ViewVC Help

View File | Revision Log | Show Annotations | Download File

/cvs/AnyEvent/lib/AnyEvent.pm

Comparing AnyEvent/lib/AnyEvent.pm (file contents):
Revision 1.144 by root, Thu May 29 00:14:35 2008 UTC vs.
Revision 1.370 by root, Wed Aug 17 02:50:35 2011 UTC

1	=head1 => NAME	1	=head1 NAME
2		2
3	AnyEvent - provide framework for multiple event loops	3	AnyEvent - the DBI of event loop programming
4		4
5	EV, Event, Glib, Tk, Perl, Event::Lib, Qt, POE - various supported event loops	5	EV, Event, Glib, Tk, Perl, Event::Lib, Irssi, rxvt-unicode, IO::Async, Qt,
		6	FLTK and POE are various supported event loops/environments.
6		7
7	=head1 SYNOPSIS	8	=head1 SYNOPSIS
8		9
9	use AnyEvent;	10	use AnyEvent;
10		11
		12	# if you prefer function calls, look at the AE manpage for
		13	# an alternative API.
		14
		15	# file handle or descriptor readable
11	my $w = AnyEvent->io (fh => $fh, poll => "r|w", cb => sub {	16	my $w = AnyEvent->io (fh => $fh, poll => "r", cb => sub { ... });
		17
		18	# one-shot or repeating timers
		19	my $w = AnyEvent->timer (after => $seconds, cb => sub { ... });
		20	my $w = AnyEvent->timer (after => $seconds, interval => $seconds, cb => ...);
		21
		22	print AnyEvent->now; # prints current event loop time
		23	print AnyEvent->time; # think Time::HiRes::time or simply CORE::time.
		24
		25	# POSIX signal
		26	my $w = AnyEvent->signal (signal => "TERM", cb => sub { ... });
		27
		28	# child process exit
		29	my $w = AnyEvent->child (pid => $pid, cb => sub {
		30	my ($pid, $status) = @_;
12	...	31	...
13	});	32	});
14		33
15	my $w = AnyEvent->timer (after => $seconds, cb => sub {	34	# called when event loop idle (if applicable)
16	...	35	my $w = AnyEvent->idle (cb => sub { ... });
17	});
18		36
19	my $w = AnyEvent->condvar; # stores whether a condition was flagged	37	my $w = AnyEvent->condvar; # stores whether a condition was flagged
20	$w->send; # wake up current and all future recv's	38	$w->send; # wake up current and all future recv's
21	$w->recv; # enters "main loop" till $condvar gets ->send	39	$w->recv; # enters "main loop" till $condvar gets ->send
		40	# use a condvar in callback mode:
		41	$w->cb (sub { $_[0]->recv });
		42
		43	=head1 INTRODUCTION/TUTORIAL
		44
		45	This manpage is mainly a reference manual. If you are interested
		46	in a tutorial or some gentle introduction, have a look at the
		47	L<AnyEvent::Intro> manpage.
		48
		49	=head1 SUPPORT
		50
		51	An FAQ document is available as L<AnyEvent::FAQ>.
		52
		53	There also is a mailinglist for discussing all things AnyEvent, and an IRC
		54	channel, too.
		55
		56	See the AnyEvent project page at the B<Schmorpforge Ta-Sa Software
		57	Repository>, at L<http://anyevent.schmorp.de>, for more info.
22		58
23	=head1 WHY YOU SHOULD USE THIS MODULE (OR NOT)	59	=head1 WHY YOU SHOULD USE THIS MODULE (OR NOT)
24		60
25	Glib, POE, IO::Async, Event... CPAN offers event models by the dozen	61	Glib, POE, IO::Async, Event... CPAN offers event models by the dozen
26	nowadays. So what is different about AnyEvent?	62	nowadays. So what is different about AnyEvent?
27		63
28	Executive Summary: AnyEvent is I<compatible>, AnyEvent is I<free of	64	Executive Summary: AnyEvent is I<compatible>, AnyEvent is I<free of
29	policy> and AnyEvent is I<small and efficient>.	65	policy> and AnyEvent is I<small and efficient>.
30		66
31	First and foremost, I<AnyEvent is not an event model> itself, it only	67	First and foremost, I<AnyEvent is not an event model> itself, it only
32	interfaces to whatever event model the main program happens to use in a	68	interfaces to whatever event model the main program happens to use, in a
33	pragmatic way. For event models and certain classes of immortals alike,	69	pragmatic way. For event models and certain classes of immortals alike,
34	the statement "there can only be one" is a bitter reality: In general,	70	the statement "there can only be one" is a bitter reality: In general,
35	only one event loop can be active at the same time in a process. AnyEvent	71	only one event loop can be active at the same time in a process. AnyEvent
36	helps hiding the differences between those event loops.	72	cannot change this, but it can hide the differences between those event
		73	loops.
37		74
38	The goal of AnyEvent is to offer module authors the ability to do event	75	The goal of AnyEvent is to offer module authors the ability to do event
39	programming (waiting for I/O or timer events) without subscribing to a	76	programming (waiting for I/O or timer events) without subscribing to a
40	religion, a way of living, and most importantly: without forcing your	77	religion, a way of living, and most importantly: without forcing your
41	module users into the same thing by forcing them to use the same event	78	module users into the same thing by forcing them to use the same event
42	model you use.	79	model you use.
43		80
44	For modules like POE or IO::Async (which is a total misnomer as it is	81	For modules like POE or IO::Async (which is a total misnomer as it is
45	actually doing all I/O I<synchronously>...), using them in your module is	82	actually doing all I/O I<synchronously>...), using them in your module is
46	like joining a cult: After you joined, you are dependent on them and you	83	like joining a cult: After you join, you are dependent on them and you
47	cannot use anything else, as it is simply incompatible to everything that	84	cannot use anything else, as they are simply incompatible to everything
48	isn't itself. What's worse, all the potential users of your module are	85	that isn't them. What's worse, all the potential users of your
49	I<also> forced to use the same event loop you use.	86	module are I<also> forced to use the same event loop you use.
50		87
51	AnyEvent is different: AnyEvent + POE works fine. AnyEvent + Glib works	88	AnyEvent is different: AnyEvent + POE works fine. AnyEvent + Glib works
52	fine. AnyEvent + Tk works fine etc. etc. but none of these work together	89	fine. AnyEvent + Tk works fine etc. etc. but none of these work together
53	with the rest: POE + IO::Async? No go. Tk + Event? No go. Again: if	90	with the rest: POE + EV? No go. Tk + Event? No go. Again: if your module
54	your module uses one of those, every user of your module has to use it,	91	uses one of those, every user of your module has to use it, too. But if
55	too. But if your module uses AnyEvent, it works transparently with all	92	your module uses AnyEvent, it works transparently with all event models it
56	event models it supports (including stuff like POE and IO::Async, as long	93	supports (including stuff like IO::Async, as long as those use one of the
57	as those use one of the supported event loops. It is trivial to add new	94	supported event loops. It is easy to add new event loops to AnyEvent, too,
58	event loops to AnyEvent, too, so it is future-proof).	95	so it is future-proof).
59		96
60	In addition to being free of having to use I<the one and only true event	97	In addition to being free of having to use I<the one and only true event
61	model>, AnyEvent also is free of bloat and policy: with POE or similar	98	model>, AnyEvent also is free of bloat and policy: with POE or similar
62	modules, you get an enormous amount of code and strict rules you have to	99	modules, you get an enormous amount of code and strict rules you have to
63	follow. AnyEvent, on the other hand, is lean and up to the point, by only	100	follow. AnyEvent, on the other hand, is lean and to the point, by only
64	offering the functionality that is necessary, in as thin as a wrapper as	101	offering the functionality that is necessary, in as thin as a wrapper as
65	technically possible.	102	technically possible.
66		103
67	Of course, AnyEvent comes with a big (and fully optional!) toolbox	104	Of course, AnyEvent comes with a big (and fully optional!) toolbox
68	of useful functionality, such as an asynchronous DNS resolver, 100%	105	of useful functionality, such as an asynchronous DNS resolver, 100%
…		…
74	useful) and you want to force your users to use the one and only event	111	useful) and you want to force your users to use the one and only event
75	model, you should I<not> use this module.	112	model, you should I<not> use this module.
76		113
77	=head1 DESCRIPTION	114	=head1 DESCRIPTION
78		115
79	L<AnyEvent> provides an identical interface to multiple event loops. This	116	L<AnyEvent> provides a uniform interface to various event loops. This
80	allows module authors to utilise an event loop without forcing module	117	allows module authors to use event loop functionality without forcing
81	users to use the same event loop (as only a single event loop can coexist	118	module users to use a specific event loop implementation (since more
82	peacefully at any one time).	119	than one event loop cannot coexist peacefully).
83		120
84	The interface itself is vaguely similar, but not identical to the L<Event>	121	The interface itself is vaguely similar, but not identical to the L<Event>
85	module.	122	module.
86		123
87	During the first call of any watcher-creation method, the module tries	124	During the first call of any watcher-creation method, the module tries
88	to detect the currently loaded event loop by probing whether one of the	125	to detect the currently loaded event loop by probing whether one of the
89	following modules is already loaded: L<EV>,	126	following modules is already loaded: L<EV>, L<AnyEvent::Loop>,
90	L<Event>, L<Glib>, L<AnyEvent::Impl::Perl>, L<Tk>, L<Event::Lib>, L<Qt>,	127	L<Event>, L<Glib>, L<Tk>, L<Event::Lib>, L<Qt>, L<POE>. The first one
91	L<POE>. The first one found is used. If none are found, the module tries	128	found is used. If none are detected, the module tries to load the first
92	to load these modules (excluding Tk, Event::Lib, Qt and POE as the pure perl	129	four modules in the order given; but note that if L<EV> is not
93	adaptor should always succeed) in the order given. The first one that can	130	available, the pure-perl L<AnyEvent::Loop> should always work, so
94	be successfully loaded will be used. If, after this, still none could be	131	the other two are not normally tried.
95	found, AnyEvent will fall back to a pure-perl event loop, which is not
96	very efficient, but should work everywhere.
97		132
98	Because AnyEvent first checks for modules that are already loaded, loading	133	Because AnyEvent first checks for modules that are already loaded, loading
99	an event model explicitly before first using AnyEvent will likely make	134	an event model explicitly before first using AnyEvent will likely make
100	that model the default. For example:	135	that model the default. For example:
101		136
…		…
103	use AnyEvent;	138	use AnyEvent;
104		139
105	# .. AnyEvent will likely default to Tk	140	# .. AnyEvent will likely default to Tk
106		141
107	The I<likely> means that, if any module loads another event model and	142	The I<likely> means that, if any module loads another event model and
108	starts using it, all bets are off. Maybe you should tell their authors to	143	starts using it, all bets are off - this case should be very rare though,
109	use AnyEvent so their modules work together with others seamlessly...	144	as very few modules hardcode event loops without announcing this very
		145	loudly.
110		146
111	The pure-perl implementation of AnyEvent is called	147	The pure-perl implementation of AnyEvent is called C<AnyEvent::Loop>. Like
112	C<AnyEvent::Impl::Perl>. Like other event modules you can load it	148	other event modules you can load it explicitly and enjoy the high
113	explicitly and enjoy the high availability of that event loop :)	149	availability of that event loop :)
114		150
115	=head1 WATCHERS	151	=head1 WATCHERS
116		152
117	AnyEvent has the central concept of a I<watcher>, which is an object that	153	AnyEvent has the central concept of a I<watcher>, which is an object that
118	stores relevant data for each kind of event you are waiting for, such as	154	stores relevant data for each kind of event you are waiting for, such as
…		…
121	These watchers are normal Perl objects with normal Perl lifetime. After	157	These watchers are normal Perl objects with normal Perl lifetime. After
122	creating a watcher it will immediately "watch" for events and invoke the	158	creating a watcher it will immediately "watch" for events and invoke the
123	callback when the event occurs (of course, only when the event model	159	callback when the event occurs (of course, only when the event model
124	is in control).	160	is in control).
125		161
		162	Note that B<callbacks must not permanently change global variables>
		163	potentially in use by the event loop (such as C<$_> or C<$[>) and that B<<
		164	callbacks must not C<die> >>. The former is good programming practice in
		165	Perl and the latter stems from the fact that exception handling differs
		166	widely between event loops.
		167
126	To disable the watcher you have to destroy it (e.g. by setting the	168	To disable a watcher you have to destroy it (e.g. by setting the
127	variable you store it in to C<undef> or otherwise deleting all references	169	variable you store it in to C<undef> or otherwise deleting all references
128	to it).	170	to it).
129		171
130	All watchers are created by calling a method on the C<AnyEvent> class.	172	All watchers are created by calling a method on the C<AnyEvent> class.
131		173
132	Many watchers either are used with "recursion" (repeating timers for	174	Many watchers either are used with "recursion" (repeating timers for
133	example), or need to refer to their watcher object in other ways.	175	example), or need to refer to their watcher object in other ways.
134		176
135	An any way to achieve that is this pattern:	177	One way to achieve that is this pattern:
136		178
137	my $w; $w = AnyEvent->type (arg => value ..., cb => sub {	179	my $w; $w = AnyEvent->type (arg => value ..., cb => sub {
138	# you can use $w here, for example to undef it	180	# you can use $w here, for example to undef it
139	undef $w;	181	undef $w;
140	});	182	});
141		183
142	Note that C<my $w; $w => combination. This is necessary because in Perl,	184	Note that C<my $w; $w => combination. This is necessary because in Perl,
143	my variables are only visible after the statement in which they are	185	my variables are only visible after the statement in which they are
144	declared.	186	declared.
145		187
146	=head2 I/O WATCHERS	188	=head2 I/O WATCHERS
147		189
		190	$w = AnyEvent->io (
		191	fh => <filehandle_or_fileno>,
		192	poll => <"r" or "w">,
		193	cb => <callback>,
		194	);
		195
148	You can create an I/O watcher by calling the C<< AnyEvent->io >> method	196	You can create an I/O watcher by calling the C<< AnyEvent->io >> method
149	with the following mandatory key-value pairs as arguments:	197	with the following mandatory key-value pairs as arguments:
150		198
151	C<fh> the Perl I<file handle> (I<not> file descriptor) to watch	199	C<fh> is the Perl I<file handle> (or a naked file descriptor) to watch
		200	for events (AnyEvent might or might not keep a reference to this file
		201	handle). Note that only file handles pointing to things for which
		202	non-blocking operation makes sense are allowed. This includes sockets,
		203	most character devices, pipes, fifos and so on, but not for example files
		204	or block devices.
		205
152	for events. C<poll> must be a string that is either C<r> or C<w>,	206	C<poll> must be a string that is either C<r> or C<w>, which creates a
153	which creates a watcher waiting for "r"eadable or "w"ritable events,	207	watcher waiting for "r"eadable or "w"ritable events, respectively.
		208
154	respectively. C<cb> is the callback to invoke each time the file handle	209	C<cb> is the callback to invoke each time the file handle becomes ready.
155	becomes ready.
156		210
157	Although the callback might get passed parameters, their value and	211	Although the callback might get passed parameters, their value and
158	presence is undefined and you cannot rely on them. Portable AnyEvent	212	presence is undefined and you cannot rely on them. Portable AnyEvent
159	callbacks cannot use arguments passed to I/O watcher callbacks.	213	callbacks cannot use arguments passed to I/O watcher callbacks.
160		214
161	The I/O watcher might use the underlying file descriptor or a copy of it.	215	The I/O watcher might use the underlying file descriptor or a copy of it.
162	You must not close a file handle as long as any watcher is active on the	216	You must not close a file handle as long as any watcher is active on the
163	underlying file descriptor.	217	underlying file descriptor.
164		218
165	Some event loops issue spurious readyness notifications, so you should	219	Some event loops issue spurious readiness notifications, so you should
166	always use non-blocking calls when reading/writing from/to your file	220	always use non-blocking calls when reading/writing from/to your file
167	handles.	221	handles.
168		222
169	Example:
170
171	# wait for readability of STDIN, then read a line and disable the watcher	223	Example: wait for readability of STDIN, then read a line and disable the
		224	watcher.
		225
172	my $w; $w = AnyEvent->io (fh => \*STDIN, poll => 'r', cb => sub {	226	my $w; $w = AnyEvent->io (fh => \*STDIN, poll => 'r', cb => sub {
173	chomp (my $input = <STDIN>);	227	chomp (my $input = <STDIN>);
174	warn "read: $input\n";	228	warn "read: $input\n";
175	undef $w;	229	undef $w;
176	});	230	});
177		231
178	=head2 TIME WATCHERS	232	=head2 TIME WATCHERS
179		233
		234	$w = AnyEvent->timer (after => <seconds>, cb => <callback>);
		235
		236	$w = AnyEvent->timer (
		237	after => <fractional_seconds>,
		238	interval => <fractional_seconds>,
		239	cb => <callback>,
		240	);
		241
180	You can create a time watcher by calling the C<< AnyEvent->timer >>	242	You can create a time watcher by calling the C<< AnyEvent->timer >>
181	method with the following mandatory arguments:	243	method with the following mandatory arguments:
182		244
183	C<after> specifies after how many seconds (fractional values are	245	C<after> specifies after how many seconds (fractional values are
184	supported) the callback should be invoked. C<cb> is the callback to invoke	246	supported) the callback should be invoked. C<cb> is the callback to invoke
…		…
186		248
187	Although the callback might get passed parameters, their value and	249	Although the callback might get passed parameters, their value and
188	presence is undefined and you cannot rely on them. Portable AnyEvent	250	presence is undefined and you cannot rely on them. Portable AnyEvent
189	callbacks cannot use arguments passed to time watcher callbacks.	251	callbacks cannot use arguments passed to time watcher callbacks.
190		252
191	The timer callback will be invoked at most once: if you want a repeating	253	The callback will normally be invoked only once. If you specify another
192	timer you have to create a new watcher (this is a limitation by both Tk	254	parameter, C<interval>, as a strictly positive number (> 0), then the
193	and Glib).	255	callback will be invoked regularly at that interval (in fractional
		256	seconds) after the first invocation. If C<interval> is specified with a
		257	false value, then it is treated as if it were not specified at all.
194		258
195	Example:	259	The callback will be rescheduled before invoking the callback, but no
		260	attempt is made to avoid timer drift in most backends, so the interval is
		261	only approximate.
196		262
197	# fire an event after 7.7 seconds	263	Example: fire an event after 7.7 seconds.
		264
198	my $w = AnyEvent->timer (after => 7.7, cb => sub {	265	my $w = AnyEvent->timer (after => 7.7, cb => sub {
199	warn "timeout\n";	266	warn "timeout\n";
200	});	267	});
201		268
202	# to cancel the timer:	269	# to cancel the timer:
203	undef $w;	270	undef $w;
204		271
205	Example 2:
206
207	# fire an event after 0.5 seconds, then roughly every second	272	Example 2: fire an event after 0.5 seconds, then roughly every second.
208	my $w;
209		273
210	my $cb = sub {
211	# cancel the old timer while creating a new one
212	$w = AnyEvent->timer (after => 1, cb => $cb);	274	my $w = AnyEvent->timer (after => 0.5, interval => 1, cb => sub {
		275	warn "timeout\n";
213	};	276	};
214
215	# start the "loop" by creating the first watcher
216	$w = AnyEvent->timer (after => 0.5, cb => $cb);
217		277
218	=head3 TIMING ISSUES	278	=head3 TIMING ISSUES
219		279
220	There are two ways to handle timers: based on real time (relative, "fire	280	There are two ways to handle timers: based on real time (relative, "fire
221	in 10 seconds") and based on wallclock time (absolute, "fire at 12	281	in 10 seconds") and based on wallclock time (absolute, "fire at 12
…		…
223		283
224	While most event loops expect timers to specified in a relative way, they	284	While most event loops expect timers to specified in a relative way, they
225	use absolute time internally. This makes a difference when your clock	285	use absolute time internally. This makes a difference when your clock
226	"jumps", for example, when ntp decides to set your clock backwards from	286	"jumps", for example, when ntp decides to set your clock backwards from
227	the wrong date of 2014-01-01 to 2008-01-01, a watcher that is supposed to	287	the wrong date of 2014-01-01 to 2008-01-01, a watcher that is supposed to
228	fire "after" a second might actually take six years to finally fire.	288	fire "after a second" might actually take six years to finally fire.
229		289
230	AnyEvent cannot compensate for this. The only event loop that is conscious	290	AnyEvent cannot compensate for this. The only event loop that is conscious
231	about these issues is L<EV>, which offers both relative (ev_timer, based	291	of these issues is L<EV>, which offers both relative (ev_timer, based
232	on true relative time) and absolute (ev_periodic, based on wallclock time)	292	on true relative time) and absolute (ev_periodic, based on wallclock time)
233	timers.	293	timers.
234		294
235	AnyEvent always prefers relative timers, if available, matching the	295	AnyEvent always prefers relative timers, if available, matching the
236	AnyEvent API.	296	AnyEvent API.
…		…
258	I<In almost all cases (in all cases if you don't care), this is the	318	I<In almost all cases (in all cases if you don't care), this is the
259	function to call when you want to know the current time.>	319	function to call when you want to know the current time.>
260		320
261	This function is also often faster then C<< AnyEvent->time >>, and	321	This function is also often faster then C<< AnyEvent->time >>, and
262	thus the preferred method if you want some timestamp (for example,	322	thus the preferred method if you want some timestamp (for example,
263	L<AnyEvent::Handle> uses this to update it's activity timeouts).	323	L<AnyEvent::Handle> uses this to update its activity timeouts).
264		324
265	The rest of this section is only of relevance if you try to be very exact	325	The rest of this section is only of relevance if you try to be very exact
266	with your timing, you can skip it without bad conscience.	326	with your timing; you can skip it without a bad conscience.
267		327
268	For a practical example of when these times differ, consider L<Event::Lib>	328	For a practical example of when these times differ, consider L<Event::Lib>
269	and L<EV> and the following set-up:	329	and L<EV> and the following set-up:
270		330
271	The event loop is running and has just invoked one of your callback at	331	The event loop is running and has just invoked one of your callbacks at
272	time=500 (assume no other callbacks delay processing). In your callback,	332	time=500 (assume no other callbacks delay processing). In your callback,
273	you wait a second by executing C<sleep 1> (blocking the process for a	333	you wait a second by executing C<sleep 1> (blocking the process for a
274	second) and then (at time=501) you create a relative timer that fires	334	second) and then (at time=501) you create a relative timer that fires
275	after three seconds.	335	after three seconds.
276		336
…		…
294	In either case, if you care (and in most cases, you don't), then you	354	In either case, if you care (and in most cases, you don't), then you
295	can get whatever behaviour you want with any event loop, by taking the	355	can get whatever behaviour you want with any event loop, by taking the
296	difference between C<< AnyEvent->time >> and C<< AnyEvent->now >> into	356	difference between C<< AnyEvent->time >> and C<< AnyEvent->now >> into
297	account.	357	account.
298		358
		359	=item AnyEvent->now_update
		360
		361	Some event loops (such as L<EV> or L<AnyEvent::Loop>) cache the current
		362	time for each loop iteration (see the discussion of L<< AnyEvent->now >>,
		363	above).
		364
		365	When a callback runs for a long time (or when the process sleeps), then
		366	this "current" time will differ substantially from the real time, which
		367	might affect timers and time-outs.
		368
		369	When this is the case, you can call this method, which will update the
		370	event loop's idea of "current time".
		371
		372	A typical example would be a script in a web server (e.g. C<mod_perl>) -
		373	when mod_perl executes the script, then the event loop will have the wrong
		374	idea about the "current time" (being potentially far in the past, when the
		375	script ran the last time). In that case you should arrange a call to C<<
		376	AnyEvent->now_update >> each time the web server process wakes up again
		377	(e.g. at the start of your script, or in a handler).
		378
		379	Note that updating the time I<might> cause some events to be handled.
		380
299	=back	381	=back
300		382
301	=head2 SIGNAL WATCHERS	383	=head2 SIGNAL WATCHERS
302		384
		385	$w = AnyEvent->signal (signal => <uppercase_signal_name>, cb => <callback>);
		386
303	You can watch for signals using a signal watcher, C<signal> is the signal	387	You can watch for signals using a signal watcher, C<signal> is the signal
304	I<name> without any C<SIG> prefix, C<cb> is the Perl callback to	388	I<name> in uppercase and without any C<SIG> prefix, C<cb> is the Perl
305	be invoked whenever a signal occurs.	389	callback to be invoked whenever a signal occurs.
306		390
307	Although the callback might get passed parameters, their value and	391	Although the callback might get passed parameters, their value and
308	presence is undefined and you cannot rely on them. Portable AnyEvent	392	presence is undefined and you cannot rely on them. Portable AnyEvent
309	callbacks cannot use arguments passed to signal watcher callbacks.	393	callbacks cannot use arguments passed to signal watcher callbacks.
310		394
…		…
312	invocation, and callback invocation will be synchronous. Synchronous means	396	invocation, and callback invocation will be synchronous. Synchronous means
313	that it might take a while until the signal gets handled by the process,	397	that it might take a while until the signal gets handled by the process,
314	but it is guaranteed not to interrupt any other callbacks.	398	but it is guaranteed not to interrupt any other callbacks.
315		399
316	The main advantage of using these watchers is that you can share a signal	400	The main advantage of using these watchers is that you can share a signal
317	between multiple watchers.	401	between multiple watchers, and AnyEvent will ensure that signals will not
		402	interrupt your program at bad times.
318		403
319	This watcher might use C<%SIG>, so programs overwriting those signals	404	This watcher might use C<%SIG> (depending on the event loop used),
320	directly will likely not work correctly.	405	so programs overwriting those signals directly will likely not work
		406	correctly.
321		407
322	Example: exit on SIGINT	408	Example: exit on SIGINT
323		409
324	my $w = AnyEvent->signal (signal => "INT", cb => sub { exit 1 });	410	my $w = AnyEvent->signal (signal => "INT", cb => sub { exit 1 });
325		411
		412	=head3 Restart Behaviour
		413
		414	While restart behaviour is up to the event loop implementation, most will
		415	not restart syscalls (that includes L<Async::Interrupt> and AnyEvent's
		416	pure perl implementation).
		417
		418	=head3 Safe/Unsafe Signals
		419
		420	Perl signals can be either "safe" (synchronous to opcode handling) or
		421	"unsafe" (asynchronous) - the former might get delayed indefinitely, the
		422	latter might corrupt your memory.
		423
		424	AnyEvent signal handlers are, in addition, synchronous to the event loop,
		425	i.e. they will not interrupt your running perl program but will only be
		426	called as part of the normal event handling (just like timer, I/O etc.
		427	callbacks, too).
		428
		429	=head3 Signal Races, Delays and Workarounds
		430
		431	Many event loops (e.g. Glib, Tk, Qt, IO::Async) do not support attaching
		432	callbacks to signals in a generic way, which is a pity, as you cannot
		433	do race-free signal handling in perl, requiring C libraries for
		434	this. AnyEvent will try to do its best, which means in some cases,
		435	signals will be delayed. The maximum time a signal might be delayed is
		436	specified in C<$AnyEvent::MAX_SIGNAL_LATENCY> (default: 10 seconds). This
		437	variable can be changed only before the first signal watcher is created,
		438	and should be left alone otherwise. This variable determines how often
		439	AnyEvent polls for signals (in case a wake-up was missed). Higher values
		440	will cause fewer spurious wake-ups, which is better for power and CPU
		441	saving.
		442
		443	All these problems can be avoided by installing the optional
		444	L<Async::Interrupt> module, which works with most event loops. It will not
		445	work with inherently broken event loops such as L<Event> or L<Event::Lib>
		446	(and not with L<POE> currently, as POE does its own workaround with
		447	one-second latency). For those, you just have to suffer the delays.
		448
326	=head2 CHILD PROCESS WATCHERS	449	=head2 CHILD PROCESS WATCHERS
327		450
		451	$w = AnyEvent->child (pid => <process id>, cb => <callback>);
		452
328	You can also watch on a child process exit and catch its exit status.	453	You can also watch for a child process exit and catch its exit status.
329		454
330	The child process is specified by the C<pid> argument (if set to C<0>, it	455	The child process is specified by the C<pid> argument (on some backends,
331	watches for any child process exit). The watcher will trigger as often	456	using C<0> watches for any child process exit, on others this will
332	as status change for the child are received. This works by installing a	457	croak). The watcher will be triggered only when the child process has
333	signal handler for C<SIGCHLD>. The callback will be called with the pid	458	finished and an exit status is available, not on any trace events
334	and exit status (as returned by waitpid), so unlike other watcher types,	459	(stopped/continued).
335	you I<can> rely on child watcher callback arguments.	460
		461	The callback will be called with the pid and exit status (as returned by
		462	waitpid), so unlike other watcher types, you I<can> rely on child watcher
		463	callback arguments.
		464
		465	This watcher type works by installing a signal handler for C<SIGCHLD>,
		466	and since it cannot be shared, nothing else should use SIGCHLD or reap
		467	random child processes (waiting for specific child processes, e.g. inside
		468	C<system>, is just fine).
336		469
337	There is a slight catch to child watchers, however: you usually start them	470	There is a slight catch to child watchers, however: you usually start them
338	I<after> the child process was created, and this means the process could	471	I<after> the child process was created, and this means the process could
339	have exited already (and no SIGCHLD will be sent anymore).	472	have exited already (and no SIGCHLD will be sent anymore).
340		473
341	Not all event models handle this correctly (POE doesn't), but even for	474	Not all event models handle this correctly (neither POE nor IO::Async do,
		475	see their AnyEvent::Impl manpages for details), but even for event models
342	event models that I<do> handle this correctly, they usually need to be	476	that I<do> handle this correctly, they usually need to be loaded before
343	loaded before the process exits (i.e. before you fork in the first place).	477	the process exits (i.e. before you fork in the first place). AnyEvent's
		478	pure perl event loop handles all cases correctly regardless of when you
		479	start the watcher.
344		480
345	This means you cannot create a child watcher as the very first thing in an	481	This means you cannot create a child watcher as the very first
346	AnyEvent program, you I<have> to create at least one watcher before you	482	thing in an AnyEvent program, you I<have> to create at least one
347	C<fork> the child (alternatively, you can call C<AnyEvent::detect>).	483	watcher before you C<fork> the child (alternatively, you can call
		484	C<AnyEvent::detect>).
		485
		486	As most event loops do not support waiting for child events, they will be
		487	emulated by AnyEvent in most cases, in which case the latency and race
		488	problems mentioned in the description of signal watchers apply.
348		489
349	Example: fork a process and wait for it	490	Example: fork a process and wait for it
350		491
351	my $done = AnyEvent->condvar;	492	my $done = AnyEvent->condvar;
352		493
353	my $pid = fork or exit 5;	494	my $pid = fork or exit 5;
354		495
355	my $w = AnyEvent->child (	496	my $w = AnyEvent->child (
356	pid => $pid,	497	pid => $pid,
357	cb => sub {	498	cb => sub {
358	my ($pid, $status) = @_;	499	my ($pid, $status) = @_;
359	warn "pid $pid exited with status $status";	500	warn "pid $pid exited with status $status";
360	$done->send;	501	$done->send;
361	},	502	},
362	);	503	);
363		504
364	# do something else, then wait for process exit	505	# do something else, then wait for process exit
365	$done->recv;	506	$done->recv;
		507
		508	=head2 IDLE WATCHERS
		509
		510	$w = AnyEvent->idle (cb => <callback>);
		511
		512	This will repeatedly invoke the callback after the process becomes idle,
		513	until either the watcher is destroyed or new events have been detected.
		514
		515	Idle watchers are useful when there is a need to do something, but it
		516	is not so important (or wise) to do it instantly. The callback will be
		517	invoked only when there is "nothing better to do", which is usually
		518	defined as "all outstanding events have been handled and no new events
		519	have been detected". That means that idle watchers ideally get invoked
		520	when the event loop has just polled for new events but none have been
		521	detected. Instead of blocking to wait for more events, the idle watchers
		522	will be invoked.
		523
		524	Unfortunately, most event loops do not really support idle watchers (only
		525	EV, Event and Glib do it in a usable fashion) - for the rest, AnyEvent
		526	will simply call the callback "from time to time".
		527
		528	Example: read lines from STDIN, but only process them when the
		529	program is otherwise idle:
		530
		531	my @lines; # read data
		532	my $idle_w;
		533	my $io_w = AnyEvent->io (fh => \*STDIN, poll => 'r', cb => sub {
		534	push @lines, scalar <STDIN>;
		535
		536	# start an idle watcher, if not already done
		537	$idle_w ||= AnyEvent->idle (cb => sub {
		538	# handle only one line, when there are lines left
		539	if (my $line = shift @lines) {
		540	print "handled when idle: $line";
		541	} else {
		542	# otherwise disable the idle watcher again
		543	undef $idle_w;
		544	}
		545	});
		546	});
366		547
367	=head2 CONDITION VARIABLES	548	=head2 CONDITION VARIABLES
		549
		550	$cv = AnyEvent->condvar;
		551
		552	$cv->send (<list>);
		553	my @res = $cv->recv;
368		554
369	If you are familiar with some event loops you will know that all of them	555	If you are familiar with some event loops you will know that all of them
370	require you to run some blocking "loop", "run" or similar function that	556	require you to run some blocking "loop", "run" or similar function that
371	will actively watch for new events and call your callbacks.	557	will actively watch for new events and call your callbacks.
372		558
373	AnyEvent is different, it expects somebody else to run the event loop and	559	AnyEvent is slightly different: it expects somebody else to run the event
374	will only block when necessary (usually when told by the user).	560	loop and will only block when necessary (usually when told by the user).
375		561
376	The instrument to do that is called a "condition variable", so called	562	The tool to do that is called a "condition variable", so called because
377	because they represent a condition that must become true.	563	they represent a condition that must become true.
		564
		565	Now is probably a good time to look at the examples further below.
378		566
379	Condition variables can be created by calling the C<< AnyEvent->condvar	567	Condition variables can be created by calling the C<< AnyEvent->condvar
380	>> method, usually without arguments. The only argument pair allowed is	568	>> method, usually without arguments. The only argument pair allowed is
381	C<cb>, which specifies a callback to be called when the condition variable	569	C<cb>, which specifies a callback to be called when the condition variable
382	becomes true.	570	becomes true, with the condition variable as the first argument (but not
		571	the results).
383		572
384	After creation, the condition variable is "false" until it becomes "true"	573	After creation, the condition variable is "false" until it becomes "true"
385	by calling the C<send> method (or calling the condition variable as if it	574	by calling the C<send> method (or calling the condition variable as if it
386	were a callback, read about the caveats in the description for the C<<	575	were a callback, read about the caveats in the description for the C<<
387	->send >> method).	576	->send >> method).
388		577
389	Condition variables are similar to callbacks, except that you can	578	Since condition variables are the most complex part of the AnyEvent API, here are
390	optionally wait for them. They can also be called merge points - points	579	some different mental models of what they are - pick the ones you can connect to:
391	in time where multiple outstanding events have been processed. And yet	580
392	another way to call them is transactions - each condition variable can be	581	=over 4
393	used to represent a transaction, which finishes at some point and delivers	582
394	a result.	583	=item * Condition variables are like callbacks - you can call them (and pass them instead
		584	of callbacks). Unlike callbacks however, you can also wait for them to be called.
		585
		586	=item * Condition variables are signals - one side can emit or send them,
		587	the other side can wait for them, or install a handler that is called when
		588	the signal fires.
		589
		590	=item * Condition variables are like "Merge Points" - points in your program
		591	where you merge multiple independent results/control flows into one.
		592
		593	=item * Condition variables represent a transaction - functions that start
		594	some kind of transaction can return them, leaving the caller the choice
		595	between waiting in a blocking fashion, or setting a callback.
		596
		597	=item * Condition variables represent future values, or promises to deliver
		598	some result, long before the result is available.
		599
		600	=back
395		601
396	Condition variables are very useful to signal that something has finished,	602	Condition variables are very useful to signal that something has finished,
397	for example, if you write a module that does asynchronous http requests,	603	for example, if you write a module that does asynchronous http requests,
398	then a condition variable would be the ideal candidate to signal the	604	then a condition variable would be the ideal candidate to signal the
399	availability of results. The user can either act when the callback is	605	availability of results. The user can either act when the callback is
…		…
412		618
413	Condition variables are represented by hash refs in perl, and the keys	619	Condition variables are represented by hash refs in perl, and the keys
414	used by AnyEvent itself are all named C<_ae_XXX> to make subclassing	620	used by AnyEvent itself are all named C<_ae_XXX> to make subclassing
415	easy (it is often useful to build your own transaction class on top of	621	easy (it is often useful to build your own transaction class on top of
416	AnyEvent). To subclass, use C<AnyEvent::CondVar> as base class and call	622	AnyEvent). To subclass, use C<AnyEvent::CondVar> as base class and call
417	it's C<new> method in your own C<new> method.	623	its C<new> method in your own C<new> method.
418		624
419	There are two "sides" to a condition variable - the "producer side" which	625	There are two "sides" to a condition variable - the "producer side" which
420	eventually calls C<< -> send >>, and the "consumer side", which waits	626	eventually calls C<< -> send >>, and the "consumer side", which waits
421	for the send to occur.	627	for the send to occur.
422		628
423	Example: wait for a timer.	629	Example: wait for a timer.
424		630
425	# wait till the result is ready	631	# condition: "wait till the timer is fired"
426	my $result_ready = AnyEvent->condvar;	632	my $timer_fired = AnyEvent->condvar;
427		633
428	# do something such as adding a timer	634	# create the timer - we could wait for, say
429	# or socket watcher the calls $result_ready->send	635	# a handle becomign ready, or even an
430	# when the "result" is ready.	636	# AnyEvent::HTTP request to finish, but
431	# in this case, we simply use a timer:	637	# in this case, we simply use a timer:
432	my $w = AnyEvent->timer (	638	my $w = AnyEvent->timer (
433	after => 1,	639	after => 1,
434	cb => sub { $result_ready->send },	640	cb => sub { $timer_fired->send },
435	);	641	);
436		642
437	# this "blocks" (while handling events) till the callback	643	# this "blocks" (while handling events) till the callback
438	# calls send	644	# calls ->send
439	$result_ready->recv;	645	$timer_fired->recv;
440		646
441	Example: wait for a timer, but take advantage of the fact that	647	Example: wait for a timer, but take advantage of the fact that condition
442	condition variables are also code references.	648	variables are also callable directly.
443		649
444	my $done = AnyEvent->condvar;	650	my $done = AnyEvent->condvar;
445	my $delay = AnyEvent->timer (after => 5, cb => $done);	651	my $delay = AnyEvent->timer (after => 5, cb => $done);
446	$done->recv;	652	$done->recv;
		653
		654	Example: Imagine an API that returns a condvar and doesn't support
		655	callbacks. This is how you make a synchronous call, for example from
		656	the main program:
		657
		658	use AnyEvent::CouchDB;
		659
		660	...
		661
		662	my @info = $couchdb->info->recv;
		663
		664	And this is how you would just set a callback to be called whenever the
		665	results are available:
		666
		667	$couchdb->info->cb (sub {
		668	my @info = $_[0]->recv;
		669	});
447		670
448	=head3 METHODS FOR PRODUCERS	671	=head3 METHODS FOR PRODUCERS
449		672
450	These methods should only be used by the producing side, i.e. the	673	These methods should only be used by the producing side, i.e. the
451	code/module that eventually sends the signal. Note that it is also	674	code/module that eventually sends the signal. Note that it is also
…		…
464	immediately from within send.	687	immediately from within send.
465		688
466	Any arguments passed to the C<send> call will be returned by all	689	Any arguments passed to the C<send> call will be returned by all
467	future C<< ->recv >> calls.	690	future C<< ->recv >> calls.
468		691
469	Condition variables are overloaded so one can call them directly	692	Condition variables are overloaded so one can call them directly (as if
470	(as a code reference). Calling them directly is the same as calling	693	they were a code reference). Calling them directly is the same as calling
471	C<send>. Note, however, that many C-based event loops do not handle	694	C<send>.
472	overloading, so as tempting as it may be, passing a condition variable
473	instead of a callback does not work. Both the pure perl and EV loops
474	support overloading, however, as well as all functions that use perl to
475	invoke a callback (as in L<AnyEvent::Socket> and L<AnyEvent::DNS> for
476	example).
477		695
478	=item $cv->croak ($error)	696	=item $cv->croak ($error)
479		697
480	Similar to send, but causes all call's to C<< ->recv >> to invoke	698	Similar to send, but causes all calls to C<< ->recv >> to invoke
481	C<Carp::croak> with the given error message/object/scalar.	699	C<Carp::croak> with the given error message/object/scalar.
482		700
483	This can be used to signal any errors to the condition variable	701	This can be used to signal any errors to the condition variable
484	user/consumer.	702	user/consumer. Doing it this way instead of calling C<croak> directly
		703	delays the error detection, but has the overwhelming advantage that it
		704	diagnoses the error at the place where the result is expected, and not
		705	deep in some event callback with no connection to the actual code causing
		706	the problem.
485		707
486	=item $cv->begin ([group callback])	708	=item $cv->begin ([group callback])
487		709
488	=item $cv->end	710	=item $cv->end
489
490	These two methods are EXPERIMENTAL and MIGHT CHANGE.
491		711
492	These two methods can be used to combine many transactions/events into	712	These two methods can be used to combine many transactions/events into
493	one. For example, a function that pings many hosts in parallel might want	713	one. For example, a function that pings many hosts in parallel might want
494	to use a condition variable for the whole process.	714	to use a condition variable for the whole process.
495		715
496	Every call to C<< ->begin >> will increment a counter, and every call to	716	Every call to C<< ->begin >> will increment a counter, and every call to
497	C<< ->end >> will decrement it. If the counter reaches C<0> in C<< ->end	717	C<< ->end >> will decrement it. If the counter reaches C<0> in C<< ->end
498	>>, the (last) callback passed to C<begin> will be executed. That callback	718	>>, the (last) callback passed to C<begin> will be executed, passing the
499	is I<supposed> to call C<< ->send >>, but that is not required. If no	719	condvar as first argument. That callback is I<supposed> to call C<< ->send
500	callback was set, C<send> will be called without any arguments.	720	>>, but that is not required. If no group callback was set, C<send> will
		721	be called without any arguments.
501		722
502	Let's clarify this with the ping example:	723	You can think of C<< $cv->send >> giving you an OR condition (one call
		724	sends), while C<< $cv->begin >> and C<< $cv->end >> giving you an AND
		725	condition (all C<begin> calls must be C<end>'ed before the condvar sends).
		726
		727	Let's start with a simple example: you have two I/O watchers (for example,
		728	STDOUT and STDERR for a program), and you want to wait for both streams to
		729	close before activating a condvar:
503		730
504	my $cv = AnyEvent->condvar;	731	my $cv = AnyEvent->condvar;
505		732
		733	$cv->begin; # first watcher
		734	my $w1 = AnyEvent->io (fh => $fh1, cb => sub {
		735	defined sysread $fh1, my $buf, 4096
		736	or $cv->end;
		737	});
		738
		739	$cv->begin; # second watcher
		740	my $w2 = AnyEvent->io (fh => $fh2, cb => sub {
		741	defined sysread $fh2, my $buf, 4096
		742	or $cv->end;
		743	});
		744
		745	$cv->recv;
		746
		747	This works because for every event source (EOF on file handle), there is
		748	one call to C<begin>, so the condvar waits for all calls to C<end> before
		749	sending.
		750
		751	The ping example mentioned above is slightly more complicated, as the
		752	there are results to be passwd back, and the number of tasks that are
		753	begun can potentially be zero:
		754
		755	my $cv = AnyEvent->condvar;
		756
506	my %result;	757	my %result;
507	$cv->begin (sub { $cv->send (\%result) });	758	$cv->begin (sub { shift->send (\%result) });
508		759
509	for my $host (@list_of_hosts) {	760	for my $host (@list_of_hosts) {
510	$cv->begin;	761	$cv->begin;
511	ping_host_then_call_callback $host, sub {	762	ping_host_then_call_callback $host, sub {
512	$result{$host} = ...;	763	$result{$host} = ...;
…		…
527	loop, which serves two important purposes: first, it sets the callback	778	loop, which serves two important purposes: first, it sets the callback
528	to be called once the counter reaches C<0>, and second, it ensures that	779	to be called once the counter reaches C<0>, and second, it ensures that
529	C<send> is called even when C<no> hosts are being pinged (the loop	780	C<send> is called even when C<no> hosts are being pinged (the loop
530	doesn't execute once).	781	doesn't execute once).
531		782
532	This is the general pattern when you "fan out" into multiple subrequests:	783	This is the general pattern when you "fan out" into multiple (but
533	use an outer C<begin>/C<end> pair to set the callback and ensure C<end>	784	potentially zero) subrequests: use an outer C<begin>/C<end> pair to set
534	is called at least once, and then, for each subrequest you start, call	785	the callback and ensure C<end> is called at least once, and then, for each
535	C<begin> and for each subrequest you finish, call C<end>.	786	subrequest you start, call C<begin> and for each subrequest you finish,
		787	call C<end>.
536		788
537	=back	789	=back
538		790
539	=head3 METHODS FOR CONSUMERS	791	=head3 METHODS FOR CONSUMERS
540		792
…		…
544	=over 4	796	=over 4
545		797
546	=item $cv->recv	798	=item $cv->recv
547		799
548	Wait (blocking if necessary) until the C<< ->send >> or C<< ->croak	800	Wait (blocking if necessary) until the C<< ->send >> or C<< ->croak
549	>> methods have been called on c<$cv>, while servicing other watchers	801	>> methods have been called on C<$cv>, while servicing other watchers
550	normally.	802	normally.
551		803
552	You can only wait once on a condition - additional calls are valid but	804	You can only wait once on a condition - additional calls are valid but
553	will return immediately.	805	will return immediately.
554		806
…		…
556	function will call C<croak>.	808	function will call C<croak>.
557		809
558	In list context, all parameters passed to C<send> will be returned,	810	In list context, all parameters passed to C<send> will be returned,
559	in scalar context only the first one will be returned.	811	in scalar context only the first one will be returned.
560		812
		813	Note that doing a blocking wait in a callback is not supported by any
		814	event loop, that is, recursive invocation of a blocking C<< ->recv
		815	>> is not allowed, and the C<recv> call will C<croak> if such a
		816	condition is detected. This condition can be slightly loosened by using
		817	L<Coro::AnyEvent>, which allows you to do a blocking C<< ->recv >> from
		818	any thread that doesn't run the event loop itself.
		819
561	Not all event models support a blocking wait - some die in that case	820	Not all event models support a blocking wait - some die in that case
562	(programs might want to do that to stay interactive), so I<if you are	821	(programs might want to do that to stay interactive), so I<if you are
563	using this from a module, never require a blocking wait>, but let the	822	using this from a module, never require a blocking wait>. Instead, let the
564	caller decide whether the call will block or not (for example, by coupling	823	caller decide whether the call will block or not (for example, by coupling
565	condition variables with some kind of request results and supporting	824	condition variables with some kind of request results and supporting
566	callbacks so the caller knows that getting the result will not block,	825	callbacks so the caller knows that getting the result will not block,
567	while still supporting blocking waits if the caller so desires).	826	while still supporting blocking waits if the caller so desires).
568		827
569	Another reason I<never> to C<< ->recv >> in a module is that you cannot
570	sensibly have two C<< ->recv >>'s in parallel, as that would require
571	multiple interpreters or coroutines/threads, none of which C<AnyEvent>
572	can supply.
573
574	The L<Coro> module, however, I<can> and I<does> supply coroutines and, in
575	fact, L<Coro::AnyEvent> replaces AnyEvent's condvars by coroutine-safe
576	versions and also integrates coroutines into AnyEvent, making blocking
577	C<< ->recv >> calls perfectly safe as long as they are done from another
578	coroutine (one that doesn't run the event loop).
579
580	You can ensure that C<< -recv >> never blocks by setting a callback and	828	You can ensure that C<< ->recv >> never blocks by setting a callback and
581	only calling C<< ->recv >> from within that callback (or at a later	829	only calling C<< ->recv >> from within that callback (or at a later
582	time). This will work even when the event loop does not support blocking	830	time). This will work even when the event loop does not support blocking
583	waits otherwise.	831	waits otherwise.
584		832
585	=item $bool = $cv->ready	833	=item $bool = $cv->ready
586		834
587	Returns true when the condition is "true", i.e. whether C<send> or	835	Returns true when the condition is "true", i.e. whether C<send> or
588	C<croak> have been called.	836	C<croak> have been called.
589		837
590	=item $cb = $cv->cb ([new callback])	838	=item $cb = $cv->cb ($cb->($cv))
591		839
592	This is a mutator function that returns the callback set and optionally	840	This is a mutator function that returns the callback set and optionally
593	replaces it before doing so.	841	replaces it before doing so.
594		842
595	The callback will be called when the condition becomes "true", i.e. when	843	The callback will be called when the condition becomes "true", i.e. when
596	C<send> or C<croak> are called. Calling C<recv> inside the callback	844	C<send> or C<croak> are called, with the only argument being the
		845	condition variable itself. If the condition is already true, the
		846	callback is called immediately when it is set. Calling C<recv> inside
597	or at any later time is guaranteed not to block.	847	the callback or at any later time is guaranteed not to block.
598		848
599	=back	849	=back
600		850
		851	=head1 SUPPORTED EVENT LOOPS/BACKENDS
		852
		853	The available backend classes are (every class has its own manpage):
		854
		855	=over 4
		856
		857	=item Backends that are autoprobed when no other event loop can be found.
		858
		859	EV is the preferred backend when no other event loop seems to be in
		860	use. If EV is not installed, then AnyEvent will fall back to its own
		861	pure-perl implementation, which is available everywhere as it comes with
		862	AnyEvent itself.
		863
		864	AnyEvent::Impl::EV based on EV (interface to libev, best choice).
		865	AnyEvent::Impl::Perl pure-perl AnyEvent::Loop, fast and portable.
		866
		867	=item Backends that are transparently being picked up when they are used.
		868
		869	These will be used if they are already loaded when the first watcher
		870	is created, in which case it is assumed that the application is using
		871	them. This means that AnyEvent will automatically pick the right backend
		872	when the main program loads an event module before anything starts to
		873	create watchers. Nothing special needs to be done by the main program.
		874
		875	AnyEvent::Impl::Event based on Event, very stable, few glitches.
		876	AnyEvent::Impl::Glib based on Glib, slow but very stable.
		877	AnyEvent::Impl::Tk based on Tk, very broken.
		878	AnyEvent::Impl::EventLib based on Event::Lib, leaks memory and worse.
		879	AnyEvent::Impl::POE based on POE, very slow, some limitations.
		880	AnyEvent::Impl::Irssi used when running within irssi.
		881	AnyEvent::Impl::IOAsync based on IO::Async.
		882	AnyEvent::Impl::Cocoa based on Cocoa::EventLoop.
		883	AnyEvent::Impl::FLTK2 based on FLTK (fltk 2 binding).
		884
		885	=item Backends with special needs.
		886
		887	Qt requires the Qt::Application to be instantiated first, but will
		888	otherwise be picked up automatically. As long as the main program
		889	instantiates the application before any AnyEvent watchers are created,
		890	everything should just work.
		891
		892	AnyEvent::Impl::Qt based on Qt.
		893
		894	=item Event loops that are indirectly supported via other backends.
		895
		896	Some event loops can be supported via other modules:
		897
		898	There is no direct support for WxWidgets (L<Wx>) or L<Prima>.
		899
		900	B<WxWidgets> has no support for watching file handles. However, you can
		901	use WxWidgets through the POE adaptor, as POE has a Wx backend that simply
		902	polls 20 times per second, which was considered to be too horrible to even
		903	consider for AnyEvent.
		904
		905	B<Prima> is not supported as nobody seems to be using it, but it has a POE
		906	backend, so it can be supported through POE.
		907
		908	AnyEvent knows about both L<Prima> and L<Wx>, however, and will try to
		909	load L<POE> when detecting them, in the hope that POE will pick them up,
		910	in which case everything will be automatic.
		911
		912	=back
		913
601	=head1 GLOBAL VARIABLES AND FUNCTIONS	914	=head1 GLOBAL VARIABLES AND FUNCTIONS
602		915
		916	These are not normally required to use AnyEvent, but can be useful to
		917	write AnyEvent extension modules.
		918
603	=over 4	919	=over 4
604		920
605	=item $AnyEvent::MODEL	921	=item $AnyEvent::MODEL
606		922
607	Contains C<undef> until the first watcher is being created. Then it	923	Contains C<undef> until the first watcher is being created, before the
		924	backend has been autodetected.
		925
608	contains the event model that is being used, which is the name of the	926	Afterwards it contains the event model that is being used, which is the
609	Perl class implementing the model. This class is usually one of the	927	name of the Perl class implementing the model. This class is usually one
610	C<AnyEvent::Impl:xxx> modules, but can be any other class in the case	928	of the C<AnyEvent::Impl::xxx> modules, but can be any other class in the
611	AnyEvent has been extended at runtime (e.g. in I<rxvt-unicode>).	929	case AnyEvent has been extended at runtime (e.g. in I<rxvt-unicode> it
612		930	will be C<urxvt::anyevent>).
613	The known classes so far are:
614
615	AnyEvent::Impl::EV based on EV (an interface to libev, best choice).
616	AnyEvent::Impl::Event based on Event, second best choice.
617	AnyEvent::Impl::Perl pure-perl implementation, fast and portable.
618	AnyEvent::Impl::Glib based on Glib, third-best choice.
619	AnyEvent::Impl::Tk based on Tk, very bad choice.
620	AnyEvent::Impl::Qt based on Qt, cannot be autoprobed (see its docs).
621	AnyEvent::Impl::EventLib based on Event::Lib, leaks memory and worse.
622	AnyEvent::Impl::POE based on POE, not generic enough for full support.
623
624	There is no support for WxWidgets, as WxWidgets has no support for
625	watching file handles. However, you can use WxWidgets through the
626	POE Adaptor, as POE has a Wx backend that simply polls 20 times per
627	second, which was considered to be too horrible to even consider for
628	AnyEvent. Likewise, other POE backends can be used by AnyEvent by using
629	it's adaptor.
630
631	AnyEvent knows about L<Prima> and L<Wx> and will try to use L<POE> when
632	autodetecting them.
633		931
634	=item AnyEvent::detect	932	=item AnyEvent::detect
635		933
636	Returns C<$AnyEvent::MODEL>, forcing autodetection of the event model	934	Returns C<$AnyEvent::MODEL>, forcing autodetection of the event model
637	if necessary. You should only call this function right before you would	935	if necessary. You should only call this function right before you would
638	have created an AnyEvent watcher anyway, that is, as late as possible at	936	have created an AnyEvent watcher anyway, that is, as late as possible at
639	runtime.	937	runtime, and not e.g. during initialisation of your module.
		938
		939	The effect of calling this function is as if a watcher had been created
		940	(specifically, actions that happen "when the first watcher is created"
		941	happen when calling detetc as well).
		942
		943	If you need to do some initialisation before AnyEvent watchers are
		944	created, use C<post_detect>.
640		945
641	=item $guard = AnyEvent::post_detect { BLOCK }	946	=item $guard = AnyEvent::post_detect { BLOCK }
642		947
643	Arranges for the code block to be executed as soon as the event model is	948	Arranges for the code block to be executed as soon as the event model is
644	autodetected (or immediately if this has already happened).	949	autodetected (or immediately if that has already happened).
		950
		951	The block will be executed I<after> the actual backend has been detected
		952	(C<$AnyEvent::MODEL> is set), but I<before> any watchers have been
		953	created, so it is possible to e.g. patch C<@AnyEvent::ISA> or do
		954	other initialisations - see the sources of L<AnyEvent::Strict> or
		955	L<AnyEvent::AIO> to see how this is used.
		956
		957	The most common usage is to create some global watchers, without forcing
		958	event module detection too early, for example, L<AnyEvent::AIO> creates
		959	and installs the global L<IO::AIO> watcher in a C<post_detect> block to
		960	avoid autodetecting the event module at load time.
645		961
646	If called in scalar or list context, then it creates and returns an object	962	If called in scalar or list context, then it creates and returns an object
647	that automatically removes the callback again when it is destroyed. See	963	that automatically removes the callback again when it is destroyed (or
		964	C<undef> when the hook was immediately executed). See L<AnyEvent::AIO> for
648	L<Coro::BDB> for a case where this is useful.	965	a case where this is useful.
		966
		967	Example: Create a watcher for the IO::AIO module and store it in
		968	C<$WATCHER>, but do so only do so after the event loop is initialised.
		969
		970	our WATCHER;
		971
		972	my $guard = AnyEvent::post_detect {
		973	$WATCHER = AnyEvent->io (fh => IO::AIO::poll_fileno, poll => 'r', cb => \&IO::AIO::poll_cb);
		974	};
		975
		976	# the ||= is important in case post_detect immediately runs the block,
		977	# as to not clobber the newly-created watcher. assigning both watcher and
		978	# post_detect guard to the same variable has the advantage of users being
		979	# able to just C<undef $WATCHER> if the watcher causes them grief.
		980
		981	$WATCHER ||= $guard;
649		982
650	=item @AnyEvent::post_detect	983	=item @AnyEvent::post_detect
651		984
652	If there are any code references in this array (you can C<push> to it	985	If there are any code references in this array (you can C<push> to it
653	before or after loading AnyEvent), then they will called directly after	986	before or after loading AnyEvent), then they will be called directly
654	the event loop has been chosen.	987	after the event loop has been chosen.
655		988
656	You should check C<$AnyEvent::MODEL> before adding to this array, though:	989	You should check C<$AnyEvent::MODEL> before adding to this array, though:
657	if it contains a true value then the event loop has already been detected,	990	if it is defined then the event loop has already been detected, and the
658	and the array will be ignored.	991	array will be ignored.
659		992
660	Best use C<AnyEvent::post_detect { BLOCK }> instead.	993	Best use C<AnyEvent::post_detect { BLOCK }> when your application allows
		994	it, as it takes care of these details.
		995
		996	This variable is mainly useful for modules that can do something useful
		997	when AnyEvent is used and thus want to know when it is initialised, but do
		998	not need to even load it by default. This array provides the means to hook
		999	into AnyEvent passively, without loading it.
		1000
		1001	Example: To load Coro::AnyEvent whenever Coro and AnyEvent are used
		1002	together, you could put this into Coro (this is the actual code used by
		1003	Coro to accomplish this):
		1004
		1005	if (defined $AnyEvent::MODEL) {
		1006	# AnyEvent already initialised, so load Coro::AnyEvent
		1007	require Coro::AnyEvent;
		1008	} else {
		1009	# AnyEvent not yet initialised, so make sure to load Coro::AnyEvent
		1010	# as soon as it is
		1011	push @AnyEvent::post_detect, sub { require Coro::AnyEvent };
		1012	}
		1013
		1014	=item AnyEvent::postpone { BLOCK }
		1015
		1016	Arranges for the block to be executed as soon as possible, but not before
		1017	the call itself returns. In practise, the block will be executed just
		1018	before the event loop polls for new events, or shortly afterwards.
		1019
		1020	This function never returns anything (to make the C<return postpone { ...
		1021	}> idiom more useful.
		1022
		1023	To understand the usefulness of this function, consider a function that
		1024	asynchronously does something for you and returns some transaction
		1025	object or guard to let you cancel the operation. For example,
		1026	C<AnyEvent::Socket::tcp_connect>:
		1027
		1028	# start a conenction attempt unless one is active
		1029	$self->{connect_guard} ||= AnyEvent::Socket::tcp_connect "www.example.net", 80, sub {
		1030	delete $self->{connect_guard};
		1031	...
		1032	};
		1033
		1034	Imagine that this function could instantly call the callback, for
		1035	example, because it detects an obvious error such as a negative port
		1036	number. Invoking the callback before the function returns causes problems
		1037	however: the callback will be called and will try to delete the guard
		1038	object. But since the function hasn't returned yet, there is nothing to
		1039	delete. When the function eventually returns it will assign the guard
		1040	object to C<< $self->{connect_guard} >>, where it will likely never be
		1041	deleted, so the program thinks it is still trying to connect.
		1042
		1043	This is where C<AnyEvent::postpone> should be used. Instead of calling the
		1044	callback directly on error:
		1045
		1046	$cb->(undef), return # signal error to callback, BAD!
		1047	if $some_error_condition;
		1048
		1049	It should use C<postpone>:
		1050
		1051	AnyEvent::postpone { $cb->(undef) }, return # signal error to callback, later
		1052	if $some_error_condition;
		1053
		1054	=item AnyEvent::log $level, $msg[, @args]
		1055
		1056	Log the given C<$msg> at the given C<$level>.
		1057
		1058	Loads AnyEvent::Log on first use and calls C<AnyEvent::Log::log> -
		1059	consequently, look at the L<AnyEvent::Log> documentation for details.
		1060
		1061	If you want to sprinkle loads of logging calls around your code, consider
		1062	creating a logger callback with the C<AnyEvent::Log::logger> function.
661		1063
662	=back	1064	=back
663		1065
664	=head1 WHAT TO DO IN A MODULE	1066	=head1 WHAT TO DO IN A MODULE
665		1067
…		…
676	because it will stall the whole program, and the whole point of using	1078	because it will stall the whole program, and the whole point of using
677	events is to stay interactive.	1079	events is to stay interactive.
678		1080
679	It is fine, however, to call C<< ->recv >> when the user of your module	1081	It is fine, however, to call C<< ->recv >> when the user of your module
680	requests it (i.e. if you create a http request object ad have a method	1082	requests it (i.e. if you create a http request object ad have a method
681	called C<results> that returns the results, it should call C<< ->recv >>	1083	called C<results> that returns the results, it may call C<< ->recv >>
682	freely, as the user of your module knows what she is doing. always).	1084	freely, as the user of your module knows what she is doing. Always).
683		1085
684	=head1 WHAT TO DO IN THE MAIN PROGRAM	1086	=head1 WHAT TO DO IN THE MAIN PROGRAM
685		1087
686	There will always be a single main program - the only place that should	1088	There will always be a single main program - the only place that should
687	dictate which event model to use.	1089	dictate which event model to use.
688		1090
689	If it doesn't care, it can just "use AnyEvent" and use it itself, or not	1091	If the program is not event-based, it need not do anything special, even
690	do anything special (it does not need to be event-based) and let AnyEvent	1092	when it depends on a module that uses an AnyEvent. If the program itself
691	decide which implementation to chose if some module relies on it.	1093	uses AnyEvent, but does not care which event loop is used, all it needs
		1094	to do is C<use AnyEvent>. In either case, AnyEvent will choose the best
		1095	available loop implementation.
692		1096
693	If the main program relies on a specific event model - for example, in	1097	If the main program relies on a specific event model - for example, in
694	Gtk2 programs you have to rely on the Glib module - you should load the	1098	Gtk2 programs you have to rely on the Glib module - you should load the
695	event module before loading AnyEvent or any module that uses it: generally	1099	event module before loading AnyEvent or any module that uses it: generally
696	speaking, you should load it as early as possible. The reason is that	1100	speaking, you should load it as early as possible. The reason is that
697	modules might create watchers when they are loaded, and AnyEvent will	1101	modules might create watchers when they are loaded, and AnyEvent will
698	decide on the event model to use as soon as it creates watchers, and it	1102	decide on the event model to use as soon as it creates watchers, and it
699	might chose the wrong one unless you load the correct one yourself.	1103	might choose the wrong one unless you load the correct one yourself.
700		1104
701	You can chose to use a pure-perl implementation by loading the	1105	You can chose to use a pure-perl implementation by loading the
702	C<AnyEvent::Impl::Perl> module, which gives you similar behaviour	1106	C<AnyEvent::Loop> module, which gives you similar behaviour
703	everywhere, but letting AnyEvent chose the model is generally better.	1107	everywhere, but letting AnyEvent chose the model is generally better.
704		1108
705	=head2 MAINLOOP EMULATION	1109	=head2 MAINLOOP EMULATION
706		1110
707	Sometimes (often for short test scripts, or even standalone programs who	1111	Sometimes (often for short test scripts, or even standalone programs who
…		…
720		1124
721		1125
722	=head1 OTHER MODULES	1126	=head1 OTHER MODULES
723		1127
724	The following is a non-exhaustive list of additional modules that use	1128	The following is a non-exhaustive list of additional modules that use
725	AnyEvent and can therefore be mixed easily with other AnyEvent modules	1129	AnyEvent as a client and can therefore be mixed easily with other
726	in the same program. Some of the modules come with AnyEvent, some are	1130	AnyEvent modules and other event loops in the same program. Some of the
727	available via CPAN.	1131	modules come as part of AnyEvent, the others are available via CPAN (see
		1132	L<http://search.cpan.org/search?m=module&q=anyevent%3A%3A*> for
		1133	a longer non-exhaustive list), and the list is heavily biased towards
		1134	modules of the AnyEvent author himself :)
728		1135
729	=over 4	1136	=over 4
730		1137
731	=item L<AnyEvent::Util>	1138	=item L<AnyEvent::Util>
732		1139
733	Contains various utility functions that replace often-used but blocking	1140	Contains various utility functions that replace often-used blocking
734	functions such as C<inet_aton> by event-/callback-based versions.	1141	functions such as C<inet_aton> with event/callback-based versions.
735
736	=item L<AnyEvent::Handle>
737
738	Provide read and write buffers and manages watchers for reads and writes.
739		1142
740	=item L<AnyEvent::Socket>	1143	=item L<AnyEvent::Socket>
741		1144
742	Provides various utility functions for (internet protocol) sockets,	1145	Provides various utility functions for (internet protocol) sockets,
743	addresses and name resolution. Also functions to create non-blocking tcp	1146	addresses and name resolution. Also functions to create non-blocking tcp
744	connections or tcp servers, with IPv6 and SRV record support and more.	1147	connections or tcp servers, with IPv6 and SRV record support and more.
745		1148
		1149	=item L<AnyEvent::Handle>
		1150
		1151	Provide read and write buffers, manages watchers for reads and writes,
		1152	supports raw and formatted I/O, I/O queued and fully transparent and
		1153	non-blocking SSL/TLS (via L<AnyEvent::TLS>).
		1154
746	=item L<AnyEvent::DNS>	1155	=item L<AnyEvent::DNS>
747		1156
748	Provides rich asynchronous DNS resolver capabilities.	1157	Provides rich asynchronous DNS resolver capabilities.
749		1158
		1159	=item L<AnyEvent::HTTP>, L<AnyEvent::IRC>, L<AnyEvent::XMPP>, L<AnyEvent::GPSD>, L<AnyEvent::IGS>, L<AnyEvent::FCP>
		1160
		1161	Implement event-based interfaces to the protocols of the same name (for
		1162	the curious, IGS is the International Go Server and FCP is the Freenet
		1163	Client Protocol).
		1164
		1165	=item L<AnyEvent::Handle::UDP>
		1166
		1167	Here be danger!
		1168
		1169	As Pauli would put it, "Not only is it not right, it's not even wrong!" -
		1170	there are so many things wrong with AnyEvent::Handle::UDP, most notably
		1171	its use of a stream-based API with a protocol that isn't streamable, that
		1172	the only way to improve it is to delete it.
		1173
		1174	It features data corruption (but typically only under load) and general
		1175	confusion. On top, the author is not only clueless about UDP but also
		1176	fact-resistant - some gems of his understanding: "connect doesn't work
		1177	with UDP", "UDP packets are not IP packets", "UDP only has datagrams, not
		1178	packets", "I don't need to implement proper error checking as UDP doesn't
		1179	support error checking" and so on - he doesn't even understand what's
		1180	wrong with his module when it is explained to him.
		1181
		1182	=item L<AnyEvent::DBI>
		1183
		1184	Executes L<DBI> requests asynchronously in a proxy process for you,
		1185	notifying you in an event-based way when the operation is finished.
		1186
		1187	=item L<AnyEvent::AIO>
		1188
		1189	Truly asynchronous (as opposed to non-blocking) I/O, should be in the
		1190	toolbox of every event programmer. AnyEvent::AIO transparently fuses
		1191	L<IO::AIO> and AnyEvent together, giving AnyEvent access to event-based
		1192	file I/O, and much more.
		1193
750	=item L<AnyEvent::HTTPD>	1194	=item L<AnyEvent::HTTPD>
751		1195
752	Provides a simple web application server framework.	1196	A simple embedded webserver.
753		1197
754	=item L<AnyEvent::FastPing>	1198	=item L<AnyEvent::FastPing>
755		1199
756	The fastest ping in the west.	1200	The fastest ping in the west.
757		1201
758	=item L<Net::IRC3>
759
760	AnyEvent based IRC client module family.
761
762	=item L<Net::XMPP2>
763
764	AnyEvent based XMPP (Jabber protocol) module family.
765
766	=item L<Net::FCP>
767
768	AnyEvent-based implementation of the Freenet Client Protocol, birthplace
769	of AnyEvent.
770
771	=item L<Event::ExecFlow>
772
773	High level API for event-based execution flow control.
774
775	=item L<Coro>	1202	=item L<Coro>
776		1203
777	Has special support for AnyEvent via L<Coro::AnyEvent>.	1204	Has special support for AnyEvent via L<Coro::AnyEvent>.
778		1205
779	=item L<AnyEvent::AIO>, L<IO::AIO>
780
781	Truly asynchronous I/O, should be in the toolbox of every event
782	programmer. AnyEvent::AIO transparently fuses IO::AIO and AnyEvent
783	together.
784
785	=item L<AnyEvent::BDB>, L<BDB>
786
787	Truly asynchronous Berkeley DB access. AnyEvent::AIO transparently fuses
788	IO::AIO and AnyEvent together.
789
790	=item L<IO::Lambda>
791
792	The lambda approach to I/O - don't ask, look there. Can use AnyEvent.
793
794	=back	1206	=back
795		1207
796	=cut	1208	=cut
797		1209
798	package AnyEvent;	1210	package AnyEvent;
799		1211
800	no warnings;	1212	# basically a tuned-down version of common::sense
801	use strict;	1213	sub common_sense {
		1214	# from common:.sense 3.4
		1215	${^WARNING_BITS} ^= ${^WARNING_BITS} ^ "\x3c\x3f\x33\x00\x0f\xf0\x0f\xc0\xf0\xfc\x33\x00";
		1216	# use strict vars subs - NO UTF-8, as Util.pm doesn't like this atm. (uts46data.pl)
		1217	$^H |= 0x00000600;
		1218	}
802		1219
		1220	BEGIN { AnyEvent::common_sense }
		1221
803	use Carp;	1222	use Carp ();
804		1223
805	our $VERSION = '4.05';	1224	our $VERSION = '6.01';
806	our $MODEL;	1225	our $MODEL;
807		1226
808	our $AUTOLOAD;
809	our @ISA;	1227	our @ISA;
810		1228
811	our @REGISTRY;	1229	our @REGISTRY;
812		1230
813	our $WIN32;	1231	our $VERBOSE;
814		1232
815	BEGIN {	1233	BEGIN {
816	my $win32 = ! ! ($^O =~ /mswin32/i);	1234	require "AnyEvent/constants.pl";
817	eval "sub WIN32(){ $win32 }";
818	}
819		1235
		1236	eval "sub TAINT (){" . (${^TAINT}*1) . "}";
		1237
		1238	delete @ENV{grep /^PERL_ANYEVENT_/, keys %ENV}
		1239	if ${^TAINT};
		1240
820	our $verbose = $ENV{PERL_ANYEVENT_VERBOSE}*1;	1241	$VERBOSE = $ENV{PERL_ANYEVENT_VERBOSE}*1;
		1242	}
		1243
		1244	our $MAX_SIGNAL_LATENCY = 10;
821		1245
822	our %PROTOCOL; # (ipv4|ipv6) => (1|2), higher numbers are preferred	1246	our %PROTOCOL; # (ipv4|ipv6) => (1|2), higher numbers are preferred
823		1247
824	{	1248	{
825	my $idx;	1249	my $idx;
826	$PROTOCOL{$_} = ++$idx	1250	$PROTOCOL{$_} = ++$idx
827	for reverse split /\s*,\s*/,	1251	for reverse split /\s*,\s*/,
828	$ENV{PERL_ANYEVENT_PROTOCOLS} || "ipv4,ipv6";	1252	$ENV{PERL_ANYEVENT_PROTOCOLS} || "ipv4,ipv6";
829	}	1253	}
830		1254
		1255	our @post_detect;
		1256
		1257	sub post_detect(&) {
		1258	my ($cb) = @_;
		1259
		1260	push @post_detect, $cb;
		1261
		1262	defined wantarray
		1263	? bless \$cb, "AnyEvent::Util::postdetect"
		1264	: ()
		1265	}
		1266
		1267	sub AnyEvent::Util::postdetect::DESTROY {
		1268	@post_detect = grep $_ != ${$_[0]}, @post_detect;
		1269	}
		1270
		1271	our $POSTPONE_W;
		1272	our @POSTPONE;
		1273
		1274	sub _postpone_exec {
		1275	undef $POSTPONE_W;
		1276
		1277	&{ shift @POSTPONE }
		1278	while @POSTPONE;
		1279	}
		1280
		1281	sub postpone(&) {
		1282	push @POSTPONE, shift;
		1283
		1284	$POSTPONE_W ||= AE::timer (0, 0, \&_postpone_exec);
		1285
		1286	()
		1287	}
		1288
		1289	sub log($$;@) {
		1290	require AnyEvent::Log;
		1291	# AnyEvent::Log overwrites this function
		1292	goto &log;
		1293	}
		1294
831	my @models = (	1295	our @models = (
832	[EV:: => AnyEvent::Impl::EV::],	1296	[EV:: => AnyEvent::Impl::EV:: , 1],
833	[Event:: => AnyEvent::Impl::Event::],	1297	[AnyEvent::Loop:: => AnyEvent::Impl::Perl:: , 1],
834	[AnyEvent::Impl::Perl:: => AnyEvent::Impl::Perl::],
835	# everything below here will not be autoprobed	1298	# everything below here will not (normally) be autoprobed
836	# as the pureperl backend should work everywhere	1299	# as the pure perl backend should work everywhere
837	# and is usually faster	1300	# and is usually faster
		1301	[Event:: => AnyEvent::Impl::Event::, 1],
		1302	[Glib:: => AnyEvent::Impl::Glib:: , 1], # becomes extremely slow with many watchers
		1303	[Event::Lib:: => AnyEvent::Impl::EventLib::], # too buggy
		1304	[Irssi:: => AnyEvent::Impl::Irssi::], # Irssi has a bogus "Event" package
838	[Tk:: => AnyEvent::Impl::Tk::], # crashes with many handles	1305	[Tk:: => AnyEvent::Impl::Tk::], # crashes with many handles
839	[Glib:: => AnyEvent::Impl::Glib::], # becomes extremely slow with many watchers
840	[Event::Lib:: => AnyEvent::Impl::EventLib::], # too buggy
841	[Qt:: => AnyEvent::Impl::Qt::], # requires special main program	1306	[Qt:: => AnyEvent::Impl::Qt::], # requires special main program
842	[POE::Kernel:: => AnyEvent::Impl::POE::], # lasciate ogni speranza	1307	[POE::Kernel:: => AnyEvent::Impl::POE::], # lasciate ogni speranza
843	[Wx:: => AnyEvent::Impl::POE::],	1308	[Wx:: => AnyEvent::Impl::POE::],
844	[Prima:: => AnyEvent::Impl::POE::],	1309	[Prima:: => AnyEvent::Impl::POE::],
		1310	[IO::Async::Loop:: => AnyEvent::Impl::IOAsync::], # a bitch to autodetect
		1311	[Cocoa::EventLoop:: => AnyEvent::Impl::Cocoa::],
		1312	[FLTK:: => AnyEvent::Impl::FLTK2::],
845	);	1313	);
846		1314
847	our %method = map +($_ => 1), qw(io timer time now signal child condvar one_event DESTROY);	1315	our @isa_hook;
848		1316
849	our @post_detect;	1317	sub _isa_set {
		1318	my @pkg = ("AnyEvent", (map $_->[0], grep defined, @isa_hook), $MODEL);
850		1319
		1320	@{"$pkg[$_-1]::ISA"} = $pkg[$_]
		1321	for 1 .. $#pkg;
		1322
		1323	grep $_ && $_->[1], @isa_hook
		1324	and AE::_reset ();
		1325	}
		1326
		1327	# used for hooking AnyEvent::Strict and AnyEvent::Debug::Wrap into the class hierarchy
		1328	sub _isa_hook($$;$) {
		1329	my ($i, $pkg, $reset_ae) = @_;
		1330
		1331	$isa_hook[$i] = $pkg ? [$pkg, $reset_ae] : undef;
		1332
		1333	_isa_set;
		1334	}
		1335
		1336	# all autoloaded methods reserve the complete glob, not just the method slot.
		1337	# due to bugs in perls method cache implementation.
		1338	our @methods = qw(io timer time now now_update signal child idle condvar);
		1339
851	sub post_detect(&) {	1340	sub detect() {
852	my ($cb) = @_;	1341	return $MODEL if $MODEL; # some programs keep references to detect
853		1342
854	if ($MODEL) {	1343	local $!; # for good measure
855	$cb->();	1344	local $SIG{__DIE__}; # we use eval
856		1345
857	1	1346	# free some memory
		1347	*detect = sub () { $MODEL };
		1348	# undef &func doesn't correctly update the method cache. grmbl.
		1349	# so we delete the whole glob. grmbl.
		1350	# otoh, perl doesn't let me undef an active usb, but it lets me free
		1351	# a glob with an active sub. hrm. i hope it works, but perl is
		1352	# usually buggy in this department. sigh.
		1353	delete @{"AnyEvent::"}{@methods};
		1354	undef @methods;
		1355
		1356	if ($ENV{PERL_ANYEVENT_MODEL} =~ /^([a-zA-Z0-9:]+)$/) {
		1357	my $model = $1;
		1358	$model = "AnyEvent::Impl::$model" unless $model =~ s/::$//;
		1359	if (eval "require $model") {
		1360	$MODEL = $model;
		1361	AnyEvent::log 7 => "loaded model '$model' (forced by \$ENV{PERL_ANYEVENT_MODEL}), using it."
		1362	if $VERBOSE >= 7;
858	} else {	1363	} else {
859	push @post_detect, $cb;	1364	AnyEvent::log warn => "unable to load model '$model' (from \$ENV{PERL_ANYEVENT_MODEL}):\n$@";
860		1365	}
861	defined wantarray
862	? bless \$cb, "AnyEvent::Util::PostDetect"
863	: ()
864	}	1366	}
865	}
866		1367
867	sub AnyEvent::Util::PostDetect::DESTROY {	1368	# check for already loaded models
868	@post_detect = grep $_ != ${$_[0]}, @post_detect;
869	}
870
871	sub detect() {
872	unless ($MODEL) {	1369	unless ($MODEL) {
873	no strict 'refs';	1370	for (@REGISTRY, @models) {
874	local $SIG{__DIE__};	1371	my ($package, $model) = @$_;
875		1372	if (${"$package\::VERSION"} > 0) {
876	if ($ENV{PERL_ANYEVENT_MODEL} =~ /^([a-zA-Z]+)$/) {
877	my $model = "AnyEvent::Impl::$1";
878	if (eval "require $model") {	1373	if (eval "require $model") {
879	$MODEL = $model;	1374	$MODEL = $model;
880	warn "AnyEvent: loaded model '$model' (forced by \$PERL_ANYEVENT_MODEL), using it.\n" if $verbose > 1;	1375	AnyEvent::log 7 => "autodetected model '$model', using it."
881	} else {	1376	if $VERBOSE >= 7;
882	warn "AnyEvent: unable to load model '$model' (from \$PERL_ANYEVENT_MODEL):\n$@" if $verbose;	1377	last;
		1378	}
883	}	1379	}
884	}	1380	}
885		1381
886	# check for already loaded models
887	unless ($MODEL) {	1382	unless ($MODEL) {
		1383	# try to autoload a model
888	for (@REGISTRY, @models) {	1384	for (@REGISTRY, @models) {
889	my ($package, $model) = @$_;	1385	my ($package, $model, $autoload) = @$_;
		1386	if (
		1387	$autoload
		1388	and eval "require $package"
890	if (${"$package\::VERSION"} > 0) {	1389	and ${"$package\::VERSION"} > 0
891	if (eval "require $model") {	1390	and eval "require $model"
		1391	) {
892	$MODEL = $model;	1392	$MODEL = $model;
893	warn "AnyEvent: autodetected model '$model', using it.\n" if $verbose > 1;	1393	AnyEvent::log 7 => "autoloaded model '$model', using it."
		1394	if $VERBOSE >= 7;
894	last;	1395	last;
895	}
896	}	1396	}
897	}	1397	}
898		1398
899	unless ($MODEL) {	1399	$MODEL
900	# try to load a model	1400	or die "AnyEvent: backend autodetection failed - did you properly install AnyEvent?";
		1401	}
		1402	}
901		1403
902	for (@REGISTRY, @models) {	1404	# free memory only needed for probing
903	my ($package, $model) = @$_;	1405	undef @models;
904	if (eval "require $package"	1406	undef @REGISTRY;
905	and ${"$package\::VERSION"} > 0	1407
906	and eval "require $model") {	1408	push @{"$MODEL\::ISA"}, "AnyEvent::Base";
907	$MODEL = $model;	1409
908	warn "AnyEvent: autoprobed model '$model', using it.\n" if $verbose > 1;	1410	# now nuke some methods that are overridden by the backend.
		1411	# SUPER usage is not allowed in these.
		1412	for (qw(time signal child idle)) {
		1413	undef &{"AnyEvent::Base::$_"}
		1414	if defined &{"$MODEL\::$_"};
		1415	}
		1416
		1417	_isa_set;
		1418
		1419	# we're officially open!
		1420
		1421	if ($ENV{PERL_ANYEVENT_STRICT}) {
		1422	require AnyEvent::Strict;
		1423	}
		1424
		1425	if ($ENV{PERL_ANYEVENT_DEBUG_WRAP}) {
		1426	require AnyEvent::Debug;
		1427	AnyEvent::Debug::wrap ($ENV{PERL_ANYEVENT_DEBUG_WRAP});
		1428	}
		1429
		1430	if (length $ENV{PERL_ANYEVENT_DEBUG_SHELL}) {
		1431	require AnyEvent::Socket;
		1432	require AnyEvent::Debug;
		1433
		1434	my $shell = $ENV{PERL_ANYEVENT_DEBUG_SHELL};
		1435	$shell =~ s/\$\$/$$/g;
		1436
		1437	my ($host, $service) = AnyEvent::Socket::parse_hostport ($shell);
		1438	$AnyEvent::Debug::SHELL = AnyEvent::Debug::shell ($host, $service);
		1439	}
		1440
		1441	# now the anyevent environment is set up as the user told us to, so
		1442	# call the actual user code - post detects
		1443
		1444	(shift @post_detect)->() while @post_detect;
		1445	undef @post_detect;
		1446
		1447	*post_detect = sub(&) {
		1448	shift->();
		1449
		1450	undef
		1451	};
		1452
		1453	$MODEL
		1454	}
		1455
		1456	for my $name (@methods) {
		1457	*$name = sub {
		1458	detect;
		1459	# we use goto because
		1460	# a) it makes the thunk more transparent
		1461	# b) it allows us to delete the thunk later
		1462	goto &{ UNIVERSAL::can AnyEvent => "SUPER::$name" }
		1463	};
		1464	}
		1465
		1466	# utility function to dup a filehandle. this is used by many backends
		1467	# to support binding more than one watcher per filehandle (they usually
		1468	# allow only one watcher per fd, so we dup it to get a different one).
		1469	sub _dupfh($$;$$) {
		1470	my ($poll, $fh, $r, $w) = @_;
		1471
		1472	# cygwin requires the fh mode to be matching, unix doesn't
		1473	my ($rw, $mode) = $poll eq "r" ? ($r, "<&") : ($w, ">&");
		1474
		1475	open my $fh2, $mode, $fh
		1476	or die "AnyEvent->io: cannot dup() filehandle in mode '$poll': $!,";
		1477
		1478	# we assume CLOEXEC is already set by perl in all important cases
		1479
		1480	($fh2, $rw)
		1481	}
		1482
		1483	=head1 SIMPLIFIED AE API
		1484
		1485	Starting with version 5.0, AnyEvent officially supports a second, much
		1486	simpler, API that is designed to reduce the calling, typing and memory
		1487	overhead by using function call syntax and a fixed number of parameters.
		1488
		1489	See the L<AE> manpage for details.
		1490
		1491	=cut
		1492
		1493	package AE;
		1494
		1495	our $VERSION = $AnyEvent::VERSION;
		1496
		1497	sub _reset() {
		1498	eval q{
		1499	# fall back to the main API by default - backends and AnyEvent::Base
		1500	# implementations can overwrite these.
		1501
		1502	sub io($$$) {
		1503	AnyEvent->io (fh => $_[0], poll => $_[1] ? "w" : "r", cb => $_[2])
		1504	}
		1505
		1506	sub timer($$$) {
		1507	AnyEvent->timer (after => $_[0], interval => $_[1], cb => $_[2])
		1508	}
		1509
		1510	sub signal($$) {
		1511	AnyEvent->signal (signal => $_[0], cb => $_[1])
		1512	}
		1513
		1514	sub child($$) {
		1515	AnyEvent->child (pid => $_[0], cb => $_[1])
		1516	}
		1517
		1518	sub idle($) {
		1519	AnyEvent->idle (cb => $_[0]);
		1520	}
		1521
		1522	sub cv(;&) {
		1523	AnyEvent->condvar (@_ ? (cb => $_[0]) : ())
		1524	}
		1525
		1526	sub now() {
		1527	AnyEvent->now
		1528	}
		1529
		1530	sub now_update() {
		1531	AnyEvent->now_update
		1532	}
		1533
		1534	sub time() {
		1535	AnyEvent->time
		1536	}
		1537
		1538	*postpone = \&AnyEvent::postpone;
		1539	*log = \&AnyEvent::log;
		1540	};
		1541	die if $@;
		1542	}
		1543
		1544	BEGIN { _reset }
		1545
		1546	package AnyEvent::Base;
		1547
		1548	# default implementations for many methods
		1549
		1550	sub time {
		1551	eval q{ # poor man's autoloading {}
		1552	# probe for availability of Time::HiRes
		1553	if (eval "use Time::HiRes (); Time::HiRes::time (); 1") {
		1554	AnyEvent::log 8 => "AnyEvent: using Time::HiRes for sub-second timing accuracy."
		1555	if $AnyEvent::VERBOSE >= 8;
		1556	*time = sub { Time::HiRes::time () };
		1557	*AE::time = \& Time::HiRes::time ;
		1558	# if (eval "use POSIX (); (POSIX::times())...
		1559	} else {
		1560	AnyEvent::log critical => "using built-in time(), WARNING, no sub-second resolution!";
		1561	*time = sub { CORE::time };
		1562	*AE::time = sub (){ CORE::time };
		1563	}
		1564
		1565	*now = \&time;
		1566	};
		1567	die if $@;
		1568
		1569	&time
		1570	}
		1571
		1572	*now = \&time;
		1573	sub now_update { }
		1574
		1575	sub _poll {
		1576	Carp::croak "$AnyEvent::MODEL does not support blocking waits. Caught";
		1577	}
		1578
		1579	# default implementation for ->condvar
		1580	# in fact, the default should not be overwritten
		1581
		1582	sub condvar {
		1583	eval q{ # poor man's autoloading {}
		1584	*condvar = sub {
		1585	bless { @_ == 3 ? (_ae_cb => $_[2]) : () }, "AnyEvent::CondVar"
		1586	};
		1587
		1588	*AE::cv = sub (;&) {
		1589	bless { @_ ? (_ae_cb => shift) : () }, "AnyEvent::CondVar"
		1590	};
		1591	};
		1592	die if $@;
		1593
		1594	&condvar
		1595	}
		1596
		1597	# default implementation for ->signal
		1598
		1599	our $HAVE_ASYNC_INTERRUPT;
		1600
		1601	sub _have_async_interrupt() {
		1602	$HAVE_ASYNC_INTERRUPT = 1*(!$ENV{PERL_ANYEVENT_AVOID_ASYNC_INTERRUPT}
		1603	&& eval "use Async::Interrupt 1.02 (); 1")
		1604	unless defined $HAVE_ASYNC_INTERRUPT;
		1605
		1606	$HAVE_ASYNC_INTERRUPT
		1607	}
		1608
		1609	our ($SIGPIPE_R, $SIGPIPE_W, %SIG_CB, %SIG_EV, $SIG_IO);
		1610	our (%SIG_ASY, %SIG_ASY_W);
		1611	our ($SIG_COUNT, $SIG_TW);
		1612
		1613	# install a dummy wakeup watcher to reduce signal catching latency
		1614	# used by Impls
		1615	sub _sig_add() {
		1616	unless ($SIG_COUNT++) {
		1617	# try to align timer on a full-second boundary, if possible
		1618	my $NOW = AE::now;
		1619
		1620	$SIG_TW = AE::timer
		1621	$MAX_SIGNAL_LATENCY - ($NOW - int $NOW),
		1622	$MAX_SIGNAL_LATENCY,
		1623	sub { } # just for the PERL_ASYNC_CHECK
		1624	;
		1625	}
		1626	}
		1627
		1628	sub _sig_del {
		1629	undef $SIG_TW
		1630	unless --$SIG_COUNT;
		1631	}
		1632
		1633	our $_sig_name_init; $_sig_name_init = sub {
		1634	eval q{ # poor man's autoloading {}
		1635	undef $_sig_name_init;
		1636
		1637	if (_have_async_interrupt) {
		1638	*sig2num = \&Async::Interrupt::sig2num;
		1639	*sig2name = \&Async::Interrupt::sig2name;
		1640	} else {
		1641	require Config;
		1642
		1643	my %signame2num;
		1644	@signame2num{ split ' ', $Config::Config{sig_name} }
		1645	= split ' ', $Config::Config{sig_num};
		1646
		1647	my @signum2name;
		1648	@signum2name[values %signame2num] = keys %signame2num;
		1649
		1650	*sig2num = sub($) {
		1651	$_[0] > 0 ? shift : $signame2num{+shift}
		1652	};
		1653	*sig2name = sub ($) {
		1654	$_[0] > 0 ? $signum2name[+shift] : shift
		1655	};
		1656	}
		1657	};
		1658	die if $@;
		1659	};
		1660
		1661	sub sig2num ($) { &$_sig_name_init; &sig2num }
		1662	sub sig2name($) { &$_sig_name_init; &sig2name }
		1663
		1664	sub signal {
		1665	eval q{ # poor man's autoloading {}
		1666	# probe for availability of Async::Interrupt
		1667	if (_have_async_interrupt) {
		1668	AnyEvent::log 8 => "using Async::Interrupt for race-free signal handling."
		1669	if $AnyEvent::VERBOSE >= 8;
		1670
		1671	$SIGPIPE_R = new Async::Interrupt::EventPipe;
		1672	$SIG_IO = AE::io $SIGPIPE_R->fileno, 0, \&_signal_exec;
		1673
		1674	} else {
		1675	AnyEvent::log 8 => "using emulated perl signal handling with latency timer."
		1676	if $AnyEvent::VERBOSE >= 8;
		1677
		1678	if (AnyEvent::WIN32) {
		1679	require AnyEvent::Util;
		1680
		1681	($SIGPIPE_R, $SIGPIPE_W) = AnyEvent::Util::portable_pipe ();
		1682	AnyEvent::Util::fh_nonblocking ($SIGPIPE_R, 1) if $SIGPIPE_R;
		1683	AnyEvent::Util::fh_nonblocking ($SIGPIPE_W, 1) if $SIGPIPE_W; # just in case
		1684	} else {
		1685	pipe $SIGPIPE_R, $SIGPIPE_W;
		1686	fcntl $SIGPIPE_R, AnyEvent::F_SETFL, AnyEvent::O_NONBLOCK if $SIGPIPE_R;
		1687	fcntl $SIGPIPE_W, AnyEvent::F_SETFL, AnyEvent::O_NONBLOCK if $SIGPIPE_W; # just in case
		1688
		1689	# not strictly required, as $^F is normally 2, but let's make sure...
		1690	fcntl $SIGPIPE_R, AnyEvent::F_SETFD, AnyEvent::FD_CLOEXEC;
		1691	fcntl $SIGPIPE_W, AnyEvent::F_SETFD, AnyEvent::FD_CLOEXEC;
		1692	}
		1693
		1694	$SIGPIPE_R
		1695	or Carp::croak "AnyEvent: unable to create a signal reporting pipe: $!\n";
		1696
		1697	$SIG_IO = AE::io $SIGPIPE_R, 0, \&_signal_exec;
		1698	}
		1699
		1700	*signal = $HAVE_ASYNC_INTERRUPT
		1701	? sub {
		1702	my (undef, %arg) = @_;
		1703
		1704	# async::interrupt
		1705	my $signal = sig2num $arg{signal};
		1706	$SIG_CB{$signal}{$arg{cb}} = $arg{cb};
		1707
		1708	$SIG_ASY{$signal} ||= new Async::Interrupt
		1709	cb => sub { undef $SIG_EV{$signal} },
		1710	signal => $signal,
		1711	pipe => [$SIGPIPE_R->filenos],
		1712	pipe_autodrain => 0,
		1713	;
		1714
		1715	bless [$signal, $arg{cb}], "AnyEvent::Base::signal"
		1716	}
		1717	: sub {
		1718	my (undef, %arg) = @_;
		1719
		1720	# pure perl
		1721	my $signal = sig2name $arg{signal};
		1722	$SIG_CB{$signal}{$arg{cb}} = $arg{cb};
		1723
		1724	$SIG{$signal} ||= sub {
909	last;	1725	local $!;
		1726	syswrite $SIGPIPE_W, "\x00", 1 unless %SIG_EV;
		1727	undef $SIG_EV{$signal};
910	}	1728	};
		1729
		1730	# can't do signal processing without introducing races in pure perl,
		1731	# so limit the signal latency.
		1732	_sig_add;
		1733
		1734	bless [$signal, $arg{cb}], "AnyEvent::Base::signal"
		1735	}
		1736	;
		1737
		1738	*AnyEvent::Base::signal::DESTROY = sub {
		1739	my ($signal, $cb) = @{$_[0]};
		1740
		1741	_sig_del;
		1742
		1743	delete $SIG_CB{$signal}{$cb};
		1744
		1745	$HAVE_ASYNC_INTERRUPT
		1746	? delete $SIG_ASY{$signal}
		1747	: # delete doesn't work with older perls - they then
		1748	# print weird messages, or just unconditionally exit
		1749	# instead of getting the default action.
		1750	undef $SIG{$signal}
		1751	unless keys %{ $SIG_CB{$signal} };
		1752	};
		1753
		1754	*_signal_exec = sub {
		1755	$HAVE_ASYNC_INTERRUPT
		1756	? $SIGPIPE_R->drain
		1757	: sysread $SIGPIPE_R, (my $dummy), 9;
		1758
		1759	while (%SIG_EV) {
		1760	for (keys %SIG_EV) {
		1761	delete $SIG_EV{$_};
		1762	&$_ for values %{ $SIG_CB{$_} || {} };
911	}	1763	}
912
913	$MODEL
914	or die "No event module selected for AnyEvent and autodetect failed. Install any one of these modules: EV, Event or Glib.";
915	}	1764	}
916	}	1765	};
917
918	unshift @ISA, $MODEL;
919	push @{"$MODEL\::ISA"}, "AnyEvent::Base";
920
921	(shift @post_detect)->() while @post_detect;
922	}
923
924	$MODEL
925	}
926
927	sub AUTOLOAD {
928	(my $func = $AUTOLOAD) =~ s/.*://;
929
930	$method{$func}
931	or croak "$func: not a valid method for AnyEvent objects";
932
933	detect unless $MODEL;
934
935	my $class = shift;
936	$class->$func (@_);
937	}
938
939	package AnyEvent::Base;
940
941	# default implementation for now and time
942
943	use Time::HiRes ();
944
945	sub time { Time::HiRes::time }
946	sub now { Time::HiRes::time }
947
948	# default implementation for ->condvar
949
950	sub condvar {
951	bless { @_ == 3 ? (_ae_cb => $_[2]) : () }, AnyEvent::CondVar::
952	}
953
954	# default implementation for ->signal
955
956	our %SIG_CB;
957
958	sub signal {
959	my (undef, %arg) = @_;
960
961	my $signal = uc $arg{signal}
962	or Carp::croak "required option 'signal' is missing";
963
964	$SIG_CB{$signal}{$arg{cb}} = $arg{cb};
965	$SIG{$signal} ||= sub {
966	$_->() for values %{ $SIG_CB{$signal} || {} };
967	};	1766	};
		1767	die if $@;
968		1768
969	bless [$signal, $arg{cb}], "AnyEvent::Base::Signal"	1769	&signal
970	}
971
972	sub AnyEvent::Base::Signal::DESTROY {
973	my ($signal, $cb) = @{$_[0]};
974
975	delete $SIG_CB{$signal}{$cb};
976
977	$SIG{$signal} = 'DEFAULT' unless keys %{ $SIG_CB{$signal} };
978	}	1770	}
979		1771
980	# default implementation for ->child	1772	# default implementation for ->child
981		1773
982	our %PID_CB;	1774	our %PID_CB;
983	our $CHLD_W;	1775	our $CHLD_W;
984	our $CHLD_DELAY_W;	1776	our $CHLD_DELAY_W;
985	our $PID_IDLE;
986	our $WNOHANG;
987		1777
988	sub _child_wait {	1778	# used by many Impl's
989	while (0 < (my $pid = waitpid -1, $WNOHANG)) {	1779	sub _emit_childstatus($$) {
		1780	my (undef, $rpid, $rstatus) = @_;
		1781
		1782	$_->($rpid, $rstatus)
990	$_->($pid, $?) for (values %{ $PID_CB{$pid} || {} }),	1783	for values %{ $PID_CB{$rpid} || {} },
991	(values %{ $PID_CB{0} || {} });	1784	values %{ $PID_CB{0} || {} };
992	}
993
994	undef $PID_IDLE;
995	}
996
997	sub _sigchld {
998	# make sure we deliver these changes "synchronous" with the event loop.
999	$CHLD_DELAY_W ||= AnyEvent->timer (after => 0, cb => sub {
1000	undef $CHLD_DELAY_W;
1001	&_child_wait;
1002	});
1003	}	1785	}
1004		1786
1005	sub child {	1787	sub child {
		1788	eval q{ # poor man's autoloading {}
		1789	*_sigchld = sub {
		1790	my $pid;
		1791
		1792	AnyEvent->_emit_childstatus ($pid, $?)
		1793	while ($pid = waitpid -1, WNOHANG) > 0;
		1794	};
		1795
		1796	*child = sub {
1006	my (undef, %arg) = @_;	1797	my (undef, %arg) = @_;
1007		1798
1008	defined (my $pid = $arg{pid} + 0)	1799	my $pid = $arg{pid};
1009	or Carp::croak "required option 'pid' is missing";	1800	my $cb = $arg{cb};
1010		1801
1011	$PID_CB{$pid}{$arg{cb}} = $arg{cb};	1802	$PID_CB{$pid}{$cb+0} = $cb;
1012		1803
1013	unless ($WNOHANG) {
1014	$WNOHANG = eval { local $SIG{__DIE__}; require POSIX; &POSIX::WNOHANG } || 1;
1015	}
1016
1017	unless ($CHLD_W) {	1804	unless ($CHLD_W) {
1018	$CHLD_W = AnyEvent->signal (signal => 'CHLD', cb => \&_sigchld);	1805	$CHLD_W = AE::signal CHLD => \&_sigchld;
1019	# child could be a zombie already, so make at least one round	1806	# child could be a zombie already, so make at least one round
1020	&_sigchld;	1807	&_sigchld;
1021	}	1808	}
1022		1809
1023	bless [$pid, $arg{cb}], "AnyEvent::Base::Child"	1810	bless [$pid, $cb+0], "AnyEvent::Base::child"
1024	}	1811	};
1025		1812
1026	sub AnyEvent::Base::Child::DESTROY {	1813	*AnyEvent::Base::child::DESTROY = sub {
1027	my ($pid, $cb) = @{$_[0]};	1814	my ($pid, $icb) = @{$_[0]};
1028		1815
1029	delete $PID_CB{$pid}{$cb};	1816	delete $PID_CB{$pid}{$icb};
1030	delete $PID_CB{$pid} unless keys %{ $PID_CB{$pid} };	1817	delete $PID_CB{$pid} unless keys %{ $PID_CB{$pid} };
1031		1818
1032	undef $CHLD_W unless keys %PID_CB;	1819	undef $CHLD_W unless keys %PID_CB;
		1820	};
		1821	};
		1822	die if $@;
		1823
		1824	&child
		1825	}
		1826
		1827	# idle emulation is done by simply using a timer, regardless
		1828	# of whether the process is idle or not, and not letting
		1829	# the callback use more than 50% of the time.
		1830	sub idle {
		1831	eval q{ # poor man's autoloading {}
		1832	*idle = sub {
		1833	my (undef, %arg) = @_;
		1834
		1835	my ($cb, $w, $rcb) = $arg{cb};
		1836
		1837	$rcb = sub {
		1838	if ($cb) {
		1839	$w = AE::time;
		1840	&$cb;
		1841	$w = AE::time - $w;
		1842
		1843	# never use more then 50% of the time for the idle watcher,
		1844	# within some limits
		1845	$w = 0.0001 if $w < 0.0001;
		1846	$w = 5 if $w > 5;
		1847
		1848	$w = AE::timer $w, 0, $rcb;
		1849	} else {
		1850	# clean up...
		1851	undef $w;
		1852	undef $rcb;
		1853	}
		1854	};
		1855
		1856	$w = AE::timer 0.05, 0, $rcb;
		1857
		1858	bless \\$cb, "AnyEvent::Base::idle"
		1859	};
		1860
		1861	*AnyEvent::Base::idle::DESTROY = sub {
		1862	undef $${$_[0]};
		1863	};
		1864	};
		1865	die if $@;
		1866
		1867	&idle
1033	}	1868	}
1034		1869
1035	package AnyEvent::CondVar;	1870	package AnyEvent::CondVar;
1036		1871
1037	our @ISA = AnyEvent::CondVar::Base::;	1872	our @ISA = AnyEvent::CondVar::Base::;
1038		1873
		1874	# only to be used for subclassing
		1875	sub new {
		1876	my $class = shift;
		1877	bless AnyEvent->condvar (@_), $class
		1878	}
		1879
1039	package AnyEvent::CondVar::Base;	1880	package AnyEvent::CondVar::Base;
1040		1881
1041	use overload	1882	#use overload
1042	'&{}' => sub { my $self = shift; sub { $self->send (@_) } },	1883	# '&{}' => sub { my $self = shift; sub { $self->send (@_) } },
1043	fallback => 1;	1884	# fallback => 1;
		1885
		1886	# save 300+ kilobytes by dirtily hardcoding overloading
		1887	${"AnyEvent::CondVar::Base::OVERLOAD"}{dummy}++; # Register with magic by touching.
		1888	*{'AnyEvent::CondVar::Base::()'} = sub { }; # "Make it findable via fetchmethod."
		1889	*{'AnyEvent::CondVar::Base::(&{}'} = sub { my $self = shift; sub { $self->send (@_) } }; # &{}
		1890	${'AnyEvent::CondVar::Base::()'} = 1; # fallback
		1891
		1892	our $WAITING;
1044		1893
1045	sub _send {	1894	sub _send {
1046	# nop	1895	# nop
		1896	}
		1897
		1898	sub _wait {
		1899	AnyEvent->_poll until $_[0]{_ae_sent};
1047	}	1900	}
1048		1901
1049	sub send {	1902	sub send {
1050	my $cv = shift;	1903	my $cv = shift;
1051	$cv->{_ae_sent} = [@_];	1904	$cv->{_ae_sent} = [@_];
…		…
1060		1913
1061	sub ready {	1914	sub ready {
1062	$_[0]{_ae_sent}	1915	$_[0]{_ae_sent}
1063	}	1916	}
1064		1917
1065	sub _wait {
1066	AnyEvent->one_event while !$_[0]{_ae_sent};
1067	}
1068
1069	sub recv {	1918	sub recv {
		1919	unless ($_[0]{_ae_sent}) {
		1920	$WAITING
		1921	and Carp::croak "AnyEvent::CondVar: recursive blocking wait attempted";
		1922
		1923	local $WAITING = 1;
1070	$_[0]->_wait;	1924	$_[0]->_wait;
		1925	}
1071		1926
1072	Carp::croak $_[0]{_ae_croak} if $_[0]{_ae_croak};	1927	$_[0]{_ae_croak}
1073	wantarray ? @{ $_[0]{_ae_sent} } : $_[0]{_ae_sent}[0]	1928	and Carp::croak $_[0]{_ae_croak};
		1929
		1930	wantarray
		1931	? @{ $_[0]{_ae_sent} }
		1932	: $_[0]{_ae_sent}[0]
1074	}	1933	}
1075		1934
1076	sub cb {	1935	sub cb {
1077	$_[0]{_ae_cb} = $_[1] if @_ > 1;	1936	my $cv = shift;
		1937
		1938	@_
		1939	and $cv->{_ae_cb} = shift
		1940	and $cv->{_ae_sent}
		1941	and (delete $cv->{_ae_cb})->($cv);
		1942
1078	$_[0]{_ae_cb}	1943	$cv->{_ae_cb}
1079	}	1944	}
1080		1945
1081	sub begin {	1946	sub begin {
1082	++$_[0]{_ae_counter};	1947	++$_[0]{_ae_counter};
1083	$_[0]{_ae_end_cb} = $_[1] if @_ > 1;	1948	$_[0]{_ae_end_cb} = $_[1] if @_ > 1;
…		…
1088	&{ $_[0]{_ae_end_cb} || sub { $_[0]->send } };	1953	&{ $_[0]{_ae_end_cb} || sub { $_[0]->send } };
1089	}	1954	}
1090		1955
1091	# undocumented/compatibility with pre-3.4	1956	# undocumented/compatibility with pre-3.4
1092	*broadcast = \&send;	1957	*broadcast = \&send;
1093	*wait = \&_wait;	1958	*wait = \&recv;
		1959
		1960	=head1 ERROR AND EXCEPTION HANDLING
		1961
		1962	In general, AnyEvent does not do any error handling - it relies on the
		1963	caller to do that if required. The L<AnyEvent::Strict> module (see also
		1964	the C<PERL_ANYEVENT_STRICT> environment variable, below) provides strict
		1965	checking of all AnyEvent methods, however, which is highly useful during
		1966	development.
		1967
		1968	As for exception handling (i.e. runtime errors and exceptions thrown while
		1969	executing a callback), this is not only highly event-loop specific, but
		1970	also not in any way wrapped by this module, as this is the job of the main
		1971	program.
		1972
		1973	The pure perl event loop simply re-throws the exception (usually
		1974	within C<< condvar->recv >>), the L<Event> and L<EV> modules call C<<
		1975	$Event/EV::DIED->() >>, L<Glib> uses C<< install_exception_handler >> and
		1976	so on.
		1977
		1978	=head1 ENVIRONMENT VARIABLES
		1979
		1980	The following environment variables are used by this module or its
		1981	submodules.
		1982
		1983	Note that AnyEvent will remove I<all> environment variables starting with
		1984	C<PERL_ANYEVENT_> from C<%ENV> when it is loaded while taint mode is
		1985	enabled.
		1986
		1987	=over 4
		1988
		1989	=item C<PERL_ANYEVENT_VERBOSE>
		1990
		1991	By default, AnyEvent will be completely silent except in fatal
		1992	conditions. You can set this environment variable to make AnyEvent more
		1993	talkative.
		1994
		1995	When set to C<5> or higher, causes AnyEvent to warn about unexpected
		1996	conditions, such as not being able to load the event model specified by
		1997	C<PERL_ANYEVENT_MODEL>.
		1998
		1999	When set to C<7> or higher, cause AnyEvent to report to STDERR which event
		2000	model it chooses.
		2001
		2002	When set to C<8> or higher, then AnyEvent will report extra information on
		2003	which optional modules it loads and how it implements certain features.
		2004
		2005	=item C<PERL_ANYEVENT_STRICT>
		2006
		2007	AnyEvent does not do much argument checking by default, as thorough
		2008	argument checking is very costly. Setting this variable to a true value
		2009	will cause AnyEvent to load C<AnyEvent::Strict> and then to thoroughly
		2010	check the arguments passed to most method calls. If it finds any problems,
		2011	it will croak.
		2012
		2013	In other words, enables "strict" mode.
		2014
		2015	Unlike C<use strict> (or its modern cousin, C<< use L<common::sense>
		2016	>>, it is definitely recommended to keep it off in production. Keeping
		2017	C<PERL_ANYEVENT_STRICT=1> in your environment while developing programs
		2018	can be very useful, however.
		2019
		2020	=item C<PERL_ANYEVENT_DEBUG_SHELL>
		2021
		2022	If this env variable is set, then its contents will be interpreted by
		2023	C<AnyEvent::Socket::parse_hostport> (after replacing every occurance of
		2024	C<$$> by the process pid) and an C<AnyEvent::Debug::shell> is bound on
		2025	that port. The shell object is saved in C<$AnyEvent::Debug::SHELL>.
		2026
		2027	This takes place when the first watcher is created.
		2028
		2029	For example, to bind a debug shell on a unix domain socket in
		2030	F<< /tmp/debug<pid>.sock >>, you could use this:
		2031
		2032	PERL_ANYEVENT_DEBUG_SHELL=/tmp/debug\$\$.sock perlprog
		2033
		2034	Note that creating sockets in F</tmp> is very unsafe on multiuser
		2035	systems.
		2036
		2037	=item C<PERL_ANYEVENT_DEBUG_WRAP>
		2038
		2039	Can be set to C<0>, C<1> or C<2> and enables wrapping of all watchers for
		2040	debugging purposes. See C<AnyEvent::Debug::wrap> for details.
		2041
		2042	=item C<PERL_ANYEVENT_MODEL>
		2043
		2044	This can be used to specify the event model to be used by AnyEvent, before
		2045	auto detection and -probing kicks in.
		2046
		2047	It normally is a string consisting entirely of ASCII letters (e.g. C<EV>
		2048	or C<IOAsync>). The string C<AnyEvent::Impl::> gets prepended and the
		2049	resulting module name is loaded and - if the load was successful - used as
		2050	event model backend. If it fails to load then AnyEvent will proceed with
		2051	auto detection and -probing.
		2052
		2053	If the string ends with C<::> instead (e.g. C<AnyEvent::Impl::EV::>) then
		2054	nothing gets prepended and the module name is used as-is (hint: C<::> at
		2055	the end of a string designates a module name and quotes it appropriately).
		2056
		2057	For example, to force the pure perl model (L<AnyEvent::Loop::Perl>) you
		2058	could start your program like this:
		2059
		2060	PERL_ANYEVENT_MODEL=Perl perl ...
		2061
		2062	=item C<PERL_ANYEVENT_PROTOCOLS>
		2063
		2064	Used by both L<AnyEvent::DNS> and L<AnyEvent::Socket> to determine preferences
		2065	for IPv4 or IPv6. The default is unspecified (and might change, or be the result
		2066	of auto probing).
		2067
		2068	Must be set to a comma-separated list of protocols or address families,
		2069	current supported: C<ipv4> and C<ipv6>. Only protocols mentioned will be
		2070	used, and preference will be given to protocols mentioned earlier in the
		2071	list.
		2072
		2073	This variable can effectively be used for denial-of-service attacks
		2074	against local programs (e.g. when setuid), although the impact is likely
		2075	small, as the program has to handle conenction and other failures anyways.
		2076
		2077	Examples: C<PERL_ANYEVENT_PROTOCOLS=ipv4,ipv6> - prefer IPv4 over IPv6,
		2078	but support both and try to use both. C<PERL_ANYEVENT_PROTOCOLS=ipv4>
		2079	- only support IPv4, never try to resolve or contact IPv6
		2080	addresses. C<PERL_ANYEVENT_PROTOCOLS=ipv6,ipv4> support either IPv4 or
		2081	IPv6, but prefer IPv6 over IPv4.
		2082
		2083	=item C<PERL_ANYEVENT_EDNS0>
		2084
		2085	Used by L<AnyEvent::DNS> to decide whether to use the EDNS0 extension
		2086	for DNS. This extension is generally useful to reduce DNS traffic, but
		2087	some (broken) firewalls drop such DNS packets, which is why it is off by
		2088	default.
		2089
		2090	Setting this variable to C<1> will cause L<AnyEvent::DNS> to announce
		2091	EDNS0 in its DNS requests.
		2092
		2093	=item C<PERL_ANYEVENT_MAX_FORKS>
		2094
		2095	The maximum number of child processes that C<AnyEvent::Util::fork_call>
		2096	will create in parallel.
		2097
		2098	=item C<PERL_ANYEVENT_MAX_OUTSTANDING_DNS>
		2099
		2100	The default value for the C<max_outstanding> parameter for the default DNS
		2101	resolver - this is the maximum number of parallel DNS requests that are
		2102	sent to the DNS server.
		2103
		2104	=item C<PERL_ANYEVENT_RESOLV_CONF>
		2105
		2106	The file to use instead of F</etc/resolv.conf> (or OS-specific
		2107	configuration) in the default resolver. When set to the empty string, no
		2108	default config will be used.
		2109
		2110	=item C<PERL_ANYEVENT_CA_FILE>, C<PERL_ANYEVENT_CA_PATH>.
		2111
		2112	When neither C<ca_file> nor C<ca_path> was specified during
		2113	L<AnyEvent::TLS> context creation, and either of these environment
		2114	variables exist, they will be used to specify CA certificate locations
		2115	instead of a system-dependent default.
		2116
		2117	=item C<PERL_ANYEVENT_AVOID_GUARD> and C<PERL_ANYEVENT_AVOID_ASYNC_INTERRUPT>
		2118
		2119	When these are set to C<1>, then the respective modules are not
		2120	loaded. Mostly good for testing AnyEvent itself.
		2121
		2122	=back
1094		2123
1095	=head1 SUPPLYING YOUR OWN EVENT MODEL INTERFACE	2124	=head1 SUPPLYING YOUR OWN EVENT MODEL INTERFACE
1096		2125
1097	This is an advanced topic that you do not normally need to use AnyEvent in	2126	This is an advanced topic that you do not normally need to use AnyEvent in
1098	a module. This section is only of use to event loop authors who want to	2127	a module. This section is only of use to event loop authors who want to
…		…
1132		2161
1133	I<rxvt-unicode> also cheats a bit by not providing blocking access to	2162	I<rxvt-unicode> also cheats a bit by not providing blocking access to
1134	condition variables: code blocking while waiting for a condition will	2163	condition variables: code blocking while waiting for a condition will
1135	C<die>. This still works with most modules/usages, and blocking calls must	2164	C<die>. This still works with most modules/usages, and blocking calls must
1136	not be done in an interactive application, so it makes sense.	2165	not be done in an interactive application, so it makes sense.
1137
1138	=head1 ENVIRONMENT VARIABLES
1139
1140	The following environment variables are used by this module:
1141
1142	=over 4
1143
1144	=item C<PERL_ANYEVENT_VERBOSE>
1145
1146	By default, AnyEvent will be completely silent except in fatal
1147	conditions. You can set this environment variable to make AnyEvent more
1148	talkative.
1149
1150	When set to C<1> or higher, causes AnyEvent to warn about unexpected
1151	conditions, such as not being able to load the event model specified by
1152	C<PERL_ANYEVENT_MODEL>.
1153
1154	When set to C<2> or higher, cause AnyEvent to report to STDERR which event
1155	model it chooses.
1156
1157	=item C<PERL_ANYEVENT_MODEL>
1158
1159	This can be used to specify the event model to be used by AnyEvent, before
1160	auto detection and -probing kicks in. It must be a string consisting
1161	entirely of ASCII letters. The string C<AnyEvent::Impl::> gets prepended
1162	and the resulting module name is loaded and if the load was successful,
1163	used as event model. If it fails to load AnyEvent will proceed with
1164	auto detection and -probing.
1165
1166	This functionality might change in future versions.
1167
1168	For example, to force the pure perl model (L<AnyEvent::Impl::Perl>) you
1169	could start your program like this:
1170
1171	PERL_ANYEVENT_MODEL=Perl perl ...
1172
1173	=item C<PERL_ANYEVENT_PROTOCOLS>
1174
1175	Used by both L<AnyEvent::DNS> and L<AnyEvent::Socket> to determine preferences
1176	for IPv4 or IPv6. The default is unspecified (and might change, or be the result
1177	of auto probing).
1178
1179	Must be set to a comma-separated list of protocols or address families,
1180	current supported: C<ipv4> and C<ipv6>. Only protocols mentioned will be
1181	used, and preference will be given to protocols mentioned earlier in the
1182	list.
1183
1184	This variable can effectively be used for denial-of-service attacks
1185	against local programs (e.g. when setuid), although the impact is likely
1186	small, as the program has to handle connection errors already-
1187
1188	Examples: C<PERL_ANYEVENT_PROTOCOLS=ipv4,ipv6> - prefer IPv4 over IPv6,
1189	but support both and try to use both. C<PERL_ANYEVENT_PROTOCOLS=ipv4>
1190	- only support IPv4, never try to resolve or contact IPv6
1191	addresses. C<PERL_ANYEVENT_PROTOCOLS=ipv6,ipv4> support either IPv4 or
1192	IPv6, but prefer IPv6 over IPv4.
1193
1194	=item C<PERL_ANYEVENT_EDNS0>
1195
1196	Used by L<AnyEvent::DNS> to decide whether to use the EDNS0 extension
1197	for DNS. This extension is generally useful to reduce DNS traffic, but
1198	some (broken) firewalls drop such DNS packets, which is why it is off by
1199	default.
1200
1201	Setting this variable to C<1> will cause L<AnyEvent::DNS> to announce
1202	EDNS0 in its DNS requests.
1203
1204	=item C<PERL_ANYEVENT_MAX_FORKS>
1205
1206	The maximum number of child processes that C<AnyEvent::Util::fork_call>
1207	will create in parallel.
1208
1209	=back
1210		2166
1211	=head1 EXAMPLE PROGRAM	2167	=head1 EXAMPLE PROGRAM
1212		2168
1213	The following program uses an I/O watcher to read data from STDIN, a timer	2169	The following program uses an I/O watcher to read data from STDIN, a timer
1214	to display a message once per second, and a condition variable to quit the	2170	to display a message once per second, and a condition variable to quit the
…		…
1227	warn "read: $input\n"; # output what has been read	2183	warn "read: $input\n"; # output what has been read
1228	$cv->send if $input =~ /^q/i; # quit program if /^q/i	2184	$cv->send if $input =~ /^q/i; # quit program if /^q/i
1229	},	2185	},
1230	);	2186	);
1231		2187
1232	my $time_watcher; # can only be used once
1233
1234	sub new_timer {
1235	$timer = AnyEvent->timer (after => 1, cb => sub {	2188	my $time_watcher = AnyEvent->timer (after => 1, interval => 1, cb => sub {
1236	warn "timeout\n"; # print 'timeout' about every second	2189	warn "timeout\n"; # print 'timeout' at most every second
1237	&new_timer; # and restart the time
1238	});	2190	});
1239	}
1240
1241	new_timer; # create first timer
1242		2191
1243	$cv->recv; # wait until user enters /^q/i	2192	$cv->recv; # wait until user enters /^q/i
1244		2193
1245	=head1 REAL-WORLD EXAMPLE	2194	=head1 REAL-WORLD EXAMPLE
1246		2195
…		…
1319		2268
1320	The actual code goes further and collects all errors (C<die>s, exceptions)	2269	The actual code goes further and collects all errors (C<die>s, exceptions)
1321	that occurred during request processing. The C<result> method detects	2270	that occurred during request processing. The C<result> method detects
1322	whether an exception as thrown (it is stored inside the $txn object)	2271	whether an exception as thrown (it is stored inside the $txn object)
1323	and just throws the exception, which means connection errors and other	2272	and just throws the exception, which means connection errors and other
1324	problems get reported tot he code that tries to use the result, not in a	2273	problems get reported to the code that tries to use the result, not in a
1325	random callback.	2274	random callback.
1326		2275
1327	All of this enables the following usage styles:	2276	All of this enables the following usage styles:
1328		2277
1329	1. Blocking:	2278	1. Blocking:
…		…
1377	through AnyEvent. The benchmark creates a lot of timers (with a zero	2326	through AnyEvent. The benchmark creates a lot of timers (with a zero
1378	timeout) and I/O watchers (watching STDOUT, a pty, to become writable,	2327	timeout) and I/O watchers (watching STDOUT, a pty, to become writable,
1379	which it is), lets them fire exactly once and destroys them again.	2328	which it is), lets them fire exactly once and destroys them again.
1380		2329
1381	Source code for this benchmark is found as F<eg/bench> in the AnyEvent	2330	Source code for this benchmark is found as F<eg/bench> in the AnyEvent
1382	distribution.	2331	distribution. It uses the L<AE> interface, which makes a real difference
		2332	for the EV and Perl backends only.
1383		2333
1384	=head3 Explanation of the columns	2334	=head3 Explanation of the columns
1385		2335
1386	I<watcher> is the number of event watchers created/destroyed. Since	2336	I<watcher> is the number of event watchers created/destroyed. Since
1387	different event models feature vastly different performances, each event	2337	different event models feature vastly different performances, each event
…		…
1408	watcher.	2358	watcher.
1409		2359
1410	=head3 Results	2360	=head3 Results
1411		2361
1412	name watchers bytes create invoke destroy comment	2362	name watchers bytes create invoke destroy comment
1413	EV/EV 400000 244 0.56 0.46 0.31 EV native interface	2363	EV/EV 100000 223 0.47 0.43 0.27 EV native interface
1414	EV/Any 100000 244 2.50 0.46 0.29 EV + AnyEvent watchers	2364	EV/Any 100000 223 0.48 0.42 0.26 EV + AnyEvent watchers
1415	CoroEV/Any 100000 244 2.49 0.44 0.29 coroutines + Coro::Signal	2365	Coro::EV/Any 100000 223 0.47 0.42 0.26 coroutines + Coro::Signal
1416	Perl/Any 100000 513 4.92 0.87 1.12 pure perl implementation	2366	Perl/Any 100000 431 2.70 0.74 0.92 pure perl implementation
1417	Event/Event 16000 516 31.88 31.30 0.85 Event native interface	2367	Event/Event 16000 516 31.16 31.84 0.82 Event native interface
1418	Event/Any 16000 590 35.75 31.42 1.08 Event + AnyEvent watchers	2368	Event/Any 16000 1203 42.61 34.79 1.80 Event + AnyEvent watchers
		2369	IOAsync/Any 16000 1911 41.92 27.45 16.81 via IO::Async::Loop::IO_Poll
		2370	IOAsync/Any 16000 1726 40.69 26.37 15.25 via IO::Async::Loop::Epoll
1419	Glib/Any 16000 1357 98.22 12.41 54.00 quadratic behaviour	2371	Glib/Any 16000 1118 89.00 12.57 51.17 quadratic behaviour
1420	Tk/Any 2000 1860 26.97 67.98 14.00 SEGV with >> 2000 watchers	2372	Tk/Any 2000 1346 20.96 10.75 8.00 SEGV with >> 2000 watchers
1421	POE/Event 2000 6644 108.64 736.02 14.73 via POE::Loop::Event	2373	POE/Any 2000 6951 108.97 795.32 14.24 via POE::Loop::Event
1422	POE/Select 2000 6343 94.13 809.12 565.96 via POE::Loop::Select	2374	POE/Any 2000 6648 94.79 774.40 575.51 via POE::Loop::Select
1423		2375
1424	=head3 Discussion	2376	=head3 Discussion
1425		2377
1426	The benchmark does I<not> measure scalability of the event loop very	2378	The benchmark does I<not> measure scalability of the event loop very
1427	well. For example, a select-based event loop (such as the pure perl one)	2379	well. For example, a select-based event loop (such as the pure perl one)
…		…
1439	benchmark machine, handling an event takes roughly 1600 CPU cycles with	2391	benchmark machine, handling an event takes roughly 1600 CPU cycles with
1440	EV, 3100 CPU cycles with AnyEvent's pure perl loop and almost 3000000 CPU	2392	EV, 3100 CPU cycles with AnyEvent's pure perl loop and almost 3000000 CPU
1441	cycles with POE.	2393	cycles with POE.
1442		2394
1443	C<EV> is the sole leader regarding speed and memory use, which are both	2395	C<EV> is the sole leader regarding speed and memory use, which are both
1444	maximal/minimal, respectively. Even when going through AnyEvent, it uses	2396	maximal/minimal, respectively. When using the L<AE> API there is zero
		2397	overhead (when going through the AnyEvent API create is about 5-6 times
		2398	slower, with other times being equal, so still uses far less memory than
1445	far less memory than any other event loop and is still faster than Event	2399	any other event loop and is still faster than Event natively).
1446	natively.
1447		2400
1448	The pure perl implementation is hit in a few sweet spots (both the	2401	The pure perl implementation is hit in a few sweet spots (both the
1449	constant timeout and the use of a single fd hit optimisations in the perl	2402	constant timeout and the use of a single fd hit optimisations in the perl
1450	interpreter and the backend itself). Nevertheless this shows that it	2403	interpreter and the backend itself). Nevertheless this shows that it
1451	adds very little overhead in itself. Like any select-based backend its	2404	adds very little overhead in itself. Like any select-based backend its
1452	performance becomes really bad with lots of file descriptors (and few of	2405	performance becomes really bad with lots of file descriptors (and few of
1453	them active), of course, but this was not subject of this benchmark.	2406	them active), of course, but this was not subject of this benchmark.
1454		2407
1455	The C<Event> module has a relatively high setup and callback invocation	2408	The C<Event> module has a relatively high setup and callback invocation
1456	cost, but overall scores in on the third place.	2409	cost, but overall scores in on the third place.
		2410
		2411	C<IO::Async> performs admirably well, about on par with C<Event>, even
		2412	when using its pure perl backend.
1457		2413
1458	C<Glib>'s memory usage is quite a bit higher, but it features a	2414	C<Glib>'s memory usage is quite a bit higher, but it features a
1459	faster callback invocation and overall ends up in the same class as	2415	faster callback invocation and overall ends up in the same class as
1460	C<Event>. However, Glib scales extremely badly, doubling the number of	2416	C<Event>. However, Glib scales extremely badly, doubling the number of
1461	watchers increases the processing time by more than a factor of four,	2417	watchers increases the processing time by more than a factor of four,
…		…
1496	(even when used without AnyEvent), but most event loops have acceptable	2452	(even when used without AnyEvent), but most event loops have acceptable
1497	performance with or without AnyEvent.	2453	performance with or without AnyEvent.
1498		2454
1499	=item * The overhead AnyEvent adds is usually much smaller than the overhead of	2455	=item * The overhead AnyEvent adds is usually much smaller than the overhead of
1500	the actual event loop, only with extremely fast event loops such as EV	2456	the actual event loop, only with extremely fast event loops such as EV
1501	adds AnyEvent significant overhead.	2457	does AnyEvent add significant overhead.
1502		2458
1503	=item * You should avoid POE like the plague if you want performance or	2459	=item * You should avoid POE like the plague if you want performance or
1504	reasonable memory usage.	2460	reasonable memory usage.
1505		2461
1506	=back	2462	=back
…		…
1522	In this benchmark, we use 10000 socket pairs (20000 sockets), of which 100	2478	In this benchmark, we use 10000 socket pairs (20000 sockets), of which 100
1523	(1%) are active. This mirrors the activity of large servers with many	2479	(1%) are active. This mirrors the activity of large servers with many
1524	connections, most of which are idle at any one point in time.	2480	connections, most of which are idle at any one point in time.
1525		2481
1526	Source code for this benchmark is found as F<eg/bench2> in the AnyEvent	2482	Source code for this benchmark is found as F<eg/bench2> in the AnyEvent
1527	distribution.	2483	distribution. It uses the L<AE> interface, which makes a real difference
		2484	for the EV and Perl backends only.
1528		2485
1529	=head3 Explanation of the columns	2486	=head3 Explanation of the columns
1530		2487
1531	I<sockets> is the number of sockets, and twice the number of "servers" (as	2488	I<sockets> is the number of sockets, and twice the number of "servers" (as
1532	each server has a read and write socket end).	2489	each server has a read and write socket end).
…		…
1539	it to another server. This includes deleting the old timeout and creating	2496	it to another server. This includes deleting the old timeout and creating
1540	a new one that moves the timeout into the future.	2497	a new one that moves the timeout into the future.
1541		2498
1542	=head3 Results	2499	=head3 Results
1543		2500
1544	name sockets create request	2501	name sockets create request
1545	EV 20000 69.01 11.16	2502	EV 20000 62.66 7.99
1546	Perl 20000 73.32 35.87	2503	Perl 20000 68.32 32.64
1547	Event 20000 212.62 257.32	2504	IOAsync 20000 174.06 101.15 epoll
1548	Glib 20000 651.16 1896.30	2505	IOAsync 20000 174.67 610.84 poll
		2506	Event 20000 202.69 242.91
		2507	Glib 20000 557.01 1689.52
1549	POE 20000 349.67 12317.24 uses POE::Loop::Event	2508	POE 20000 341.54 12086.32 uses POE::Loop::Event
1550		2509
1551	=head3 Discussion	2510	=head3 Discussion
1552		2511
1553	This benchmark I<does> measure scalability and overall performance of the	2512	This benchmark I<does> measure scalability and overall performance of the
1554	particular event loop.	2513	particular event loop.
…		…
1556	EV is again fastest. Since it is using epoll on my system, the setup time	2515	EV is again fastest. Since it is using epoll on my system, the setup time
1557	is relatively high, though.	2516	is relatively high, though.
1558		2517
1559	Perl surprisingly comes second. It is much faster than the C-based event	2518	Perl surprisingly comes second. It is much faster than the C-based event
1560	loops Event and Glib.	2519	loops Event and Glib.
		2520
		2521	IO::Async performs very well when using its epoll backend, and still quite
		2522	good compared to Glib when using its pure perl backend.
1561		2523
1562	Event suffers from high setup time as well (look at its code and you will	2524	Event suffers from high setup time as well (look at its code and you will
1563	understand why). Callback invocation also has a high overhead compared to	2525	understand why). Callback invocation also has a high overhead compared to
1564	the C<< $_->() for .. >>-style loop that the Perl event loop uses. Event	2526	the C<< $_->() for .. >>-style loop that the Perl event loop uses. Event
1565	uses select or poll in basically all documented configurations.	2527	uses select or poll in basically all documented configurations.
…		…
1628	=item * C-based event loops perform very well with small number of	2590	=item * C-based event loops perform very well with small number of
1629	watchers, as the management overhead dominates.	2591	watchers, as the management overhead dominates.
1630		2592
1631	=back	2593	=back
1632		2594
		2595	=head2 THE IO::Lambda BENCHMARK
		2596
		2597	Recently I was told about the benchmark in the IO::Lambda manpage, which
		2598	could be misinterpreted to make AnyEvent look bad. In fact, the benchmark
		2599	simply compares IO::Lambda with POE, and IO::Lambda looks better (which
		2600	shouldn't come as a surprise to anybody). As such, the benchmark is
		2601	fine, and mostly shows that the AnyEvent backend from IO::Lambda isn't
		2602	very optimal. But how would AnyEvent compare when used without the extra
		2603	baggage? To explore this, I wrote the equivalent benchmark for AnyEvent.
		2604
		2605	The benchmark itself creates an echo-server, and then, for 500 times,
		2606	connects to the echo server, sends a line, waits for the reply, and then
		2607	creates the next connection. This is a rather bad benchmark, as it doesn't
		2608	test the efficiency of the framework or much non-blocking I/O, but it is a
		2609	benchmark nevertheless.
		2610
		2611	name runtime
		2612	Lambda/select 0.330 sec
		2613	+ optimized 0.122 sec
		2614	Lambda/AnyEvent 0.327 sec
		2615	+ optimized 0.138 sec
		2616	Raw sockets/select 0.077 sec
		2617	POE/select, components 0.662 sec
		2618	POE/select, raw sockets 0.226 sec
		2619	POE/select, optimized 0.404 sec
		2620
		2621	AnyEvent/select/nb 0.085 sec
		2622	AnyEvent/EV/nb 0.068 sec
		2623	+state machine 0.134 sec
		2624
		2625	The benchmark is also a bit unfair (my fault): the IO::Lambda/POE
		2626	benchmarks actually make blocking connects and use 100% blocking I/O,
		2627	defeating the purpose of an event-based solution. All of the newly
		2628	written AnyEvent benchmarks use 100% non-blocking connects (using
		2629	AnyEvent::Socket::tcp_connect and the asynchronous pure perl DNS
		2630	resolver), so AnyEvent is at a disadvantage here, as non-blocking connects
		2631	generally require a lot more bookkeeping and event handling than blocking
		2632	connects (which involve a single syscall only).
		2633
		2634	The last AnyEvent benchmark additionally uses L<AnyEvent::Handle>, which
		2635	offers similar expressive power as POE and IO::Lambda, using conventional
		2636	Perl syntax. This means that both the echo server and the client are 100%
		2637	non-blocking, further placing it at a disadvantage.
		2638
		2639	As you can see, the AnyEvent + EV combination even beats the
		2640	hand-optimised "raw sockets benchmark", while AnyEvent + its pure perl
		2641	backend easily beats IO::Lambda and POE.
		2642
		2643	And even the 100% non-blocking version written using the high-level (and
		2644	slow :) L<AnyEvent::Handle> abstraction beats both POE and IO::Lambda
		2645	higher level ("unoptimised") abstractions by a large margin, even though
		2646	it does all of DNS, tcp-connect and socket I/O in a non-blocking way.
		2647
		2648	The two AnyEvent benchmarks programs can be found as F<eg/ae0.pl> and
		2649	F<eg/ae2.pl> in the AnyEvent distribution, the remaining benchmarks are
		2650	part of the IO::Lambda distribution and were used without any changes.
		2651
		2652
		2653	=head1 SIGNALS
		2654
		2655	AnyEvent currently installs handlers for these signals:
		2656
		2657	=over 4
		2658
		2659	=item SIGCHLD
		2660
		2661	A handler for C<SIGCHLD> is installed by AnyEvent's child watcher
		2662	emulation for event loops that do not support them natively. Also, some
		2663	event loops install a similar handler.
		2664
		2665	Additionally, when AnyEvent is loaded and SIGCHLD is set to IGNORE, then
		2666	AnyEvent will reset it to default, to avoid losing child exit statuses.
		2667
		2668	=item SIGPIPE
		2669
		2670	A no-op handler is installed for C<SIGPIPE> when C<$SIG{PIPE}> is C<undef>
		2671	when AnyEvent gets loaded.
		2672
		2673	The rationale for this is that AnyEvent users usually do not really depend
		2674	on SIGPIPE delivery (which is purely an optimisation for shell use, or
		2675	badly-written programs), but C<SIGPIPE> can cause spurious and rare
		2676	program exits as a lot of people do not expect C<SIGPIPE> when writing to
		2677	some random socket.
		2678
		2679	The rationale for installing a no-op handler as opposed to ignoring it is
		2680	that this way, the handler will be restored to defaults on exec.
		2681
		2682	Feel free to install your own handler, or reset it to defaults.
		2683
		2684	=back
		2685
		2686	=cut
		2687
		2688	undef $SIG{CHLD}
		2689	if $SIG{CHLD} eq 'IGNORE';
		2690
		2691	$SIG{PIPE} = sub { }
		2692	unless defined $SIG{PIPE};
		2693
		2694	=head1 RECOMMENDED/OPTIONAL MODULES
		2695
		2696	One of AnyEvent's main goals is to be 100% Pure-Perl(tm): only perl (and
		2697	its built-in modules) are required to use it.
		2698
		2699	That does not mean that AnyEvent won't take advantage of some additional
		2700	modules if they are installed.
		2701
		2702	This section explains which additional modules will be used, and how they
		2703	affect AnyEvent's operation.
		2704
		2705	=over 4
		2706
		2707	=item L<Async::Interrupt>
		2708
		2709	This slightly arcane module is used to implement fast signal handling: To
		2710	my knowledge, there is no way to do completely race-free and quick
		2711	signal handling in pure perl. To ensure that signals still get
		2712	delivered, AnyEvent will start an interval timer to wake up perl (and
		2713	catch the signals) with some delay (default is 10 seconds, look for
		2714	C<$AnyEvent::MAX_SIGNAL_LATENCY>).
		2715
		2716	If this module is available, then it will be used to implement signal
		2717	catching, which means that signals will not be delayed, and the event loop
		2718	will not be interrupted regularly, which is more efficient (and good for
		2719	battery life on laptops).
		2720
		2721	This affects not just the pure-perl event loop, but also other event loops
		2722	that have no signal handling on their own (e.g. Glib, Tk, Qt).
		2723
		2724	Some event loops (POE, Event, Event::Lib) offer signal watchers natively,
		2725	and either employ their own workarounds (POE) or use AnyEvent's workaround
		2726	(using C<$AnyEvent::MAX_SIGNAL_LATENCY>). Installing L<Async::Interrupt>
		2727	does nothing for those backends.
		2728
		2729	=item L<EV>
		2730
		2731	This module isn't really "optional", as it is simply one of the backend
		2732	event loops that AnyEvent can use. However, it is simply the best event
		2733	loop available in terms of features, speed and stability: It supports
		2734	the AnyEvent API optimally, implements all the watcher types in XS, does
		2735	automatic timer adjustments even when no monotonic clock is available,
		2736	can take avdantage of advanced kernel interfaces such as C<epoll> and
		2737	C<kqueue>, and is the fastest backend I<by far>. You can even embed
		2738	L<Glib>/L<Gtk2> in it (or vice versa, see L<EV::Glib> and L<Glib::EV>).
		2739
		2740	If you only use backends that rely on another event loop (e.g. C<Tk>),
		2741	then this module will do nothing for you.
		2742
		2743	=item L<Guard>
		2744
		2745	The guard module, when used, will be used to implement
		2746	C<AnyEvent::Util::guard>. This speeds up guards considerably (and uses a
		2747	lot less memory), but otherwise doesn't affect guard operation much. It is
		2748	purely used for performance.
		2749
		2750	=item L<JSON> and L<JSON::XS>
		2751
		2752	One of these modules is required when you want to read or write JSON data
		2753	via L<AnyEvent::Handle>. L<JSON> is also written in pure-perl, but can take
		2754	advantage of the ultra-high-speed L<JSON::XS> module when it is installed.
		2755
		2756	=item L<Net::SSLeay>
		2757
		2758	Implementing TLS/SSL in Perl is certainly interesting, but not very
		2759	worthwhile: If this module is installed, then L<AnyEvent::Handle> (with
		2760	the help of L<AnyEvent::TLS>), gains the ability to do TLS/SSL.
		2761
		2762	=item L<Time::HiRes>
		2763
		2764	This module is part of perl since release 5.008. It will be used when the
		2765	chosen event library does not come with a timing source of its own. The
		2766	pure-perl event loop (L<AnyEvent::Loop>) will additionally load it to
		2767	try to use a monotonic clock for timing stability.
		2768
		2769	=back
		2770
1633		2771
1634	=head1 FORK	2772	=head1 FORK
1635		2773
1636	Most event libraries are not fork-safe. The ones who are usually are	2774	Most event libraries are not fork-safe. The ones who are usually are
1637	because they rely on inefficient but fork-safe C<select> or C<poll>	2775	because they rely on inefficient but fork-safe C<select> or C<poll> calls
1638	calls. Only L<EV> is fully fork-aware.	2776	- higher performance APIs such as BSD's kqueue or the dreaded Linux epoll
		2777	are usually badly thought-out hacks that are incompatible with fork in
		2778	one way or another. Only L<EV> is fully fork-aware and ensures that you
		2779	continue event-processing in both parent and child (or both, if you know
		2780	what you are doing).
		2781
		2782	This means that, in general, you cannot fork and do event processing in
		2783	the child if the event library was initialised before the fork (which
		2784	usually happens when the first AnyEvent watcher is created, or the library
		2785	is loaded).
1639		2786
1640	If you have to fork, you must either do so I<before> creating your first	2787	If you have to fork, you must either do so I<before> creating your first
1641	watcher OR you must not use AnyEvent at all in the child.	2788	watcher OR you must not use AnyEvent at all in the child OR you must do
		2789	something completely out of the scope of AnyEvent.
		2790
		2791	The problem of doing event processing in the parent I<and> the child
		2792	is much more complicated: even for backends that I<are> fork-aware or
		2793	fork-safe, their behaviour is not usually what you want: fork clones all
		2794	watchers, that means all timers, I/O watchers etc. are active in both
		2795	parent and child, which is almost never what you want. USing C<exec>
		2796	to start worker children from some kind of manage rprocess is usually
		2797	preferred, because it is much easier and cleaner, at the expense of having
		2798	to have another binary.
1642		2799
1643		2800
1644	=head1 SECURITY CONSIDERATIONS	2801	=head1 SECURITY CONSIDERATIONS
1645		2802
1646	AnyEvent can be forced to load any event model via	2803	AnyEvent can be forced to load any event model via
…		…
1651	specified in the variable.	2808	specified in the variable.
1652		2809
1653	You can make AnyEvent completely ignore this variable by deleting it	2810	You can make AnyEvent completely ignore this variable by deleting it
1654	before the first watcher gets created, e.g. with a C<BEGIN> block:	2811	before the first watcher gets created, e.g. with a C<BEGIN> block:
1655		2812
1656	BEGIN { delete $ENV{PERL_ANYEVENT_MODEL} }	2813	BEGIN { delete $ENV{PERL_ANYEVENT_MODEL} }
1657		2814
1658	use AnyEvent;	2815	use AnyEvent;
1659		2816
1660	Similar considerations apply to $ENV{PERL_ANYEVENT_VERBOSE}, as that can	2817	Similar considerations apply to $ENV{PERL_ANYEVENT_VERBOSE}, as that can
1661	be used to probe what backend is used and gain other information (which is	2818	be used to probe what backend is used and gain other information (which is
1662	probably even less useful to an attacker than PERL_ANYEVENT_MODEL).	2819	probably even less useful to an attacker than PERL_ANYEVENT_MODEL), and
		2820	$ENV{PERL_ANYEVENT_STRICT}.
		2821
		2822	Note that AnyEvent will remove I<all> environment variables starting with
		2823	C<PERL_ANYEVENT_> from C<%ENV> when it is loaded while taint mode is
		2824	enabled.
		2825
		2826
		2827	=head1 BUGS
		2828
		2829	Perl 5.8 has numerous memleaks that sometimes hit this module and are hard
		2830	to work around. If you suffer from memleaks, first upgrade to Perl 5.10
		2831	and check wether the leaks still show up. (Perl 5.10.0 has other annoying
		2832	memleaks, such as leaking on C<map> and C<grep> but it is usually not as
		2833	pronounced).
1663		2834
1664		2835
1665	=head1 SEE ALSO	2836	=head1 SEE ALSO
1666		2837
1667	Utility functions: L<AnyEvent::Util>.	2838	Tutorial/Introduction: L<AnyEvent::Intro>.
1668		2839
1669	Event modules: L<EV>, L<EV::Glib>, L<Glib::EV>, L<Event>, L<Glib::Event>,	2840	FAQ: L<AnyEvent::FAQ>.
1670	L<Glib>, L<Tk>, L<Event::Lib>, L<Qt>, L<POE>.	2841
		2842	Utility functions: L<AnyEvent::Util> (misc. grab-bag), L<AnyEvent::Log>
		2843	(simply logging).
		2844
		2845	Development/Debugging: L<AnyEvent::Strict> (stricter checking),
		2846	L<AnyEvent::Debug> (interactive shell, watcher tracing).
		2847
		2848	Supported event modules: L<AnyEvent::Loop>, L<EV>, L<EV::Glib>,
		2849	L<Glib::EV>, L<Event>, L<Glib::Event>, L<Glib>, L<Tk>, L<Event::Lib>,
		2850	L<Qt>, L<POE>, L<FLTK>.
1671		2851
1672	Implementations: L<AnyEvent::Impl::EV>, L<AnyEvent::Impl::Event>,	2852	Implementations: L<AnyEvent::Impl::EV>, L<AnyEvent::Impl::Event>,
1673	L<AnyEvent::Impl::Glib>, L<AnyEvent::Impl::Tk>, L<AnyEvent::Impl::Perl>,	2853	L<AnyEvent::Impl::Glib>, L<AnyEvent::Impl::Tk>, L<AnyEvent::Impl::Perl>,
1674	L<AnyEvent::Impl::EventLib>, L<AnyEvent::Impl::Qt>,	2854	L<AnyEvent::Impl::EventLib>, L<AnyEvent::Impl::Qt>,
		2855	L<AnyEvent::Impl::POE>, L<AnyEvent::Impl::IOAsync>, L<Anyevent::Impl::Irssi>,
1675	L<AnyEvent::Impl::POE>.	2856	L<AnyEvent::Impl::FLTK>.
1676		2857
1677	Non-blocking file handles, sockets, TCP clients and	2858	Non-blocking handles, pipes, stream sockets, TCP clients and
1678	servers: L<AnyEvent::Handle>, L<AnyEvent::Socket>.	2859	servers: L<AnyEvent::Handle>, L<AnyEvent::Socket>, L<AnyEvent::TLS>.
1679		2860
1680	Asynchronous DNS: L<AnyEvent::DNS>.	2861	Asynchronous DNS: L<AnyEvent::DNS>.
1681		2862
1682	Coroutine support: L<Coro>, L<Coro::AnyEvent>, L<Coro::EV>, L<Coro::Event>,	2863	Thread support: L<Coro>, L<Coro::AnyEvent>, L<Coro::EV>, L<Coro::Event>.
1683		2864
1684	Nontrivial usage examples: L<Net::FCP>, L<Net::XMPP2>, L<AnyEvent::DNS>.	2865	Nontrivial usage examples: L<AnyEvent::GPSD>, L<AnyEvent::IRC>,
		2866	L<AnyEvent::HTTP>.
1685		2867
1686		2868
1687	=head1 AUTHOR	2869	=head1 AUTHOR
1688		2870
1689	Marc Lehmann <schmorp@schmorp.de>	2871	Marc Lehmann <schmorp@schmorp.de>
1690	http://home.schmorp.de/	2872	http://home.schmorp.de/
1691		2873
1692	=cut	2874	=cut
1693		2875
1694	1	2876	1
1695		2877

Diff Legend

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