http://pod.tst.eu/http://cvs.schmorp.de/libev/ev.pod#FUNCTIONS_CONTROLLING_THE_EVENT_LOOP>
		334	(locally installed as F<EV::libev>) for a more detailed discussion.
		335
		336	=item $count = EV::pending_count
		337
		338	=item $count = $loop->pending_count
		339
		340	Returns the number of currently pending watchers.
		341
		342	=item EV::invoke_pending
		343
		344	=item $loop->invoke_pending
		345
		346	Invoke all currently pending watchers.
		347
		348	=back
		349
		350
		351	=head1 WATCHER OBJECTS
130		352
131	A watcher is an object that gets created to record your interest in some	353	A watcher is an object that gets created to record your interest in some
132	event. For instance, if you want to wait for STDIN to become readable, you	354	event. For instance, if you want to wait for STDIN to become readable, you
133	would create an EV::io watcher for that:	355	would create an EV::io watcher for that:
134		356
135	my $watcher = EV::io *STDIN, EV::READ, sub {	357	my $watcher = EV::io *STDIN, EV::READ, sub {
136	my ($watcher, $revents) = @_;	358	my ($watcher, $revents) = @_;
137	warn "yeah, STDIN should not be readable without blocking!\n"	359	warn "yeah, STDIN should now be readable without blocking!\n"
138	};	360	};
139		361
140	All watchers can be active (waiting for events) or inactive (paused). Only	362	All watchers can be active (waiting for events) or inactive (paused). Only
141	active watchers will have their callbacks invoked. All callbacks will be	363	active watchers will have their callbacks invoked. All callbacks will be
142	called with at least two arguments: the watcher and a bitmask of received	364	called with at least two arguments: the watcher and a bitmask of received
143	events.	365	events.
144		366
145	Each watcher type has its associated bit in revents, so you can use the	367	Each watcher type has its associated bit in revents, so you can use the
146	same callback for multiple watchers. The event mask is named after the	368	same callback for multiple watchers. The event mask is named after the
147	type, i..e. EV::child sets EV::CHILD, EV::prepare sets EV::PREPARE,	369	type, i.e. EV::child sets EV::CHILD, EV::prepare sets EV::PREPARE,
148	EV::periodic sets EV::PERIODIC and so on, with the exception of IO events	370	EV::periodic sets EV::PERIODIC and so on, with the exception of I/O events
149	(which can set both EV::READ and EV::WRITE bits), and EV::timer (which	371	(which can set both EV::READ and EV::WRITE bits).
150	uses EV::TIMEOUT).
151		372
152	In the rare case where one wants to create a watcher but not start it at	373	In the rare case where one wants to create a watcher but not start it at
153	the same time, each constructor has a variant with a trailing C<_ns> in	374	the same time, each constructor has a variant with a trailing C<_ns> in
154	its name, e.g. EV::io has a non-starting variant EV::io_ns and so on.	375	its name, e.g. EV::io has a non-starting variant EV::io_ns and so on.
155		376
…		…
159		380
160	Also, all methods changing some aspect of a watcher (->set, ->priority,	381	Also, all methods changing some aspect of a watcher (->set, ->priority,
161	->fh and so on) automatically stop and start it again if it is active,	382	->fh and so on) automatically stop and start it again if it is active,
162	which means pending events get lost.	383	which means pending events get lost.
163		384
164	=head2 WATCHER TYPES	385	=head2 COMMON WATCHER METHODS
165		386
166	Now lets move to the existing watcher types and asociated methods.	387	This section lists methods common to all watchers.
167
168	The following methods are available for all watchers. Then followes a
169	description of each watcher constructor (EV::io, EV::timer, EV::periodic,
170	EV::signal, EV::child, EV::idle, EV::prepare and EV::check), followed by
171	any type-specific methods (if any).
172		388
173	=over 4	389	=over 4
174		390
175	=item $w->start	391	=item $w->start
176		392
…		…
180		396
181	=item $w->stop	397	=item $w->stop
182		398
183	Stop a watcher if it is active. Also clear any pending events (events that	399	Stop a watcher if it is active. Also clear any pending events (events that
184	have been received but that didn't yet result in a callback invocation),	400	have been received but that didn't yet result in a callback invocation),
185	regardless of wether the watcher was active or not.	401	regardless of whether the watcher was active or not.
186		402
187	=item $bool = $w->is_active	403	=item $bool = $w->is_active
188		404
189	Returns true if the watcher is active, false otherwise.	405	Returns true if the watcher is active, false otherwise.
		406
		407	=item $current_data = $w->data
		408
		409	=item $old_data = $w->data ($new_data)
		410
		411	Queries a freely usable data scalar on the watcher and optionally changes
		412	it. This is a way to associate custom data with a watcher:
		413
		414	my $w = EV::timer 60, 0, sub {
		415	warn $_[0]->data;
		416	};
		417	$w->data ("print me!");
190		418
191	=item $current_cb = $w->cb	419	=item $current_cb = $w->cb
192		420
193	=item $old_cb = $w->cb ($new_cb)	421	=item $old_cb = $w->cb ($new_cb)
194		422
…		…
203	watchers with higher priority will be invoked first. The valid range of	431	watchers with higher priority will be invoked first. The valid range of
204	priorities lies between EV::MAXPRI (default 2) and EV::MINPRI (default	432	priorities lies between EV::MAXPRI (default 2) and EV::MINPRI (default
205	-2). If the priority is outside this range it will automatically be	433	-2). If the priority is outside this range it will automatically be
206	normalised to the nearest valid priority.	434	normalised to the nearest valid priority.
207		435
208	The default priority of any newly-created weatcher is 0.	436	The default priority of any newly-created watcher is 0.
209		437
		438	Note that the priority semantics have not yet been fleshed out and are
		439	subject to almost certain change.
		440
210	=item $w->trigger ($revents)	441	=item $w->invoke ($revents)
211		442
212	Call the callback now with the given event mask.	443	Call the callback now with the given event mask.
213		444
		445	=item $w->feed_event ($revents)
		446
		447	Feed some events on this watcher into EV. EV will react to this call as if
		448	the watcher had received the given C<$revents> mask.
		449
		450	=item $revents = $w->clear_pending
		451
		452	If the watcher is pending, this function clears its pending status and
		453	returns its C<$revents> bitset (as if its callback was invoked). If the
		454	watcher isn't pending it does nothing and returns C<0>.
		455
		456	=item $previous_state = $w->keepalive ($bool)
		457
		458	Normally, C<EV::loop> will return when there are no active watchers
		459	(which is a "deadlock" because no progress can be made anymore). This is
		460	convenient because it allows you to start your watchers (and your jobs),
		461	call C<EV::loop> once and when it returns you know that all your jobs are
		462	finished (or they forgot to register some watchers for their task :).
		463
		464	Sometimes, however, this gets in your way, for example when the module
		465	that calls C<EV::loop> (usually the main program) is not the same module
		466	as a long-living watcher (for example a DNS client module written by
		467	somebody else even). Then you might want any outstanding requests to be
		468	handled, but you would not want to keep C<EV::loop> from returning just
		469	because you happen to have this long-running UDP port watcher.
		470
		471	In this case you can clear the keepalive status, which means that even
		472	though your watcher is active, it won't keep C<EV::loop> from returning.
		473
		474	The initial value for keepalive is true (enabled), and you can change it
		475	any time.
		476
		477	Example: Register an I/O watcher for some UDP socket but do not keep the
		478	event loop from running just because of that watcher.
		479
		480	my $udp_socket = ...
		481	my $udp_watcher = EV::io $udp_socket, EV::READ, sub { ... };
		482	$udp_watcher->keepalive (0);
		483
		484	=item $loop = $w->loop
		485
		486	Return the loop that this watcher is attached to.
		487
		488	=back
		489
		490
		491	=head1 WATCHER TYPES
		492
		493	Each of the following subsections describes a single watcher type.
		494
		495	=head3 I/O WATCHERS - is this file descriptor readable or writable?
		496
		497	=over 4
214		498
215	=item $w = EV::io $fileno_or_fh, $eventmask, $callback	499	=item $w = EV::io $fileno_or_fh, $eventmask, $callback
216		500
217	=item $w = EV::io_ns $fileno_or_fh, $eventmask, $callback	501	=item $w = EV::io_ns $fileno_or_fh, $eventmask, $callback
218		502
		503	=item $w = $loop->io ($fileno_or_fh, $eventmask, $callback)
		504
		505	=item $w = $loop->io_ns ($fileno_or_fh, $eventmask, $callback)
		506
219	As long as the returned watcher object is alive, call the C<$callback>	507	As long as the returned watcher object is alive, call the C<$callback>
220	when the events specified in C<$eventmask>.	508	when at least one of events specified in C<$eventmask> occurs.
221		509
222	The $eventmask can be one or more of these constants ORed together:	510	The $eventmask can be one or more of these constants ORed together:
223		511
224	EV::READ wait until read() wouldn't block anymore	512	EV::READ wait until read() wouldn't block anymore
225	EV::WRITE wait until write() wouldn't block anymore	513	EV::WRITE wait until write() wouldn't block anymore
…		…
241		529
242	=item $old_eventmask = $w->events ($new_eventmask)	530	=item $old_eventmask = $w->events ($new_eventmask)
243		531
244	Returns the previously set event mask and optionally set a new one.	532	Returns the previously set event mask and optionally set a new one.
245		533
		534	=back
		535
		536
		537	=head3 TIMER WATCHERS - relative and optionally repeating timeouts
		538
		539	=over 4
246		540
247	=item $w = EV::timer $after, $repeat, $callback	541	=item $w = EV::timer $after, $repeat, $callback
248		542
249	=item $w = EV::timer_ns $after, $repeat, $callback	543	=item $w = EV::timer_ns $after, $repeat, $callback
250		544
251	Calls the callback after C<$after> seconds. If C<$repeat> is non-zero,	545	=item $w = $loop->timer ($after, $repeat, $callback)
252	the timer will be restarted (with the $repeat value as $after) after the	546
253	callback returns.	547	=item $w = $loop->timer_ns ($after, $repeat, $callback)
		548
		549	Calls the callback after C<$after> seconds (which may be fractional). If
		550	C<$repeat> is non-zero, the timer will be restarted (with the $repeat
		551	value as $after) after the callback returns.
254		552
255	This means that the callback would be called roughly after C<$after>	553	This means that the callback would be called roughly after C<$after>
256	seconds, and then every C<$repeat> seconds. "Roughly" because the time of	554	seconds, and then every C<$repeat> seconds. The timer does his best not
257	callback processing is not taken into account, so the timer will slowly	555	to drift, but it will not invoke the timer more often then once per event
258	drift. If that isn't acceptable, look at EV::periodic.	556	loop iteration, and might drift in other cases. If that isn't acceptable,
		557	look at EV::periodic, which can provide long-term stable timers.
259		558
260	The timer is based on a monotonic clock, that is if somebody is sitting	559	The timer is based on a monotonic clock, that is, if somebody is sitting
261	in front of the machine while the timer is running and changes the system	560	in front of the machine while the timer is running and changes the system
262	clock, the timer will nevertheless run (roughly) the same time.	561	clock, the timer will nevertheless run (roughly) the same time.
263		562
264	The C<timer_ns> variant doesn't start (activate) the newly created watcher.	563	The C<timer_ns> variant doesn't start (activate) the newly created watcher.
265		564
266	=item $w->set ($after, $repeat)	565	=item $w->set ($after, $repeat)
267		566
268	Reconfigures the watcher, see the constructor above for details. Can be at	567	Reconfigures the watcher, see the constructor above for details. Can be called at
269	any time.	568	any time.
270		569
271	=item $w->again	570	=item $w->again
272		571
273	Similar to the C<start> method, but has special semantics for repeating timers:	572	Similar to the C<start> method, but has special semantics for repeating timers:
		573
		574	If the timer is active and non-repeating, it will be stopped.
274		575
275	If the timer is active and repeating, reset the timeout to occur	576	If the timer is active and repeating, reset the timeout to occur
276	C<$repeat> seconds after now.	577	C<$repeat> seconds after now.
277		578
278	If the timer is active and non-repeating, it will be stopped.
279
280	If the timer is in active and repeating, start it.	579	If the timer is inactive and repeating, start it using the repeat value.
281		580
282	Otherwise do nothing.	581	Otherwise do nothing.
283		582
284	This behaviour is useful when you have a timeout for some IO	583	This behaviour is useful when you have a timeout for some IO
285	operation. You create a timer object with the same value for C<$after> and	584	operation. You create a timer object with the same value for C<$after> and
286	C<$repeat>, and then, in the read/write watcher, run the C<again> method	585	C<$repeat>, and then, in the read/write watcher, run the C<again> method
287	on the timeout.	586	on the timeout.
288		587
		588	=back
289		589
		590
		591	=head3 PERIODIC WATCHERS - to cron or not to cron?
		592
		593	=over 4
		594
290	=item $w = EV::periodic $at, $interval, $callback	595	=item $w = EV::periodic $at, $interval, $reschedule_cb, $callback
291		596
292	=item $w = EV::periodic_ns $at, $interval, $callback	597	=item $w = EV::periodic_ns $at, $interval, $reschedule_cb, $callback
293		598
294	Similar to EV::timer, but the time is given as an absolute point in time	599	=item $w = $loop->periodic ($at, $interval, $reschedule_cb, $callback)
295	(C<$at>), plus an optional C<$interval>.
296		600
297	If the C<$interval> is zero, then the callback will be called at the time	601	=item $w = $loop->periodic_ns ($at, $interval, $reschedule_cb, $callback)
298	C<$at> if that is in the future, or as soon as possible if it is in the
299	past. It will not automatically repeat.
300		602
301	If the C<$interval> is nonzero, then the watcher will always be scheduled	603	Similar to EV::timer, but is not based on relative timeouts but on
302	to time out at the next C<$at + N * $interval> time.	604	absolute times. Apart from creating "simple" timers that trigger "at" the
		605	specified time, it can also be used for non-drifting absolute timers and
		606	more complex, cron-like, setups that are not adversely affected by time
		607	jumps (i.e. when the system clock is changed by explicit date -s or other
		608	means such as ntpd). It is also the most complex watcher type in EV.
303		609
304	This can be used to schedule a callback to run at very regular intervals,	610	It has three distinct "modes":
305	as long as the processing time is less then the interval (otherwise	611
306	obviously events will be skipped).	612	=over 4
		613
		614	=item * absolute timer ($interval = $reschedule_cb = 0)
		615
		616	This time simply fires at the wallclock time C<$at> and doesn't repeat. It
		617	will not adjust when a time jump occurs, that is, if it is to be run
		618	at January 1st 2011 then it will run when the system time reaches or
		619	surpasses this time.
		620
		621	=item * repeating interval timer ($interval > 0, $reschedule_cb = 0)
		622
		623	In this mode the watcher will always be scheduled to time out at the
		624	next C<$at + N * $interval> time (for some integer N) and then repeat,
		625	regardless of any time jumps.
		626
		627	This can be used to create timers that do not drift with respect to system
		628	time:
		629
		630	my $hourly = EV::periodic 0, 3600, 0, sub { print "once/hour\n" };
		631
		632	That doesn't mean there will always be 3600 seconds in between triggers,
		633	but only that the the clalback will be called when the system time shows a
		634	full hour (UTC).
307		635
308	Another way to think about it (for the mathematically inclined) is that	636	Another way to think about it (for the mathematically inclined) is that
309	EV::periodic will try to run the callback at the next possible time where	637	EV::periodic will try to run the callback in this mode at the next
310	C<$time = $at (mod $interval)>, regardless of any time jumps.	638	possible time where C<$time = $at (mod $interval)>, regardless of any time
		639	jumps.
311		640
312	This periodic timer is based on "wallclock time", that is, if the clock	641	=item * manual reschedule mode ($reschedule_cb = coderef)
313	changes (C<ntp>, C<date -s> etc.), then the timer will nevertheless run at	642
314	the specified time. This means it will never drift (it might jitter, but	643	In this mode $interval and $at are both being ignored. Instead, each
315	it will not drift).	644	time the periodic watcher gets scheduled, the reschedule callback
		645	($reschedule_cb) will be called with the watcher as first, and the current
		646	time as second argument.
		647
		648	I<This callback MUST NOT stop or destroy this or any other periodic
		649	watcher, ever, and MUST NOT call any event loop functions or methods>. If
		650	you need to stop it, return 1e30 and stop it afterwards. You may create
		651	and start a C<EV::prepare> watcher for this task.
		652
		653	It must return the next time to trigger, based on the passed time value
		654	(that is, the lowest time value larger than or equal to to the second
		655	argument). It will usually be called just before the callback will be
		656	triggered, but might be called at other times, too.
		657
		658	This can be used to create very complex timers, such as a timer that
		659	triggers on each midnight, local time (actually 24 hours after the last
		660	midnight, to keep the example simple. If you know a way to do it correctly
		661	in about the same space (without requiring elaborate modules), drop me a
		662	note :):
		663
		664	my $daily = EV::periodic 0, 0, sub {
		665	my ($w, $now) = @_;
		666
		667	use Time::Local ();
		668	my (undef, undef, undef, $d, $m, $y) = localtime $now;
		669	86400 + Time::Local::timelocal 0, 0, 0, $d, $m, $y
		670	}, sub {
		671	print "it's midnight or likely shortly after, now\n";
		672	};
		673
		674	=back
316		675
317	The C<periodic_ns> variant doesn't start (activate) the newly created watcher.	676	The C<periodic_ns> variant doesn't start (activate) the newly created watcher.
318		677
319	=item $w->set ($at, $interval)	678	=item $w->set ($at, $interval, $reschedule_cb)
320		679
321	Reconfigures the watcher, see the constructor above for details. Can be at	680	Reconfigures the watcher, see the constructor above for details. Can be called at
322	any time.	681	any time.
323		682
		683	=item $w->again
		684
		685	Simply stops and starts the watcher again.
		686
		687	=item $time = $w->at
		688
		689	Return the time that the watcher is expected to trigger next.
		690
		691	=back
		692
		693
		694	=head3 SIGNAL WATCHERS - signal me when a signal gets signalled!
		695
		696	=over 4
324		697
325	=item $w = EV::signal $signal, $callback	698	=item $w = EV::signal $signal, $callback
326		699
327	=item $w = EV::signal_ns $signal, $callback	700	=item $w = EV::signal_ns $signal, $callback
328		701
		702	=item $w = $loop->signal ($signal, $callback)
		703
		704	=item $w = $loop->signal_ns ($signal, $callback)
		705
329	Call the callback when $signal is received (the signal can be specified	706	Call the callback when $signal is received (the signal can be specified by
330	by number or by name, just as with kill or %SIG).	707	number or by name, just as with C<kill> or C<%SIG>).
		708
		709	Only one event loop can grab a given signal - attempting to grab the same
		710	signal from two EV loops will crash the program immediately or cause data
		711	corruption.
331		712
332	EV will grab the signal for the process (the kernel only allows one	713	EV will grab the signal for the process (the kernel only allows one
333	component to receive a signal at a time) when you start a signal watcher,	714	component to receive a signal at a time) when you start a signal watcher,
334	and removes it again when you stop it. Perl does the same when you	715	and removes it again when you stop it. Perl does the same when you
335	add/remove callbacks to %SIG, so watch out.	716	add/remove callbacks to C<%SIG>, so watch out.
336		717
337	You can have as many signal watchers per signal as you want.	718	You can have as many signal watchers per signal as you want.
338		719
339	The C<signal_ns> variant doesn't start (activate) the newly created watcher.	720	The C<signal_ns> variant doesn't start (activate) the newly created watcher.
340		721
341	=item $w->set ($signal)	722	=item $w->set ($signal)
342		723
343	Reconfigures the watcher, see the constructor above for details. Can be at	724	Reconfigures the watcher, see the constructor above for details. Can be
344	any time.	725	called at any time.
345		726
346	=item $current_signum = $w->signal	727	=item $current_signum = $w->signal
347		728
348	=item $old_signum = $w->signal ($new_signal)	729	=item $old_signum = $w->signal ($new_signal)
349		730
350	Returns the previously set signal (always as a number not name) and	731	Returns the previously set signal (always as a number not name) and
351	optionally set a new one.	732	optionally set a new one.
352		733
		734	=back
353		735
		736
		737	=head3 CHILD WATCHERS - watch out for process status changes
		738
		739	=over 4
		740
354	=item $w = EV::child $pid, $callback	741	=item $w = EV::child $pid, $trace, $callback
355		742
356	=item $w = EV::child_ns $pid, $callback	743	=item $w = EV::child_ns $pid, $trace, $callback
		744
		745	=item $w = $loop->child ($pid, $trace, $callback)
		746
		747	=item $w = $loop->child_ns ($pid, $trace, $callback)
357		748
358	Call the callback when a status change for pid C<$pid> (or any pid	749	Call the callback when a status change for pid C<$pid> (or any pid
359	if C<$pid> is 0) has been received. More precisely: when the process	750	if C<$pid> is 0) has been received (a status change happens when the
		751	process terminates or is killed, or, when trace is true, additionally when
		752	it is stopped or continued). More precisely: when the process receives
360	receives a SIGCHLD, EV will fetch the outstanding exit/wait status for all	753	a C<SIGCHLD>, EV will fetch the outstanding exit/wait status for all
361	changed/zombie children and call the callback.	754	changed/zombie children and call the callback.
362		755
363	You can access both status and pid by using the C<rstatus> and C<rpid>	756	It is valid (and fully supported) to install a child watcher after a child
364	methods on the watcher object.	757	has exited but before the event loop has started its next iteration (for
		758	example, first you C<fork>, then the new child process might exit, and
		759	only then do you install a child watcher in the parent for the new pid).
365		760
		761	You can access both exit (or tracing) status and pid by using the
		762	C<rstatus> and C<rpid> methods on the watcher object.
		763
366	You can have as many pid watchers per pid as you want.	764	You can have as many pid watchers per pid as you want, they will all be
		765	called.
367		766
368	The C<child_ns> variant doesn't start (activate) the newly created watcher.	767	The C<child_ns> variant doesn't start (activate) the newly created watcher.
369		768
370	=item $w->set ($pid)	769	=item $w->set ($pid, $trace)
371		770
372	Reconfigures the watcher, see the constructor above for details. Can be at	771	Reconfigures the watcher, see the constructor above for details. Can be called at
373	any time.	772	any time.
374		773
375	=item $current_pid = $w->pid	774	=item $current_pid = $w->pid
376
377	=item $old_pid = $w->pid ($new_pid)
378		775
379	Returns the previously set process id and optionally set a new one.	776	Returns the previously set process id and optionally set a new one.
380		777
381	=item $exit_status = $w->rstatus	778	=item $exit_status = $w->rstatus
382		779
…		…
386	=item $pid = $w->rpid	783	=item $pid = $w->rpid
387		784
388	Return the pid of the awaited child (useful when you have installed a	785	Return the pid of the awaited child (useful when you have installed a
389	watcher for all pids).	786	watcher for all pids).
390		787
		788	=back
		789
		790
		791	=head3 STAT WATCHERS - did the file attributes just change?
		792
		793	=over 4
		794
		795	=item $w = EV::stat $path, $interval, $callback
		796
		797	=item $w = EV::stat_ns $path, $interval, $callback
		798
		799	=item $w = $loop->stat ($path, $interval, $callback)
		800
		801	=item $w = $loop->stat_ns ($path, $interval, $callback)
		802
		803	Call the callback when a file status change has been detected on
		804	C<$path>. The C<$path> does not need to exist, changing from "path exists"
		805	to "path does not exist" is a status change like any other.
		806
		807	The C<$interval> is a recommended polling interval for systems where
		808	OS-supported change notifications don't exist or are not supported. If
		809	you use C<0> then an unspecified default is used (which is highly
		810	recommended!), which is to be expected to be around five seconds usually.
		811
		812	This watcher type is not meant for massive numbers of stat watchers,
		813	as even with OS-supported change notifications, this can be
		814	resource-intensive.
		815
		816	The C<stat_ns> variant doesn't start (activate) the newly created watcher.
		817
		818	=item ... = $w->stat
		819
		820	This call is very similar to the perl C<stat> built-in: It stats (using
		821	C<lstat>) the path specified in the watcher and sets perls stat cache (as
		822	well as EV's idea of the current stat values) to the values found.
		823
		824	In scalar context, a boolean is return indicating success or failure of
		825	the stat. In list context, the same 13-value list as with stat is returned
		826	(except that the blksize and blocks fields are not reliable).
		827
		828	In the case of an error, errno is set to C<ENOENT> (regardless of the
		829	actual error value) and the C<nlink> value is forced to zero (if the stat
		830	was successful then nlink is guaranteed to be non-zero).
		831
		832	See also the next two entries for more info.
		833
		834	=item ... = $w->attr
		835
		836	Just like C<< $w->stat >>, but without the initial stat'ing: this returns
		837	the values most recently detected by EV. See the next entry for more info.
		838
		839	=item ... = $w->prev
		840
		841	Just like C<< $w->stat >>, but without the initial stat'ing: this returns
		842	the previous set of values, before the change.
		843
		844	That is, when the watcher callback is invoked, C<< $w->prev >> will be set
		845	to the values found I<before> a change was detected, while C<< $w->attr >>
		846	returns the values found leading to the change detection. The difference (if any)
		847	between C<prev> and C<attr> is what triggered the callback.
		848
		849	If you did something to the filesystem object and do not want to trigger
		850	yet another change, you can call C<stat> to update EV's idea of what the
		851	current attributes are.
		852
		853	=item $w->set ($path, $interval)
		854
		855	Reconfigures the watcher, see the constructor above for details. Can be
		856	called at any time.
		857
		858	=item $current_path = $w->path
		859
		860	=item $old_path = $w->path ($new_path)
		861
		862	Returns the previously set path and optionally set a new one.
		863
		864	=item $current_interval = $w->interval
		865
		866	=item $old_interval = $w->interval ($new_interval)
		867
		868	Returns the previously set interval and optionally set a new one. Can be
		869	used to query the actual interval used.
		870
		871	=back
		872
		873
		874	=head3 IDLE WATCHERS - when you've got nothing better to do...
		875
		876	=over 4
391		877
392	=item $w = EV::idle $callback	878	=item $w = EV::idle $callback
393		879
394	=item $w = EV::idle_ns $callback	880	=item $w = EV::idle_ns $callback
395		881
396	Call the callback when there are no pending io, timer/periodic, signal or	882	=item $w = $loop->idle ($callback)
397	child events, i.e. when the process is idle.	883
		884	=item $w = $loop->idle_ns ($callback)
		885
		886	Call the callback when there are no other pending watchers of the same or
		887	higher priority (excluding check, prepare and other idle watchers of the
		888	same or lower priority, of course). They are called idle watchers because
		889	when the watcher is the highest priority pending event in the process, the
		890	process is considered to be idle at that priority.
		891
		892	If you want a watcher that is only ever called when I<no> other events are
		893	outstanding you have to set the priority to C<EV::MINPRI>.
398		894
399	The process will not block as long as any idle watchers are active, and	895	The process will not block as long as any idle watchers are active, and
400	they will be called repeatedly until stopped.	896	they will be called repeatedly until stopped.
401		897
		898	For example, if you have idle watchers at priority C<0> and C<1>, and
		899	an I/O watcher at priority C<0>, then the idle watcher at priority C<1>
		900	and the I/O watcher will always run when ready. Only when the idle watcher
		901	at priority C<1> is stopped and the I/O watcher at priority C<0> is not
		902	pending with the C<0>-priority idle watcher be invoked.
		903
402	The C<idle_ns> variant doesn't start (activate) the newly created watcher.	904	The C<idle_ns> variant doesn't start (activate) the newly created watcher.
403		905
		906	=back
		907
		908
		909	=head3 PREPARE WATCHERS - customise your event loop!
		910
		911	=over 4
404		912
405	=item $w = EV::prepare $callback	913	=item $w = EV::prepare $callback
406		914
407	=item $w = EV::prepare_ns $callback	915	=item $w = EV::prepare_ns $callback
		916
		917	=item $w = $loop->prepare ($callback)
		918
		919	=item $w = $loop->prepare_ns ($callback)
408		920
409	Call the callback just before the process would block. You can still	921	Call the callback just before the process would block. You can still
410	create/modify any watchers at this point.	922	create/modify any watchers at this point.
411		923
412	See the EV::check watcher, below, for explanations and an example.	924	See the EV::check watcher, below, for explanations and an example.
413		925
414	The C<prepare_ns> variant doesn't start (activate) the newly created watcher.	926	The C<prepare_ns> variant doesn't start (activate) the newly created watcher.
415		927
		928	=back
		929
		930
		931	=head3 CHECK WATCHERS - customise your event loop even more!
		932
		933	=over 4
416		934
417	=item $w = EV::check $callback	935	=item $w = EV::check $callback
418		936
419	=item $w = EV::check_ns $callback	937	=item $w = EV::check_ns $callback
		938
		939	=item $w = $loop->check ($callback)
		940
		941	=item $w = $loop->check_ns ($callback)
420		942
421	Call the callback just after the process wakes up again (after it has	943	Call the callback just after the process wakes up again (after it has
422	gathered events), but before any other callbacks have been invoked.	944	gathered events), but before any other callbacks have been invoked.
423		945
424	This is used to integrate other event-based software into the EV	946	This is used to integrate other event-based software into the EV
…		…
432	# do nothing unless active	954	# do nothing unless active
433	$dispatcher->{_event_queue_h}	955	$dispatcher->{_event_queue_h}
434	or return;	956	or return;
435		957
436	# make the dispatcher handle any outstanding stuff	958	# make the dispatcher handle any outstanding stuff
		959	... not shown
437		960
438	# create an IO watcher for each and every socket	961	# create an I/O watcher for each and every socket
439	@snmp_watcher = (	962	@snmp_watcher = (
440	(map { EV::io $_, EV::READ, sub { } }	963	(map { EV::io $_, EV::READ, sub { } }
441	keys %{ $dispatcher->{_descriptors} }),	964	keys %{ $dispatcher->{_descriptors} }),
		965
		966	EV::timer +($event->[Net::SNMP::Dispatcher::_ACTIVE]
		967	? $event->[Net::SNMP::Dispatcher::_TIME] - EV::now : 0),
		968	0, sub { },
442	);	969	);
443
444	# if there are any timeouts, also create a timer
445	push @snmp_watcher, EV::timer $event->[Net::SNMP::Dispatcher::_TIME] - EV::now, 0, sub { }
446	if $event->[Net::SNMP::Dispatcher::_ACTIVE];
447	};	970	};
448		971
449	The callbacks are irrelevant, the only purpose of those watchers is	972	The callbacks are irrelevant (and are not even being called), the
450	to wake up the process as soon as one of those events occurs (socket	973	only purpose of those watchers is to wake up the process as soon as
451	readable, or timer timed out). The corresponding EV::check watcher will then	974	one of those events occurs (socket readable, or timer timed out). The
452	clean up:	975	corresponding EV::check watcher will then clean up:
453		976
454	our $snmp_check = EV::check sub {	977	our $snmp_check = EV::check sub {
455	# destroy all watchers	978	# destroy all watchers
456	@snmp_watcher = ();	979	@snmp_watcher = ();
457		980
458	# make the dispatcher handle any new stuff	981	# make the dispatcher handle any new stuff
		982	... not shown
459	};	983	};
460		984
461	The callbacks of the created watchers will not be called as the watchers	985	The callbacks of the created watchers will not be called as the watchers
462	are destroyed before this cna happen (remember EV::check gets called	986	are destroyed before this can happen (remember EV::check gets called
463	first).	987	first).
464		988
465	The C<check_ns> variant doesn't start (activate) the newly created watcher.	989	The C<check_ns> variant doesn't start (activate) the newly created watcher.
466		990
467	=back	991	=back
468		992
		993
		994	=head3 FORK WATCHERS - the audacity to resume the event loop after a fork
		995
		996	Fork watchers are called when a C<fork ()> was detected. The invocation
		997	is done before the event loop blocks next and before C<check> watchers
		998	are being called, and only in the child after the fork.
		999
		1000	=over 4
		1001
		1002	=item $w = EV::fork $callback
		1003
		1004	=item $w = EV::fork_ns $callback
		1005
		1006	=item $w = $loop->fork ($callback)
		1007
		1008	=item $w = $loop->fork_ns ($callback)
		1009
		1010	Call the callback before the event loop is resumed in the child process
		1011	after a fork.
		1012
		1013	The C<fork_ns> variant doesn't start (activate) the newly created watcher.
		1014
		1015	=back
		1016
		1017
		1018	=head3 EMBED WATCHERS - when one backend isn't enough...
		1019
		1020	This is a rather advanced watcher type that lets you embed one event loop
		1021	into another (currently only IO events are supported in the embedded
		1022	loop, other types of watchers might be handled in a delayed or incorrect
		1023	fashion and must not be used).
		1024
		1025	See the libev documentation at
		1026	L<http://pod.tst.eu/http://cvs.schmorp.de/libev/ev.pod#code_ev_embed_code_when_one_backend_>
		1027	(locally installed as F<EV::libev>) for more details.
		1028
		1029	In short, this watcher is most useful on BSD systems without working
		1030	kqueue to still be able to handle a large number of sockets:
		1031
		1032	my $socket_loop;
		1033
		1034	# check wether we use SELECT or POLL _and_ KQUEUE is supported
		1035	if (
		1036	(EV::backend & (EV::BACKEND_POLL \| EV::BACKEND_SELECT))
		1037	&& (EV::supported_backends & EV::embeddable_backends & EV::BACKEND_KQUEUE)
		1038	) {
		1039	# use kqueue for sockets
		1040	$socket_loop = new EV::Loop EV::BACKEND_KQUEUE \| EV::FLAG_NOENV;
		1041	}
		1042
		1043	# use the default loop otherwise
		1044	$socket_loop \|\|= EV::default_loop;
		1045
		1046	=over 4
		1047
		1048	=item $w = EV::embed $otherloop[, $callback]
		1049
		1050	=item $w = EV::embed_ns $otherloop[, $callback]
		1051
		1052	=item $w = $loop->embed ($otherloop[, $callback])
		1053
		1054	=item $w = $loop->embed_ns ($otherloop[, $callback])
		1055
		1056	Call the callback when the embedded event loop (C<$otherloop>) has any
		1057	I/O activity. The C<$callback> is optional: if it is missing, then the
		1058	embedded event loop will be managed automatically (which is recommended),
		1059	otherwise you have to invoke C<sweep> yourself.
		1060
		1061	The C<embed_ns> variant doesn't start (activate) the newly created watcher.
		1062
		1063	=back
		1064
		1065	=head3 ASYNC WATCHERS - how to wake up another event loop
		1066
		1067	Async watchers are provided by EV, but have little use in perl directly,
		1068	as perl neither supports threads running in parallel nor direct access to
		1069	signal handlers or other contexts where they could be of value.
		1070
		1071	It is, however, possible to use them from the XS level.
		1072
		1073	Please see the libev documentation for further details.
		1074
		1075	=over 4
		1076
		1077	=item $w = EV::async $callback
		1078
		1079	=item $w = EV::async_ns $callback
		1080
		1081	=item $w->send
		1082
		1083	=item $bool = $w->async_pending
		1084
		1085	=back
		1086
		1087
		1088	=head1 PERL SIGNALS
		1089
		1090	While Perl signal handling (C<%SIG>) is not affected by EV, the behaviour
		1091	with EV is as the same as any other C library: Perl-signals will only be
		1092	handled when Perl runs, which means your signal handler might be invoked
		1093	only the next time an event callback is invoked.
		1094
		1095	The solution is to use EV signal watchers (see C<EV::signal>), which will
		1096	ensure proper operations with regards to other event watchers.
		1097
		1098	If you cannot do this for whatever reason, you can also force a watcher
		1099	to be called on every event loop iteration by installing a C<EV::check>
		1100	watcher:
		1101
		1102	my $async_check = EV::check sub { };
		1103
		1104	This ensures that perl gets into control for a short time to handle any
		1105	pending signals, and also ensures (slightly) slower overall operation.
		1106
469	=head1 THREADS	1107	=head1 ITHREADS
470		1108
471	Threads are not supported by this in any way. Perl pseudo-threads is evil	1109	Ithreads are not supported by this module in any way. Perl pseudo-threads
472	stuff and must die.	1110	is evil stuff and must die. Real threads as provided by Coro are fully
		1111	supported (and enhanced support is available via L<Coro::EV>).
		1112
		1113	=head1 FORK
		1114
		1115	Most of the "improved" event delivering mechanisms of modern operating
		1116	systems have quite a few problems with fork(2) (to put it bluntly: it is
		1117	not supported and usually destructive). Libev makes it possible to work
		1118	around this by having a function that recreates the kernel state after
		1119	fork in the child.
		1120
		1121	On non-win32 platforms, this module requires the pthread_atfork
		1122	functionality to do this automatically for you. This function is quite
		1123	buggy on most BSDs, though, so YMMV. The overhead for this is quite
		1124	negligible, because everything the function currently does is set a flag
		1125	that is checked only when the event loop gets used the next time, so when
		1126	you do fork but not use EV, the overhead is minimal.
		1127
		1128	On win32, there is no notion of fork so all this doesn't apply, of course.
473		1129
474	=cut	1130	=cut
475		1131
476	our $DIED = sub {	1132	our $DIED = sub {
477	warn "EV: error in callback (ignoring): $@";	1133	warn "EV: error in callback (ignoring): $@";
478	};	1134	};
479		1135
480	init	1136	default_loop
481	or die 'EV: cannot initialise libev backend. bad $ENV{LIBEV_METHODS}?';	1137	or die 'EV: cannot initialise libev backend. bad $ENV{LIBEV_FLAGS}?';
482
483	push @AnyEvent::REGISTRY, [EV => "EV::AnyEvent"];
484		1138
485	1;	1139	1;
486		1140
487	=head1 SEE ALSO	1141	=head1 SEE ALSO
488		1142
489	L<EV::DNS>, L<EV::AnyEvent>.	1143	L<EV::ADNS> (asynchronous DNS), L<Glib::EV> (makes Glib/Gtk2 use EV as
		1144	event loop), L<EV::Glib> (embed Glib into EV), L<Coro::EV> (efficient
		1145	coroutines with EV), L<Net::SNMP::EV> (asynchronous SNMP), L<AnyEvent> for
		1146	event-loop agnostic and portable event driven programming.
490		1147
491	=head1 AUTHOR	1148	=head1 AUTHOR
492		1149
493	Marc Lehmann <schmorp@schmorp.de>	1150	Marc Lehmann <schmorp@schmorp.de>
494	http://home.schmorp.de/	1151	http://home.schmorp.de/
495		1152
496	=cut	1153	=cut
497		1154

Diff Legend

-–
+Removed lines
-+
+Added lines
-<
+Changed lines
->
+Changed lines

Comparing EV/EV.pm (file contents): Revision 1.27 by root, Sat Nov 3 09:19:58 2007 UTC vs. Revision 1.126 by root, Tue Mar 16 17:11:48 2010 UTC

Diff Legend

Comparing EV/EV.pm (file contents):
Revision 1.27 by root, Sat Nov 3 09:19:58 2007 UTC vs.
Revision 1.126 by root, Tue Mar 16 17:11:48 2010 UTC