ViewVC Help

View File | Revision Log | Show Annotations | Download File

/cvs/AnyEvent/lib/AnyEvent.pm

Comparing AnyEvent/lib/AnyEvent.pm (file contents):
Revision 1.48 by root, Mon Apr 14 19:00:23 2008 UTC vs.
Revision 1.410 by root, Fri Mar 1 06:03:21 2013 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	Event, Coro, Glib, Tk, Perl - 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 wether a condition was flagged	37	my $w = AnyEvent->condvar; # stores whether a condition was flagged
		38	$w->send; # wake up current and all future recv's
20	$w->wait; # enters "main loop" till $condvar gets ->broadcast	39	$w->recv; # enters "main loop" till $condvar gets ->send
21	$w->broadcast; # wake up current and all future wait's	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, and AnyEvent	70	the statement "there can only be one" is a bitter reality: In general,
35	helps hiding the differences.	71	only one event loop can be active at the same time in a process. AnyEvent
		72	cannot change this, but it can hide the differences between those event
		73	loops.
36		74
37	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
38	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
39	religion, a way of living, and most importantly: without forcing your	77	religion, a way of living, and most importantly: without forcing your
40	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
41	model you use.	79	model you use.
42		80
43	For modules like POE or IO::Async (which is actually doing all I/O	81	For modules like POE or IO::Async (which is a total misnomer as it is
44	I<synchronously>...), using them in your module is like joining a	82	actually doing all I/O I<synchronously>...), using them in your module is
45	cult: After you joined, you are dependent on them and you cannot use	83	like joining a cult: After you join, you are dependent on them and you
46	anything else, as it is simply incompatible to everything that isn't	84	cannot use anything else, as they are simply incompatible to everything
47	itself.	85	that isn't them. What's worse, all the potential users of your
		86	module are I<also> forced to use the same event loop you use.
48		87
49	AnyEvent + POE works fine. AnyEvent + Glib works fine. AnyEvent + Tk	88	AnyEvent is different: AnyEvent + POE works fine. AnyEvent + Glib works
50	works fine etc. etc. but none of these work together with the rest: POE	89	fine. AnyEvent + Tk works fine etc. etc. but none of these work together
51	+ IO::Async? no go. Tk + Event? no go. If your module uses one of	90	with the rest: POE + EV? No go. Tk + Event? No go. Again: if your module
52	those, every user of your module has to use it, too. If your module	91	uses one of those, every user of your module has to use it, too. But if
53	uses AnyEvent, it works transparently with all event models it supports	92	your module uses AnyEvent, it works transparently with all event models it
54	(including stuff like POE and IO::Async).	93	supports (including stuff like IO::Async, as long as those use one of the
		94	supported event loops. It is easy to add new event loops to AnyEvent, too,
		95	so it is future-proof).
55		96
56	In addition of 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
57	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
58	modules, you get an enourmous amount of code and strict rules you have	99	modules, you get an enormous amount of code and strict rules you have to
59	to follow. AnyEvent, on the other hand, is lean and to the point by only	100	follow. AnyEvent, on the other hand, is lean and to the point, by only
60	offering the functionality that is useful, in as thin as a wrapper as	101	offering the functionality that is necessary, in as thin as a wrapper as
61	technically possible.	102	technically possible.
62		103
		104	Of course, AnyEvent comes with a big (and fully optional!) toolbox
		105	of useful functionality, such as an asynchronous DNS resolver, 100%
		106	non-blocking connects (even with TLS/SSL, IPv6 and on broken platforms
		107	such as Windows) and lots of real-world knowledge and workarounds for
		108	platform bugs and differences.
		109
63	Of course, if you want lots of policy (this can arguably be somewhat	110	Now, if you I<do want> lots of policy (this can arguably be somewhat
64	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
65	model, you should I<not> use this module.	112	model, you should I<not> use this module.
66		113
67
68	=head1 DESCRIPTION	114	=head1 DESCRIPTION
69		115
70	L<AnyEvent> provides an identical interface to multiple event loops. This	116	L<AnyEvent> provides a uniform interface to various event loops. This
71	allows module authors to utilise an event loop without forcing module	117	allows module authors to use event loop functionality without forcing
72	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
73	peacefully at any one time).	119	than one event loop cannot coexist peacefully).
74		120
75	The interface itself is vaguely similar but not identical to the Event	121	The interface itself is vaguely similar, but not identical to the L<Event>
76	module.	122	module.
77		123
78	On the first call of any method, the module tries to detect the currently	124	During the first call of any watcher-creation method, the module tries
79	loaded event loop by probing wether any of the following modules is	125	to detect the currently loaded event loop by probing whether one of the
80	loaded: L<Coro::Event>, L<Event>, L<Glib>, L<Tk>. The first one found is	126	following modules is already loaded: L<EV>, L<AnyEvent::Loop>,
81	used. If none is found, the module tries to load these modules in the	127	L<Event>, L<Glib>, L<Tk>, L<Event::Lib>, L<Qt>, L<POE>. The first one
82	order given. The first one that could be successfully loaded will be	128	found is used. If none are detected, the module tries to load the first
83	used. If still none could be found, AnyEvent will fall back to a pure-perl	129	four modules in the order given; but note that if L<EV> is not
84	event loop, which is also not very efficient.	130	available, the pure-perl L<AnyEvent::Loop> should always work, so
		131	the other two are not normally tried.
85		132
86	Because AnyEvent first checks for modules that are already loaded, loading	133	Because AnyEvent first checks for modules that are already loaded, loading
87	an Event model explicitly before first using AnyEvent will likely make	134	an event model explicitly before first using AnyEvent will likely make
88	that model the default. For example:	135	that model the default. For example:
89		136
90	use Tk;	137	use Tk;
91	use AnyEvent;	138	use AnyEvent;
92		139
93	# .. AnyEvent will likely default to Tk	140	# .. AnyEvent will likely default to Tk
94		141
		142	The I<likely> means that, if any module loads another event model and
		143	starts using it, all bets are off - this case should be very rare though,
		144	as very few modules hardcode event loops without announcing this very
		145	loudly.
		146
95	The pure-perl implementation of AnyEvent is called	147	The pure-perl implementation of AnyEvent is called C<AnyEvent::Loop>. Like
96	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
97	explicitly.	149	availability of that event loop :)
98		150
99	=head1 WATCHERS	151	=head1 WATCHERS
100		152
101	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
102	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
103	the callback to call, the filehandle to watch, etc.	155	the callback to call, the file handle to watch, etc.
104		156
105	These watchers are normal Perl objects with normal Perl lifetime. After	157	These watchers are normal Perl objects with normal Perl lifetime. After
106	creating a watcher it will immediately "watch" for events and invoke	158	creating a watcher it will immediately "watch" for events and invoke the
		159	callback when the event occurs (of course, only when the event model
		160	is in control).
		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
107	the callback. To disable the watcher you have to destroy it (e.g. by	168	To disable a watcher you have to destroy it (e.g. by setting the
108	setting the variable that stores it to C<undef> or otherwise deleting all	169	variable you store it in to C<undef> or otherwise deleting all references
109	references to it).	170	to it).
110		171
111	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.
112		173
		174	Many watchers either are used with "recursion" (repeating timers for
		175	example), or need to refer to their watcher object in other ways.
		176
		177	One way to achieve that is this pattern:
		178
		179	my $w; $w = AnyEvent->type (arg => value ..., cb => sub {
		180	# you can use $w here, for example to undef it
		181	undef $w;
		182	});
		183
		184	Note that C<my $w; $w => combination. This is necessary because in Perl,
		185	my variables are only visible after the statement in which they are
		186	declared.
		187
113	=head2 IO WATCHERS	188	=head2 I/O WATCHERS
114		189
		190	$w = AnyEvent->io (
		191	fh => <filehandle_or_fileno>,
		192	poll => <"r" or "w">,
		193	cb => <callback>,
		194	);
		195
115	You can create I/O watcher by calling the C<< AnyEvent->io >> method with	196	You can create an I/O watcher by calling the C<< AnyEvent->io >> method
116	the following mandatory arguments:	197	with the following mandatory key-value pairs as arguments:
117		198
118	C<fh> the Perl I<filehandle> (not filedescriptor) to watch for	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
119	events. C<poll> must be a string that is either C<r> or C<w>, that creates	206	C<poll> must be a string that is either C<r> or C<w>, which creates a
120	a watcher waiting for "r"eadable or "w"ritable events. C<cb> the callback	207	watcher waiting for "r"eadable or "w"ritable events, respectively.
121	to invoke everytime the filehandle becomes ready.
122		208
123	Only one io watcher per C<fh> and C<poll> combination is allowed (i.e. on	209	C<cb> is the callback to invoke each time the file handle becomes ready.
124	a socket you can have one r + one w, not any more (limitation comes from
125	Tk - if you are sure you are not using Tk this limitation is gone).
126		210
127	Filehandles will be kept alive, so as long as the watcher exists, the	211	Although the callback might get passed parameters, their value and
128	filehandle exists, too.	212	presence is undefined and you cannot rely on them. Portable AnyEvent
		213	callbacks cannot use arguments passed to I/O watcher callbacks.
129		214
130	Example:	215	The I/O watcher might use the underlying file descriptor or a copy of it.
		216	You must not close a file handle as long as any watcher is active on the
		217	underlying file descriptor.
131		218
		219	Some event loops issue spurious readiness notifications, so you should
		220	always use non-blocking calls when reading/writing from/to your file
		221	handles.
		222
132	# 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
133	my $w; $w = AnyEvent->io (fh => \*STDIN, poll => 'r', cb => sub {	226	my $w; $w = AnyEvent->io (fh => \*STDIN, poll => 'r', cb => sub {
134	chomp (my $input = <STDIN>);	227	chomp (my $input = <STDIN>);
135	warn "read: $input\n";	228	warn "read: $input\n";
136	undef $w;	229	undef $w;
137	});	230	});
138		231
139	=head2 TIME WATCHERS	232	=head2 TIME WATCHERS
140		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
141	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 >>
142	method with the following mandatory arguments:	243	method with the following mandatory arguments:
143		244
144	C<after> after how many seconds (fractions are supported) should the timer	245	C<after> specifies after how many seconds (fractional values are
145	activate. C<cb> the callback to invoke.	246	supported) the callback should be invoked. C<cb> is the callback to invoke
		247	in that case.
146		248
147	The timer callback will be invoked at most once: if you want a repeating	249	Although the callback might get passed parameters, their value and
148	timer you have to create a new watcher (this is a limitation by both Tk	250	presence is undefined and you cannot rely on them. Portable AnyEvent
149	and Glib).	251	callbacks cannot use arguments passed to time watcher callbacks.
150		252
151	Example:	253	The callback will normally be invoked only once. If you specify another
		254	parameter, C<interval>, as a strictly positive number (> 0), then the
		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.
152		258
		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.
		262
153	# fire an event after 7.7 seconds	263	Example: fire an event after 7.7 seconds.
		264
154	my $w = AnyEvent->timer (after => 7.7, cb => sub {	265	my $w = AnyEvent->timer (after => 7.7, cb => sub {
155	warn "timeout\n";	266	warn "timeout\n";
156	});	267	});
157		268
158	# to cancel the timer:	269	# to cancel the timer:
159	undef $w;	270	undef $w;
160		271
161	=head2 CONDITION WATCHERS	272	Example 2: fire an event after 0.5 seconds, then roughly every second.
162		273
163	Condition watchers can be created by calling the C<< AnyEvent->condvar >>	274	my $w = AnyEvent->timer (after => 0.5, interval => 1, cb => sub {
164	method without any arguments.	275	warn "timeout\n";
		276	};
165		277
166	A condition watcher watches for a condition - precisely that the C<<	278	=head3 TIMING ISSUES
167	->broadcast >> method has been called.
168		279
169	Note that condition watchers recurse into the event loop - if you have	280	There are two ways to handle timers: based on real time (relative, "fire
170	two watchers that call C<< ->wait >> in a round-robbin fashion, you	281	in 10 seconds") and based on wallclock time (absolute, "fire at 12
171	lose. Therefore, condition watchers are good to export to your caller, but	282	o'clock").
172	you should avoid making a blocking wait, at least in callbacks, as this
173	usually asks for trouble.
174		283
175	The watcher has only two methods:	284	While most event loops expect timers to specified in a relative way, they
		285	use absolute time internally. This makes a difference when your clock
		286	"jumps", for example, when ntp decides to set your clock backwards from
		287	the wrong date of 2014-01-01 to 2008-01-01, a watcher that is supposed to
		288	fire "after a second" might actually take six years to finally fire.
		289
		290	AnyEvent cannot compensate for this. The only event loop that is conscious
		291	of these issues is L<EV>, which offers both relative (ev_timer, based
		292	on true relative time) and absolute (ev_periodic, based on wallclock time)
		293	timers.
		294
		295	AnyEvent always prefers relative timers, if available, matching the
		296	AnyEvent API.
		297
		298	AnyEvent has two additional methods that return the "current time":
176		299
177	=over 4	300	=over 4
178		301
		302	=item AnyEvent->time
		303
		304	This returns the "current wallclock time" as a fractional number of
		305	seconds since the Epoch (the same thing as C<time> or C<Time::HiRes::time>
		306	return, and the result is guaranteed to be compatible with those).
		307
		308	It progresses independently of any event loop processing, i.e. each call
		309	will check the system clock, which usually gets updated frequently.
		310
		311	=item AnyEvent->now
		312
		313	This also returns the "current wallclock time", but unlike C<time>, above,
		314	this value might change only once per event loop iteration, depending on
		315	the event loop (most return the same time as C<time>, above). This is the
		316	time that AnyEvent's timers get scheduled against.
		317
		318	I<In almost all cases (in all cases if you don't care), this is the
		319	function to call when you want to know the current time.>
		320
		321	This function is also often faster then C<< AnyEvent->time >>, and
		322	thus the preferred method if you want some timestamp (for example,
		323	L<AnyEvent::Handle> uses this to update its activity timeouts).
		324
		325	The rest of this section is only of relevance if you try to be very exact
		326	with your timing; you can skip it without a bad conscience.
		327
		328	For a practical example of when these times differ, consider L<Event::Lib>
		329	and L<EV> and the following set-up:
		330
		331	The event loop is running and has just invoked one of your callbacks at
		332	time=500 (assume no other callbacks delay processing). In your callback,
		333	you wait a second by executing C<sleep 1> (blocking the process for a
		334	second) and then (at time=501) you create a relative timer that fires
		335	after three seconds.
		336
		337	With L<Event::Lib>, C<< AnyEvent->time >> and C<< AnyEvent->now >> will
		338	both return C<501>, because that is the current time, and the timer will
		339	be scheduled to fire at time=504 (C<501> + C<3>).
		340
		341	With L<EV>, C<< AnyEvent->time >> returns C<501> (as that is the current
		342	time), but C<< AnyEvent->now >> returns C<500>, as that is the time the
		343	last event processing phase started. With L<EV>, your timer gets scheduled
		344	to run at time=503 (C<500> + C<3>).
		345
		346	In one sense, L<Event::Lib> is more exact, as it uses the current time
		347	regardless of any delays introduced by event processing. However, most
		348	callbacks do not expect large delays in processing, so this causes a
		349	higher drift (and a lot more system calls to get the current time).
		350
		351	In another sense, L<EV> is more exact, as your timer will be scheduled at
		352	the same time, regardless of how long event processing actually took.
		353
		354	In either case, if you care (and in most cases, you don't), then you
		355	can get whatever behaviour you want with any event loop, by taking the
		356	difference between C<< AnyEvent->time >> and C<< AnyEvent->now >> into
		357	account.
		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
		381	=back
		382
		383	=head2 SIGNAL WATCHERS
		384
		385	$w = AnyEvent->signal (signal => <uppercase_signal_name>, cb => <callback>);
		386
		387	You can watch for signals using a signal watcher, C<signal> is the signal
		388	I<name> in uppercase and without any C<SIG> prefix, C<cb> is the Perl
		389	callback to be invoked whenever a signal occurs.
		390
		391	Although the callback might get passed parameters, their value and
		392	presence is undefined and you cannot rely on them. Portable AnyEvent
		393	callbacks cannot use arguments passed to signal watcher callbacks.
		394
		395	Multiple signal occurrences can be clumped together into one callback
		396	invocation, and callback invocation will be synchronous. Synchronous means
		397	that it might take a while until the signal gets handled by the process,
		398	but it is guaranteed not to interrupt any other callbacks.
		399
		400	The main advantage of using these watchers is that you can share a signal
		401	between multiple watchers, and AnyEvent will ensure that signals will not
		402	interrupt your program at bad times.
		403
		404	This watcher might use C<%SIG> (depending on the event loop used),
		405	so programs overwriting those signals directly will likely not work
		406	correctly.
		407
		408	Example: exit on SIGINT
		409
		410	my $w = AnyEvent->signal (signal => "INT", cb => sub { exit 1 });
		411
		412	=head3 Restart Behaviour
		413
		414	While restart behaviour is up to the event loop implementation, most will
		415	not restart syscalls (that includes L<Async::Interrupt> and AnyEvent's
		416	pure perl implementation).
		417
		418	=head3 Safe/Unsafe Signals
		419
		420	Perl signals can be either "safe" (synchronous to opcode handling)
		421	or "unsafe" (asynchronous) - the former might delay signal delivery
		422	indefinitely, the latter might corrupt your memory.
		423
		424	AnyEvent signal handlers are, in addition, synchronous to the event loop,
		425	i.e. they will not interrupt your running perl program but will only be
		426	called as part of the normal event handling (just like timer, I/O etc.
		427	callbacks, too).
		428
		429	=head3 Signal Races, Delays and Workarounds
		430
		431	Many event loops (e.g. Glib, Tk, Qt, IO::Async) do not support
		432	attaching callbacks to signals in a generic way, which is a pity,
		433	as you cannot do race-free signal handling in perl, requiring
		434	C libraries for this. AnyEvent will try to do its best, which
		435	means in some cases, signals will be delayed. The maximum time
		436	a signal might be delayed is 10 seconds by default, but can
		437	be overriden via C<$ENV{PERL_ANYEVENT_MAX_SIGNAL_LATENCY}> or
		438	C<$AnyEvent::MAX_SIGNAL_LATENCY> - see the L<ENVIRONMENT VARIABLES>
		439	section for details.
		440
		441	All these problems can be avoided by installing the optional
		442	L<Async::Interrupt> module, which works with most event loops. It will not
		443	work with inherently broken event loops such as L<Event> or L<Event::Lib>
		444	(and not with L<POE> currently). For those, you just have to suffer the
		445	delays.
		446
		447	=head2 CHILD PROCESS WATCHERS
		448
		449	$w = AnyEvent->child (pid => <process id>, cb => <callback>);
		450
		451	You can also watch for a child process exit and catch its exit status.
		452
		453	The child process is specified by the C<pid> argument (on some backends,
		454	using C<0> watches for any child process exit, on others this will
		455	croak). The watcher will be triggered only when the child process has
		456	finished and an exit status is available, not on any trace events
		457	(stopped/continued).
		458
		459	The callback will be called with the pid and exit status (as returned by
		460	waitpid), so unlike other watcher types, you I<can> rely on child watcher
		461	callback arguments.
		462
		463	This watcher type works by installing a signal handler for C<SIGCHLD>,
		464	and since it cannot be shared, nothing else should use SIGCHLD or reap
		465	random child processes (waiting for specific child processes, e.g. inside
		466	C<system>, is just fine).
		467
		468	There is a slight catch to child watchers, however: you usually start them
		469	I<after> the child process was created, and this means the process could
		470	have exited already (and no SIGCHLD will be sent anymore).
		471
		472	Not all event models handle this correctly (neither POE nor IO::Async do,
		473	see their AnyEvent::Impl manpages for details), but even for event models
		474	that I<do> handle this correctly, they usually need to be loaded before
		475	the process exits (i.e. before you fork in the first place). AnyEvent's
		476	pure perl event loop handles all cases correctly regardless of when you
		477	start the watcher.
		478
		479	This means you cannot create a child watcher as the very first
		480	thing in an AnyEvent program, you I<have> to create at least one
		481	watcher before you C<fork> the child (alternatively, you can call
		482	C<AnyEvent::detect>).
		483
		484	As most event loops do not support waiting for child events, they will be
		485	emulated by AnyEvent in most cases, in which case the latency and race
		486	problems mentioned in the description of signal watchers apply.
		487
		488	Example: fork a process and wait for it
		489
		490	my $done = AnyEvent->condvar;
		491
		492	my $pid = fork or exit 5;
		493
		494	my $w = AnyEvent->child (
		495	pid => $pid,
		496	cb => sub {
		497	my ($pid, $status) = @_;
		498	warn "pid $pid exited with status $status";
		499	$done->send;
		500	},
		501	);
		502
		503	# do something else, then wait for process exit
		504	$done->recv;
		505
		506	=head2 IDLE WATCHERS
		507
		508	$w = AnyEvent->idle (cb => <callback>);
		509
		510	This will repeatedly invoke the callback after the process becomes idle,
		511	until either the watcher is destroyed or new events have been detected.
		512
		513	Idle watchers are useful when there is a need to do something, but it
		514	is not so important (or wise) to do it instantly. The callback will be
		515	invoked only when there is "nothing better to do", which is usually
		516	defined as "all outstanding events have been handled and no new events
		517	have been detected". That means that idle watchers ideally get invoked
		518	when the event loop has just polled for new events but none have been
		519	detected. Instead of blocking to wait for more events, the idle watchers
		520	will be invoked.
		521
		522	Unfortunately, most event loops do not really support idle watchers (only
		523	EV, Event and Glib do it in a usable fashion) - for the rest, AnyEvent
		524	will simply call the callback "from time to time".
		525
		526	Example: read lines from STDIN, but only process them when the
		527	program is otherwise idle:
		528
		529	my @lines; # read data
		530	my $idle_w;
		531	my $io_w = AnyEvent->io (fh => \*STDIN, poll => 'r', cb => sub {
		532	push @lines, scalar <STDIN>;
		533
		534	# start an idle watcher, if not already done
		535	$idle_w ||= AnyEvent->idle (cb => sub {
		536	# handle only one line, when there are lines left
		537	if (my $line = shift @lines) {
		538	print "handled when idle: $line";
		539	} else {
		540	# otherwise disable the idle watcher again
		541	undef $idle_w;
		542	}
		543	});
		544	});
		545
		546	=head2 CONDITION VARIABLES
		547
		548	$cv = AnyEvent->condvar;
		549
		550	$cv->send (<list>);
		551	my @res = $cv->recv;
		552
		553	If you are familiar with some event loops you will know that all of them
		554	require you to run some blocking "loop", "run" or similar function that
		555	will actively watch for new events and call your callbacks.
		556
		557	AnyEvent is slightly different: it expects somebody else to run the event
		558	loop and will only block when necessary (usually when told by the user).
		559
		560	The tool to do that is called a "condition variable", so called because
		561	they represent a condition that must become true.
		562
		563	Now is probably a good time to look at the examples further below.
		564
		565	Condition variables can be created by calling the C<< AnyEvent->condvar
		566	>> method, usually without arguments. The only argument pair allowed is
		567	C<cb>, which specifies a callback to be called when the condition variable
		568	becomes true, with the condition variable as the first argument (but not
		569	the results).
		570
		571	After creation, the condition variable is "false" until it becomes "true"
		572	by calling the C<send> method (or calling the condition variable as if it
		573	were a callback, read about the caveats in the description for the C<<
		574	->send >> method).
		575
		576	Since condition variables are the most complex part of the AnyEvent API, here are
		577	some different mental models of what they are - pick the ones you can connect to:
		578
		579	=over 4
		580
		581	=item * Condition variables are like callbacks - you can call them (and pass them instead
		582	of callbacks). Unlike callbacks however, you can also wait for them to be called.
		583
		584	=item * Condition variables are signals - one side can emit or send them,
		585	the other side can wait for them, or install a handler that is called when
		586	the signal fires.
		587
		588	=item * Condition variables are like "Merge Points" - points in your program
		589	where you merge multiple independent results/control flows into one.
		590
		591	=item * Condition variables represent a transaction - functions that start
		592	some kind of transaction can return them, leaving the caller the choice
		593	between waiting in a blocking fashion, or setting a callback.
		594
		595	=item * Condition variables represent future values, or promises to deliver
		596	some result, long before the result is available.
		597
		598	=back
		599
		600	Condition variables are very useful to signal that something has finished,
		601	for example, if you write a module that does asynchronous http requests,
		602	then a condition variable would be the ideal candidate to signal the
		603	availability of results. The user can either act when the callback is
		604	called or can synchronously C<< ->recv >> for the results.
		605
		606	You can also use them to simulate traditional event loops - for example,
		607	you can block your main program until an event occurs - for example, you
		608	could C<< ->recv >> in your main program until the user clicks the Quit
		609	button of your app, which would C<< ->send >> the "quit" event.
		610
		611	Note that condition variables recurse into the event loop - if you have
		612	two pieces of code that call C<< ->recv >> in a round-robin fashion, you
		613	lose. Therefore, condition variables are good to export to your caller, but
		614	you should avoid making a blocking wait yourself, at least in callbacks,
		615	as this asks for trouble.
		616
		617	Condition variables are represented by hash refs in perl, and the keys
		618	used by AnyEvent itself are all named C<_ae_XXX> to make subclassing
		619	easy (it is often useful to build your own transaction class on top of
		620	AnyEvent). To subclass, use C<AnyEvent::CondVar> as base class and call
		621	its C<new> method in your own C<new> method.
		622
		623	There are two "sides" to a condition variable - the "producer side" which
		624	eventually calls C<< -> send >>, and the "consumer side", which waits
		625	for the send to occur.
		626
		627	Example: wait for a timer.
		628
		629	# condition: "wait till the timer is fired"
		630	my $timer_fired = AnyEvent->condvar;
		631
		632	# create the timer - we could wait for, say
		633	# a handle becomign ready, or even an
		634	# AnyEvent::HTTP request to finish, but
		635	# in this case, we simply use a timer:
		636	my $w = AnyEvent->timer (
		637	after => 1,
		638	cb => sub { $timer_fired->send },
		639	);
		640
		641	# this "blocks" (while handling events) till the callback
		642	# calls ->send
		643	$timer_fired->recv;
		644
		645	Example: wait for a timer, but take advantage of the fact that condition
		646	variables are also callable directly.
		647
		648	my $done = AnyEvent->condvar;
		649	my $delay = AnyEvent->timer (after => 5, cb => $done);
		650	$done->recv;
		651
		652	Example: Imagine an API that returns a condvar and doesn't support
		653	callbacks. This is how you make a synchronous call, for example from
		654	the main program:
		655
		656	use AnyEvent::CouchDB;
		657
		658	...
		659
		660	my @info = $couchdb->info->recv;
		661
		662	And this is how you would just set a callback to be called whenever the
		663	results are available:
		664
		665	$couchdb->info->cb (sub {
		666	my @info = $_[0]->recv;
		667	});
		668
		669	=head3 METHODS FOR PRODUCERS
		670
		671	These methods should only be used by the producing side, i.e. the
		672	code/module that eventually sends the signal. Note that it is also
		673	the producer side which creates the condvar in most cases, but it isn't
		674	uncommon for the consumer to create it as well.
		675
		676	=over 4
		677
		678	=item $cv->send (...)
		679
		680	Flag the condition as ready - a running C<< ->recv >> and all further
		681	calls to C<recv> will (eventually) return after this method has been
		682	called. If nobody is waiting the send will be remembered.
		683
		684	If a callback has been set on the condition variable, it is called
		685	immediately from within send.
		686
		687	Any arguments passed to the C<send> call will be returned by all
		688	future C<< ->recv >> calls.
		689
		690	Condition variables are overloaded so one can call them directly (as if
		691	they were a code reference). Calling them directly is the same as calling
		692	C<send>.
		693
		694	=item $cv->croak ($error)
		695
		696	Similar to send, but causes all calls to C<< ->recv >> to invoke
		697	C<Carp::croak> with the given error message/object/scalar.
		698
		699	This can be used to signal any errors to the condition variable
		700	user/consumer. Doing it this way instead of calling C<croak> directly
		701	delays the error detection, but has the overwhelming advantage that it
		702	diagnoses the error at the place where the result is expected, and not
		703	deep in some event callback with no connection to the actual code causing
		704	the problem.
		705
		706	=item $cv->begin ([group callback])
		707
179	=item $cv->wait	708	=item $cv->end
180		709
181	Wait (blocking if necessary) until the C<< ->broadcast >> method has been	710	These two methods can be used to combine many transactions/events into
182	called on c<$cv>, while servicing other watchers normally.	711	one. For example, a function that pings many hosts in parallel might want
		712	to use a condition variable for the whole process.
183		713
		714	Every call to C<< ->begin >> will increment a counter, and every call to
		715	C<< ->end >> will decrement it. If the counter reaches C<0> in C<< ->end
		716	>>, the (last) callback passed to C<begin> will be executed, passing the
		717	condvar as first argument. That callback is I<supposed> to call C<< ->send
		718	>>, but that is not required. If no group callback was set, C<send> will
		719	be called without any arguments.
		720
		721	You can think of C<< $cv->send >> giving you an OR condition (one call
		722	sends), while C<< $cv->begin >> and C<< $cv->end >> giving you an AND
		723	condition (all C<begin> calls must be C<end>'ed before the condvar sends).
		724
		725	Let's start with a simple example: you have two I/O watchers (for example,
		726	STDOUT and STDERR for a program), and you want to wait for both streams to
		727	close before activating a condvar:
		728
		729	my $cv = AnyEvent->condvar;
		730
		731	$cv->begin; # first watcher
		732	my $w1 = AnyEvent->io (fh => $fh1, cb => sub {
		733	defined sysread $fh1, my $buf, 4096
		734	or $cv->end;
		735	});
		736
		737	$cv->begin; # second watcher
		738	my $w2 = AnyEvent->io (fh => $fh2, cb => sub {
		739	defined sysread $fh2, my $buf, 4096
		740	or $cv->end;
		741	});
		742
		743	$cv->recv;
		744
		745	This works because for every event source (EOF on file handle), there is
		746	one call to C<begin>, so the condvar waits for all calls to C<end> before
		747	sending.
		748
		749	The ping example mentioned above is slightly more complicated, as the
		750	there are results to be passwd back, and the number of tasks that are
		751	begun can potentially be zero:
		752
		753	my $cv = AnyEvent->condvar;
		754
		755	my %result;
		756	$cv->begin (sub { shift->send (\%result) });
		757
		758	for my $host (@list_of_hosts) {
		759	$cv->begin;
		760	ping_host_then_call_callback $host, sub {
		761	$result{$host} = ...;
		762	$cv->end;
		763	};
		764	}
		765
		766	$cv->end;
		767
		768	...
		769
		770	my $results = $cv->recv;
		771
		772	This code fragment supposedly pings a number of hosts and calls
		773	C<send> after results for all then have have been gathered - in any
		774	order. To achieve this, the code issues a call to C<begin> when it starts
		775	each ping request and calls C<end> when it has received some result for
		776	it. Since C<begin> and C<end> only maintain a counter, the order in which
		777	results arrive is not relevant.
		778
		779	There is an additional bracketing call to C<begin> and C<end> outside the
		780	loop, which serves two important purposes: first, it sets the callback
		781	to be called once the counter reaches C<0>, and second, it ensures that
		782	C<send> is called even when C<no> hosts are being pinged (the loop
		783	doesn't execute once).
		784
		785	This is the general pattern when you "fan out" into multiple (but
		786	potentially zero) subrequests: use an outer C<begin>/C<end> pair to set
		787	the callback and ensure C<end> is called at least once, and then, for each
		788	subrequest you start, call C<begin> and for each subrequest you finish,
		789	call C<end>.
		790
		791	=back
		792
		793	=head3 METHODS FOR CONSUMERS
		794
		795	These methods should only be used by the consuming side, i.e. the
		796	code awaits the condition.
		797
		798	=over 4
		799
		800	=item $cv->recv
		801
		802	Wait (blocking if necessary) until the C<< ->send >> or C<< ->croak
		803	>> methods have been called on C<$cv>, while servicing other watchers
		804	normally.
		805
184	You can only wait once on a condition - additional calls will return	806	You can only wait once on a condition - additional calls are valid but
185	immediately.	807	will return immediately.
		808
		809	If an error condition has been set by calling C<< ->croak >>, then this
		810	function will call C<croak>.
		811
		812	In list context, all parameters passed to C<send> will be returned,
		813	in scalar context only the first one will be returned.
		814
		815	Note that doing a blocking wait in a callback is not supported by any
		816	event loop, that is, recursive invocation of a blocking C<< ->recv >> is
		817	not allowed and the C<recv> call will C<croak> if such a condition is
		818	detected. This requirement can be dropped by relying on L<Coro::AnyEvent>
		819	, which allows you to do a blocking C<< ->recv >> from any thread
		820	that doesn't run the event loop itself. L<Coro::AnyEvent> is loaded
		821	automatically when L<Coro> is used with L<AnyEvent>, so code does not need
		822	to do anything special to take advantage of that: any code that would
		823	normally block your program because it calls C<recv>, be executed in an
		824	C<async> thread instead without blocking other threads.
186		825
187	Not all event models support a blocking wait - some die in that case	826	Not all event models support a blocking wait - some die in that case
188	(programs might want to do that so they stay interactive), so I<if you	827	(programs might want to do that to stay interactive), so I<if you are
189	are using this from a module, never require a blocking wait>, but let the	828	using this from a module, never require a blocking wait>. Instead, let the
190	caller decide wether the call will block or not (for example, by coupling	829	caller decide whether the call will block or not (for example, by coupling
191	condition variables with some kind of request results and supporting	830	condition variables with some kind of request results and supporting
192	callbacks so the caller knows that getting the result will not block,	831	callbacks so the caller knows that getting the result will not block,
193	while still suppporting blocking waits if the caller so desires).	832	while still supporting blocking waits if the caller so desires).
194		833
195	Another reason I<never> to C<< ->wait >> in a module is that you cannot	834	You can ensure that C<< ->recv >> never blocks by setting a callback and
196	sensibly have two C<< ->wait >>'s in parallel, as that would require	835	only calling C<< ->recv >> from within that callback (or at a later
197	multiple interpreters or coroutines/threads, none of which C<AnyEvent>	836	time). This will work even when the event loop does not support blocking
198	can supply (the coroutine-aware backends C<Coro::EV> and C<Coro::Event>	837	waits otherwise.
199	explicitly support concurrent C<< ->wait >>'s from different coroutines,
200	however).
201		838
202	=item $cv->broadcast	839	=item $bool = $cv->ready
203		840
204	Flag the condition as ready - a running C<< ->wait >> and all further	841	Returns true when the condition is "true", i.e. whether C<send> or
205	calls to C<wait> will return after this method has been called. If nobody	842	C<croak> have been called.
206	is waiting the broadcast will be remembered..
207		843
208	Example:	844	=item $cb = $cv->cb ($cb->($cv))
209		845
210	# wait till the result is ready	846	This is a mutator function that returns the callback set and optionally
211	my $result_ready = AnyEvent->condvar;	847	replaces it before doing so.
212		848
213	# do something such as adding a timer	849	The callback will be called when the condition becomes "true", i.e. when
214	# or socket watcher the calls $result_ready->broadcast	850	C<send> or C<croak> are called, with the only argument being the
215	# when the "result" is ready.	851	condition variable itself. If the condition is already true, the
216		852	callback is called immediately when it is set. Calling C<recv> inside
217	$result_ready->wait;	853	the callback or at any later time is guaranteed not to block.
218		854
219	=back	855	=back
220		856
221	=head2 SIGNAL WATCHERS	857	=head1 SUPPORTED EVENT LOOPS/BACKENDS
222		858
223	You can listen for signals using a signal watcher, C<signal> is the signal	859	The available backend classes are (every class has its own manpage):
224	I<name> without any C<SIG> prefix. Multiple signals events can be clumped
225	together into one callback invocation, and callback invocation might or
226	might not be asynchronous.
227
228	These watchers might use C<%SIG>, so programs overwriting those signals
229	directly will likely not work correctly.
230
231	Example: exit on SIGINT
232
233	my $w = AnyEvent->signal (signal => "INT", cb => sub { exit 1 });
234
235	=head2 CHILD PROCESS WATCHERS
236
237	You can also listen for the status of a child process specified by the
238	C<pid> argument (or any child if the pid argument is 0). The watcher will
239	trigger as often as status change for the child are received. This works
240	by installing a signal handler for C<SIGCHLD>. The callback will be called with
241	the pid and exit status (as returned by waitpid).
242
243	Example: wait for pid 1333
244
245	my $w = AnyEvent->child (pid => 1333, cb => sub { warn "exit status $?" });
246
247	=head1 GLOBALS
248		860
249	=over 4	861	=over 4
250		862
		863	=item Backends that are autoprobed when no other event loop can be found.
		864
		865	EV is the preferred backend when no other event loop seems to be in
		866	use. If EV is not installed, then AnyEvent will fall back to its own
		867	pure-perl implementation, which is available everywhere as it comes with
		868	AnyEvent itself.
		869
		870	AnyEvent::Impl::EV based on EV (interface to libev, best choice).
		871	AnyEvent::Impl::Perl pure-perl AnyEvent::Loop, fast and portable.
		872
		873	=item Backends that are transparently being picked up when they are used.
		874
		875	These will be used if they are already loaded when the first watcher
		876	is created, in which case it is assumed that the application is using
		877	them. This means that AnyEvent will automatically pick the right backend
		878	when the main program loads an event module before anything starts to
		879	create watchers. Nothing special needs to be done by the main program.
		880
		881	AnyEvent::Impl::Event based on Event, very stable, few glitches.
		882	AnyEvent::Impl::Glib based on Glib, slow but very stable.
		883	AnyEvent::Impl::Tk based on Tk, very broken.
		884	AnyEvent::Impl::EventLib based on Event::Lib, leaks memory and worse.
		885	AnyEvent::Impl::POE based on POE, very slow, some limitations.
		886	AnyEvent::Impl::Irssi used when running within irssi.
		887	AnyEvent::Impl::IOAsync based on IO::Async.
		888	AnyEvent::Impl::Cocoa based on Cocoa::EventLoop.
		889	AnyEvent::Impl::FLTK based on FLTK (fltk 2 binding).
		890
		891	=item Backends with special needs.
		892
		893	Qt requires the Qt::Application to be instantiated first, but will
		894	otherwise be picked up automatically. As long as the main program
		895	instantiates the application before any AnyEvent watchers are created,
		896	everything should just work.
		897
		898	AnyEvent::Impl::Qt based on Qt.
		899
		900	=item Event loops that are indirectly supported via other backends.
		901
		902	Some event loops can be supported via other modules:
		903
		904	There is no direct support for WxWidgets (L<Wx>) or L<Prima>.
		905
		906	B<WxWidgets> has no support for watching file handles. However, you can
		907	use WxWidgets through the POE adaptor, as POE has a Wx backend that simply
		908	polls 20 times per second, which was considered to be too horrible to even
		909	consider for AnyEvent.
		910
		911	B<Prima> is not supported as nobody seems to be using it, but it has a POE
		912	backend, so it can be supported through POE.
		913
		914	AnyEvent knows about both L<Prima> and L<Wx>, however, and will try to
		915	load L<POE> when detecting them, in the hope that POE will pick them up,
		916	in which case everything will be automatic.
		917
		918	=back
		919
		920	=head1 GLOBAL VARIABLES AND FUNCTIONS
		921
		922	These are not normally required to use AnyEvent, but can be useful to
		923	write AnyEvent extension modules.
		924
		925	=over 4
		926
251	=item $AnyEvent::MODEL	927	=item $AnyEvent::MODEL
252		928
253	Contains C<undef> until the first watcher is being created. Then it	929	Contains C<undef> until the first watcher is being created, before the
		930	backend has been autodetected.
		931
254	contains the event model that is being used, which is the name of the	932	Afterwards it contains the event model that is being used, which is the
255	Perl class implementing the model. This class is usually one of the	933	name of the Perl class implementing the model. This class is usually one
256	C<AnyEvent::Impl:xxx> modules, but can be any other class in the case	934	of the C<AnyEvent::Impl::xxx> modules, but can be any other class in the
257	AnyEvent has been extended at runtime (e.g. in I<rxvt-unicode>).	935	case AnyEvent has been extended at runtime (e.g. in I<rxvt-unicode> it
258		936	will be C<urxvt::anyevent>).
259	The known classes so far are:
260
261	AnyEvent::Impl::CoroEV based on Coro::EV, best choice.
262	AnyEvent::Impl::EV based on EV (an interface to libev, also best choice).
263	AnyEvent::Impl::CoroEvent based on Coro::Event, second best choice.
264	AnyEvent::Impl::Event based on Event, also second best choice :)
265	AnyEvent::Impl::Glib based on Glib, third-best choice.
266	AnyEvent::Impl::Tk based on Tk, very bad choice.
267	AnyEvent::Impl::Perl pure-perl implementation, inefficient but portable.
268		937
269	=item AnyEvent::detect	938	=item AnyEvent::detect
270		939
271	Returns C<$AnyEvent::MODEL>, forcing autodetection of the event model if	940	Returns C<$AnyEvent::MODEL>, forcing autodetection of the event model
272	necessary. You should only call this function right before you would have	941	if necessary. You should only call this function right before you would
273	created an AnyEvent watcher anyway, that is, very late at runtime.	942	have created an AnyEvent watcher anyway, that is, as late as possible at
		943	runtime, and not e.g. during initialisation of your module.
		944
		945	The effect of calling this function is as if a watcher had been created
		946	(specifically, actions that happen "when the first watcher is created"
		947	happen when calling detetc as well).
		948
		949	If you need to do some initialisation before AnyEvent watchers are
		950	created, use C<post_detect>.
		951
		952	=item $guard = AnyEvent::post_detect { BLOCK }
		953
		954	Arranges for the code block to be executed as soon as the event model is
		955	autodetected (or immediately if that has already happened).
		956
		957	The block will be executed I<after> the actual backend has been detected
		958	(C<$AnyEvent::MODEL> is set), but I<before> any watchers have been
		959	created, so it is possible to e.g. patch C<@AnyEvent::ISA> or do
		960	other initialisations - see the sources of L<AnyEvent::Strict> or
		961	L<AnyEvent::AIO> to see how this is used.
		962
		963	The most common usage is to create some global watchers, without forcing
		964	event module detection too early, for example, L<AnyEvent::AIO> creates
		965	and installs the global L<IO::AIO> watcher in a C<post_detect> block to
		966	avoid autodetecting the event module at load time.
		967
		968	If called in scalar or list context, then it creates and returns an object
		969	that automatically removes the callback again when it is destroyed (or
		970	C<undef> when the hook was immediately executed). See L<AnyEvent::AIO> for
		971	a case where this is useful.
		972
		973	Example: Create a watcher for the IO::AIO module and store it in
		974	C<$WATCHER>, but do so only do so after the event loop is initialised.
		975
		976	our WATCHER;
		977
		978	my $guard = AnyEvent::post_detect {
		979	$WATCHER = AnyEvent->io (fh => IO::AIO::poll_fileno, poll => 'r', cb => \&IO::AIO::poll_cb);
		980	};
		981
		982	# the ||= is important in case post_detect immediately runs the block,
		983	# as to not clobber the newly-created watcher. assigning both watcher and
		984	# post_detect guard to the same variable has the advantage of users being
		985	# able to just C<undef $WATCHER> if the watcher causes them grief.
		986
		987	$WATCHER ||= $guard;
		988
		989	=item @AnyEvent::post_detect
		990
		991	If there are any code references in this array (you can C<push> to it
		992	before or after loading AnyEvent), then they will be called directly
		993	after the event loop has been chosen.
		994
		995	You should check C<$AnyEvent::MODEL> before adding to this array, though:
		996	if it is defined then the event loop has already been detected, and the
		997	array will be ignored.
		998
		999	Best use C<AnyEvent::post_detect { BLOCK }> when your application allows
		1000	it, as it takes care of these details.
		1001
		1002	This variable is mainly useful for modules that can do something useful
		1003	when AnyEvent is used and thus want to know when it is initialised, but do
		1004	not need to even load it by default. This array provides the means to hook
		1005	into AnyEvent passively, without loading it.
		1006
		1007	Example: To load Coro::AnyEvent whenever Coro and AnyEvent are used
		1008	together, you could put this into Coro (this is the actual code used by
		1009	Coro to accomplish this):
		1010
		1011	if (defined $AnyEvent::MODEL) {
		1012	# AnyEvent already initialised, so load Coro::AnyEvent
		1013	require Coro::AnyEvent;
		1014	} else {
		1015	# AnyEvent not yet initialised, so make sure to load Coro::AnyEvent
		1016	# as soon as it is
		1017	push @AnyEvent::post_detect, sub { require Coro::AnyEvent };
		1018	}
		1019
		1020	=item AnyEvent::postpone { BLOCK }
		1021
		1022	Arranges for the block to be executed as soon as possible, but not before
		1023	the call itself returns. In practise, the block will be executed just
		1024	before the event loop polls for new events, or shortly afterwards.
		1025
		1026	This function never returns anything (to make the C<return postpone { ...
		1027	}> idiom more useful.
		1028
		1029	To understand the usefulness of this function, consider a function that
		1030	asynchronously does something for you and returns some transaction
		1031	object or guard to let you cancel the operation. For example,
		1032	C<AnyEvent::Socket::tcp_connect>:
		1033
		1034	# start a conenction attempt unless one is active
		1035	$self->{connect_guard} ||= AnyEvent::Socket::tcp_connect "www.example.net", 80, sub {
		1036	delete $self->{connect_guard};
		1037	...
		1038	};
		1039
		1040	Imagine that this function could instantly call the callback, for
		1041	example, because it detects an obvious error such as a negative port
		1042	number. Invoking the callback before the function returns causes problems
		1043	however: the callback will be called and will try to delete the guard
		1044	object. But since the function hasn't returned yet, there is nothing to
		1045	delete. When the function eventually returns it will assign the guard
		1046	object to C<< $self->{connect_guard} >>, where it will likely never be
		1047	deleted, so the program thinks it is still trying to connect.
		1048
		1049	This is where C<AnyEvent::postpone> should be used. Instead of calling the
		1050	callback directly on error:
		1051
		1052	$cb->(undef), return # signal error to callback, BAD!
		1053	if $some_error_condition;
		1054
		1055	It should use C<postpone>:
		1056
		1057	AnyEvent::postpone { $cb->(undef) }, return # signal error to callback, later
		1058	if $some_error_condition;
		1059
		1060	=item AnyEvent::log $level, $msg[, @args]
		1061
		1062	Log the given C<$msg> at the given C<$level>.
		1063
		1064	If L<AnyEvent::Log> is not loaded then this function makes a simple test
		1065	to see whether the message will be logged. If the test succeeds it will
		1066	load AnyEvent::Log and call C<AnyEvent::Log::log> - consequently, look at
		1067	the L<AnyEvent::Log> documentation for details.
		1068
		1069	If the test fails it will simply return. Right now this happens when a
		1070	numerical loglevel is used and it is larger than the level specified via
		1071	C<$ENV{PERL_ANYEVENT_VERBOSE}>.
		1072
		1073	If you want to sprinkle loads of logging calls around your code, consider
		1074	creating a logger callback with the C<AnyEvent::Log::logger> function,
		1075	which can reduce typing, codesize and can reduce the logging overhead
		1076	enourmously.
274		1077
275	=back	1078	=back
276		1079
277	=head1 WHAT TO DO IN A MODULE	1080	=head1 WHAT TO DO IN A MODULE
278		1081
279	As a module author, you should "use AnyEvent" and call AnyEvent methods	1082	As a module author, you should C<use AnyEvent> and call AnyEvent methods
280	freely, but you should not load a specific event module or rely on it.	1083	freely, but you should not load a specific event module or rely on it.
281		1084
282	Be careful when you create watchers in the module body - Anyevent will	1085	Be careful when you create watchers in the module body - AnyEvent will
283	decide which event module to use as soon as the first method is called, so	1086	decide which event module to use as soon as the first method is called, so
284	by calling AnyEvent in your module body you force the user of your module	1087	by calling AnyEvent in your module body you force the user of your module
285	to load the event module first.	1088	to load the event module first.
286		1089
		1090	Never call C<< ->recv >> on a condition variable unless you I<know> that
		1091	the C<< ->send >> method has been called on it already. This is
		1092	because it will stall the whole program, and the whole point of using
		1093	events is to stay interactive.
		1094
		1095	It is fine, however, to call C<< ->recv >> when the user of your module
		1096	requests it (i.e. if you create a http request object ad have a method
		1097	called C<results> that returns the results, it may call C<< ->recv >>
		1098	freely, as the user of your module knows what she is doing. Always).
		1099
287	=head1 WHAT TO DO IN THE MAIN PROGRAM	1100	=head1 WHAT TO DO IN THE MAIN PROGRAM
288		1101
289	There will always be a single main program - the only place that should	1102	There will always be a single main program - the only place that should
290	dictate which event model to use.	1103	dictate which event model to use.
291		1104
292	If it doesn't care, it can just "use AnyEvent" and use it itself, or not	1105	If the program is not event-based, it need not do anything special, even
293	do anything special and let AnyEvent decide which implementation to chose.	1106	when it depends on a module that uses an AnyEvent. If the program itself
		1107	uses AnyEvent, but does not care which event loop is used, all it needs
		1108	to do is C<use AnyEvent>. In either case, AnyEvent will choose the best
		1109	available loop implementation.
294		1110
295	If the main program relies on a specific event model (for example, in Gtk2	1111	If the main program relies on a specific event model - for example, in
296	programs you have to rely on either Glib or Glib::Event), you should load	1112	Gtk2 programs you have to rely on the Glib module - you should load the
297	it before loading AnyEvent or any module that uses it, generally, as early	1113	event module before loading AnyEvent or any module that uses it: generally
298	as possible. The reason is that modules might create watchers when they	1114	speaking, you should load it as early as possible. The reason is that
299	are loaded, and AnyEvent will decide on the event model to use as soon as	1115	modules might create watchers when they are loaded, and AnyEvent will
300	it creates watchers, and it might chose the wrong one unless you load the	1116	decide on the event model to use as soon as it creates watchers, and it
301	correct one yourself.	1117	might choose the wrong one unless you load the correct one yourself.
302		1118
303	You can chose to use a rather inefficient pure-perl implementation by	1119	You can chose to use a pure-perl implementation by loading the
304	loading the C<AnyEvent::Impl::Perl> module, but letting AnyEvent chose is	1120	C<AnyEvent::Loop> module, which gives you similar behaviour
305	generally better.	1121	everywhere, but letting AnyEvent chose the model is generally better.
		1122
		1123	=head2 MAINLOOP EMULATION
		1124
		1125	Sometimes (often for short test scripts, or even standalone programs who
		1126	only want to use AnyEvent), you do not want to run a specific event loop.
		1127
		1128	In that case, you can use a condition variable like this:
		1129
		1130	AnyEvent->condvar->recv;
		1131
		1132	This has the effect of entering the event loop and looping forever.
		1133
		1134	Note that usually your program has some exit condition, in which case
		1135	it is better to use the "traditional" approach of storing a condition
		1136	variable somewhere, waiting for it, and sending it when the program should
		1137	exit cleanly.
		1138
		1139
		1140	=head1 OTHER MODULES
		1141
		1142	The following is a non-exhaustive list of additional modules that use
		1143	AnyEvent as a client and can therefore be mixed easily with other
		1144	AnyEvent modules and other event loops in the same program. Some of the
		1145	modules come as part of AnyEvent, the others are available via CPAN (see
		1146	L<http://search.cpan.org/search?m=module&q=anyevent%3A%3A*> for
		1147	a longer non-exhaustive list), and the list is heavily biased towards
		1148	modules of the AnyEvent author himself :)
		1149
		1150	=over 4
		1151
		1152	=item L<AnyEvent::Util>
		1153
		1154	Contains various utility functions that replace often-used blocking
		1155	functions such as C<inet_aton> with event/callback-based versions.
		1156
		1157	=item L<AnyEvent::Socket>
		1158
		1159	Provides various utility functions for (internet protocol) sockets,
		1160	addresses and name resolution. Also functions to create non-blocking tcp
		1161	connections or tcp servers, with IPv6 and SRV record support and more.
		1162
		1163	=item L<AnyEvent::Handle>
		1164
		1165	Provide read and write buffers, manages watchers for reads and writes,
		1166	supports raw and formatted I/O, I/O queued and fully transparent and
		1167	non-blocking SSL/TLS (via L<AnyEvent::TLS>).
		1168
		1169	=item L<AnyEvent::DNS>
		1170
		1171	Provides rich asynchronous DNS resolver capabilities.
		1172
		1173	=item L<AnyEvent::HTTP>, L<AnyEvent::IRC>, L<AnyEvent::XMPP>, L<AnyEvent::GPSD>, L<AnyEvent::IGS>, L<AnyEvent::FCP>
		1174
		1175	Implement event-based interfaces to the protocols of the same name (for
		1176	the curious, IGS is the International Go Server and FCP is the Freenet
		1177	Client Protocol).
		1178
		1179	=item L<AnyEvent::AIO>
		1180
		1181	Truly asynchronous (as opposed to non-blocking) I/O, should be in the
		1182	toolbox of every event programmer. AnyEvent::AIO transparently fuses
		1183	L<IO::AIO> and AnyEvent together, giving AnyEvent access to event-based
		1184	file I/O, and much more.
		1185
		1186	=item L<AnyEvent::Filesys::Notify>
		1187
		1188	AnyEvent is good for non-blocking stuff, but it can't detect file or
		1189	path changes (e.g. "watch this directory for new files", "watch this
		1190	file for changes"). The L<AnyEvent::Filesys::Notify> module promises to
		1191	do just that in a portbale fashion, supporting inotify on GNU/Linux and
		1192	some weird, without doubt broken, stuff on OS X to monitor files. It can
		1193	fall back to blocking scans at regular intervals transparently on other
		1194	platforms, so it's about as portable as it gets.
		1195
		1196	(I haven't used it myself, but I haven't heard anybody complaining about
		1197	it yet).
		1198
		1199	=item L<AnyEvent::DBI>
		1200
		1201	Executes L<DBI> requests asynchronously in a proxy process for you,
		1202	notifying you in an event-based way when the operation is finished.
		1203
		1204	=item L<AnyEvent::HTTPD>
		1205
		1206	A simple embedded webserver.
		1207
		1208	=item L<AnyEvent::FastPing>
		1209
		1210	The fastest ping in the west.
		1211
		1212	=item L<Coro>
		1213
		1214	Has special support for AnyEvent via L<Coro::AnyEvent>, which allows you
		1215	to simply invert the flow control - don't call us, we will call you:
		1216
		1217	async {
		1218	Coro::AnyEvent::sleep 5; # creates a 5s timer and waits for it
		1219	print "5 seconds later!\n";
		1220
		1221	Coro::AnyEvent::readable *STDIN; # uses an I/O watcher
		1222	my $line = <STDIN>; # works for ttys
		1223
		1224	AnyEvent::HTTP::http_get "url", Coro::rouse_cb;
		1225	my ($body, $hdr) = Coro::rouse_wait;
		1226	};
		1227
		1228	=back
306		1229
307	=cut	1230	=cut
308		1231
309	package AnyEvent;	1232	package AnyEvent;
310		1233
311	no warnings;	1234	# basically a tuned-down version of common::sense
312	use strict;	1235	sub common_sense {
		1236	# from common:.sense 3.5
		1237	local $^W;
		1238	${^WARNING_BITS} ^= ${^WARNING_BITS} ^ "\x3c\x3f\x33\x00\x0f\xf0\x0f\xc0\xf0\xfc\x33\x00";
		1239	# use strict vars subs - NO UTF-8, as Util.pm doesn't like this atm. (uts46data.pl)
		1240	$^H |= 0x00000600;
		1241	}
313		1242
		1243	BEGIN { AnyEvent::common_sense }
		1244
314	use Carp;	1245	use Carp ();
315		1246
316	our $VERSION = '3.0';	1247	our $VERSION = '7.04';
317	our $MODEL;	1248	our $MODEL;
318
319	our $AUTOLOAD;
320	our @ISA;	1249	our @ISA;
321
322	our $verbose = $ENV{PERL_ANYEVENT_VERBOSE}*1;
323
324	our @REGISTRY;	1250	our @REGISTRY;
		1251	our $VERBOSE;
		1252	our %PROTOCOL; # (ipv4|ipv6) => (1|2), higher numbers are preferred
		1253	our $MAX_SIGNAL_LATENCY = $ENV{PERL_ANYEVENT_MAX_SIGNAL_LATENCY} || 10; # executes after the BEGIN block below (tainting!)
325		1254
		1255	BEGIN {
		1256	require "AnyEvent/constants.pl";
		1257
		1258	eval "sub TAINT (){" . (${^TAINT}*1) . "}";
		1259
		1260	delete @ENV{grep /^PERL_ANYEVENT_/, keys %ENV}
		1261	if ${^TAINT};
		1262
		1263	$ENV{"PERL_ANYEVENT_$_"} = $ENV{"AE_$_"}
		1264	for grep s/^AE_// && !exists $ENV{"PERL_ANYEVENT_$_"}, keys %ENV;
		1265
		1266	@ENV{grep /^PERL_ANYEVENT_/, keys %ENV} = ()
		1267	if ${^TAINT};
		1268
		1269	# $ENV{PERL_ANYEVENT_xxx} now valid
		1270
		1271	$VERBOSE = length $ENV{PERL_ANYEVENT_VERBOSE} ? $ENV{PERL_ANYEVENT_VERBOSE}*1 : 4;
		1272
		1273	my $idx;
		1274	$PROTOCOL{$_} = ++$idx
		1275	for reverse split /\s*,\s*/,
		1276	$ENV{PERL_ANYEVENT_PROTOCOLS} || "ipv4,ipv6";
		1277	}
		1278
		1279	our @post_detect;
		1280
		1281	sub post_detect(&) {
		1282	my ($cb) = @_;
		1283
		1284	push @post_detect, $cb;
		1285
		1286	defined wantarray
		1287	? bless \$cb, "AnyEvent::Util::postdetect"
		1288	: ()
		1289	}
		1290
		1291	sub AnyEvent::Util::postdetect::DESTROY {
		1292	@post_detect = grep $_ != ${$_[0]}, @post_detect;
		1293	}
		1294
		1295	our $POSTPONE_W;
		1296	our @POSTPONE;
		1297
		1298	sub _postpone_exec {
		1299	undef $POSTPONE_W;
		1300
		1301	&{ shift @POSTPONE }
		1302	while @POSTPONE;
		1303	}
		1304
		1305	sub postpone(&) {
		1306	push @POSTPONE, shift;
		1307
		1308	$POSTPONE_W ||= AE::timer (0, 0, \&_postpone_exec);
		1309
		1310	()
		1311	}
		1312
		1313	sub log($$;@) {
		1314	# only load the big bloated module when we actually are about to log something
		1315	if ($_[0] <= ($VERBOSE || 1)) { # also catches non-numeric levels(!) and fatal
		1316	local ($!, $@);
		1317	require AnyEvent::Log; # among other things, sets $VERBOSE to 9
		1318	# AnyEvent::Log overwrites this function
		1319	goto &log;
		1320	}
		1321
		1322	0 # not logged
		1323	}
		1324
		1325	sub _logger($;$) {
		1326	my ($level, $renabled) = @_;
		1327
		1328	$$renabled = $level <= $VERBOSE;
		1329
		1330	my $logger = [(caller)[0], $level, $renabled];
		1331
		1332	$AnyEvent::Log::LOGGER{$logger+0} = $logger;
		1333
		1334	# return unless defined wantarray;
		1335	#
		1336	# require AnyEvent::Util;
		1337	# my $guard = AnyEvent::Util::guard (sub {
		1338	# # "clean up"
		1339	# delete $LOGGER{$logger+0};
		1340	# });
		1341	#
		1342	# sub {
		1343	# return 0 unless $$renabled;
		1344	#
		1345	# $guard if 0; # keep guard alive, but don't cause runtime overhead
		1346	# require AnyEvent::Log unless $AnyEvent::Log::VERSION;
		1347	# package AnyEvent::Log;
		1348	# _log ($logger->[0], $level, @_) # logger->[0] has been converted at load time
		1349	# }
		1350	}
		1351
		1352	if (length $ENV{PERL_ANYEVENT_LOG}) {
		1353	require AnyEvent::Log; # AnyEvent::Log does the thing for us
		1354	}
		1355
326	my @models = (	1356	our @models = (
327	[Coro::EV:: => AnyEvent::Impl::CoroEV::],
328	[EV:: => AnyEvent::Impl::EV::],	1357	[EV:: => AnyEvent::Impl::EV::],
329	[Coro::Event:: => AnyEvent::Impl::CoroEvent::],	1358	[AnyEvent::Loop:: => AnyEvent::Impl::Perl::],
		1359	# everything below here will not (normally) be autoprobed
		1360	# as the pure perl backend should work everywhere
		1361	# and is usually faster
		1362	[Irssi:: => AnyEvent::Impl::Irssi::], # Irssi has a bogus "Event" package, so msut be near the top
330	[Event:: => AnyEvent::Impl::Event::],	1363	[Event:: => AnyEvent::Impl::Event::], # slow, stable
331	[Glib:: => AnyEvent::Impl::Glib::],	1364	[Glib:: => AnyEvent::Impl::Glib::], # becomes extremely slow with many watchers
		1365	# everything below here should not be autoloaded
		1366	[Event::Lib:: => AnyEvent::Impl::EventLib::], # too buggy
		1367	[Tk:: => AnyEvent::Impl::Tk::], # crashes with many handles
		1368	[Qt:: => AnyEvent::Impl::Qt::], # requires special main program
		1369	[POE::Kernel:: => AnyEvent::Impl::POE::], # lasciate ogni speranza
332	[Tk:: => AnyEvent::Impl::Tk::],	1370	[Wx:: => AnyEvent::Impl::POE::],
333	[AnyEvent::Impl::Perl:: => AnyEvent::Impl::Perl::],	1371	[Prima:: => AnyEvent::Impl::POE::],
		1372	[IO::Async::Loop:: => AnyEvent::Impl::IOAsync::], # a bitch to autodetect
		1373	[Cocoa::EventLoop:: => AnyEvent::Impl::Cocoa::],
		1374	[FLTK:: => AnyEvent::Impl::FLTK::],
334	);	1375	);
335		1376
336	our %method = map +($_ => 1), qw(io timer condvar broadcast wait signal one_event DESTROY);	1377	our @isa_hook;
		1378
		1379	sub _isa_set {
		1380	my @pkg = ("AnyEvent", (map $_->[0], grep defined, @isa_hook), $MODEL);
		1381
		1382	@{"$pkg[$_-1]::ISA"} = $pkg[$_]
		1383	for 1 .. $#pkg;
		1384
		1385	grep $_ && $_->[1], @isa_hook
		1386	and AE::_reset ();
		1387	}
		1388
		1389	# used for hooking AnyEvent::Strict and AnyEvent::Debug::Wrap into the class hierarchy
		1390	sub _isa_hook($$;$) {
		1391	my ($i, $pkg, $reset_ae) = @_;
		1392
		1393	$isa_hook[$i] = $pkg ? [$pkg, $reset_ae] : undef;
		1394
		1395	_isa_set;
		1396	}
		1397
		1398	# all autoloaded methods reserve the complete glob, not just the method slot.
		1399	# due to bugs in perls method cache implementation.
		1400	our @methods = qw(io timer time now now_update signal child idle condvar);
337		1401
338	sub detect() {	1402	sub detect() {
		1403	return $MODEL if $MODEL; # some programs keep references to detect
		1404
		1405	# IO::Async::Loop::AnyEvent is extremely evil, refuse to work with it
		1406	# the author knows about the problems and what it does to AnyEvent as a whole
		1407	# (and the ability of others to use AnyEvent), but simply wants to abuse AnyEvent
		1408	# anyway.
		1409	AnyEvent::log fatal => "IO::Async::Loop::AnyEvent detected - that module is broken by\n"
		1410	. "design, abuses internals and breaks AnyEvent - will not continue."
		1411	if exists $INC{"IO/Async/Loop/AnyEvent.pm"};
		1412
		1413	local $!; # for good measure
		1414	local $SIG{__DIE__}; # we use eval
		1415
		1416	# free some memory
		1417	*detect = sub () { $MODEL };
		1418	# undef &func doesn't correctly update the method cache. grmbl.
		1419	# so we delete the whole glob. grmbl.
		1420	# otoh, perl doesn't let me undef an active usb, but it lets me free
		1421	# a glob with an active sub. hrm. i hope it works, but perl is
		1422	# usually buggy in this department. sigh.
		1423	delete @{"AnyEvent::"}{@methods};
		1424	undef @methods;
		1425
		1426	if ($ENV{PERL_ANYEVENT_MODEL} =~ /^([a-zA-Z0-9:]+)$/) {
		1427	my $model = $1;
		1428	$model = "AnyEvent::Impl::$model" unless $model =~ s/::$//;
		1429	if (eval "require $model") {
		1430	AnyEvent::log 7 => "Loaded model '$model' (forced by \$ENV{PERL_ANYEVENT_MODEL}), using it.";
		1431	$MODEL = $model;
		1432	} else {
		1433	AnyEvent::log 4 => "Unable to load model '$model' (from \$ENV{PERL_ANYEVENT_MODEL}):\n$@";
		1434	}
		1435	}
		1436
		1437	# check for already loaded models
339	unless ($MODEL) {	1438	unless ($MODEL) {
340	no strict 'refs';
341
342	# check for already loaded models
343	for (@REGISTRY, @models) {	1439	for (@REGISTRY, @models) {
344	my ($package, $model) = @$_;	1440	my ($package, $model) = @$_;
345	if (${"$package\::VERSION"} > 0) {	1441	if (${"$package\::VERSION"} > 0) {
346	if (eval "require $model") {	1442	if (eval "require $model") {
		1443	AnyEvent::log 7 => "Autodetected model '$model', using it.";
347	$MODEL = $model;	1444	$MODEL = $model;
348	warn "AnyEvent: found model '$model', using it.\n" if $verbose > 1;	1445	last;
		1446	} else {
		1447	AnyEvent::log 8 => "Detected event loop $package, but cannot load '$model', skipping: $@";
		1448	}
		1449	}
		1450	}
		1451
		1452	unless ($MODEL) {
		1453	# try to autoload a model
		1454	for (@REGISTRY, @models) {
		1455	my ($package, $model) = @$_;
		1456	if (
		1457	eval "require $package"
		1458	and ${"$package\::VERSION"} > 0
		1459	and eval "require $model"
		1460	) {
		1461	AnyEvent::log 7 => "Autoloaded model '$model', using it.";
		1462	$MODEL = $model;
349	last;	1463	last;
350	}	1464	}
351	}	1465	}
		1466
		1467	$MODEL
		1468	or AnyEvent::log fatal => "Backend autodetection failed - did you properly install AnyEvent?";
352	}	1469	}
		1470	}
353		1471
354	unless ($MODEL) {	1472	# free memory only needed for probing
355	# try to load a model	1473	undef @models;
		1474	undef @REGISTRY;
356		1475
357	for (@REGISTRY, @models) {	1476	push @{"$MODEL\::ISA"}, "AnyEvent::Base";
358	my ($package, $model) = @$_;	1477
359	if (eval "require $package"	1478	# now nuke some methods that are overridden by the backend.
360	and ${"$package\::VERSION"} > 0	1479	# SUPER usage is not allowed in these.
361	and eval "require $model") {	1480	for (qw(time signal child idle)) {
362	$MODEL = $model;	1481	undef &{"AnyEvent::Base::$_"}
363	warn "AnyEvent: autoprobed and loaded model '$model', using it.\n" if $verbose > 1;	1482	if defined &{"$MODEL\::$_"};
		1483	}
		1484
		1485	_isa_set;
		1486
		1487	# we're officially open!
		1488
		1489	if ($ENV{PERL_ANYEVENT_STRICT}) {
		1490	require AnyEvent::Strict;
		1491	}
		1492
		1493	if ($ENV{PERL_ANYEVENT_DEBUG_WRAP}) {
		1494	require AnyEvent::Debug;
		1495	AnyEvent::Debug::wrap ($ENV{PERL_ANYEVENT_DEBUG_WRAP});
		1496	}
		1497
		1498	if (length $ENV{PERL_ANYEVENT_DEBUG_SHELL}) {
		1499	require AnyEvent::Socket;
		1500	require AnyEvent::Debug;
		1501
		1502	my $shell = $ENV{PERL_ANYEVENT_DEBUG_SHELL};
		1503	$shell =~ s/\$\$/$$/g;
		1504
		1505	my ($host, $service) = AnyEvent::Socket::parse_hostport ($shell);
		1506	$AnyEvent::Debug::SHELL = AnyEvent::Debug::shell ($host, $service);
		1507	}
		1508
		1509	# now the anyevent environment is set up as the user told us to, so
		1510	# call the actual user code - post detects
		1511
		1512	(shift @post_detect)->() while @post_detect;
		1513	undef @post_detect;
		1514
		1515	*post_detect = sub(&) {
		1516	shift->();
		1517
		1518	undef
		1519	};
		1520
		1521	$MODEL
		1522	}
		1523
		1524	for my $name (@methods) {
		1525	*$name = sub {
		1526	detect;
		1527	# we use goto because
		1528	# a) it makes the thunk more transparent
		1529	# b) it allows us to delete the thunk later
		1530	goto &{ UNIVERSAL::can AnyEvent => "SUPER::$name" }
		1531	};
		1532	}
		1533
		1534	# utility function to dup a filehandle. this is used by many backends
		1535	# to support binding more than one watcher per filehandle (they usually
		1536	# allow only one watcher per fd, so we dup it to get a different one).
		1537	sub _dupfh($$;$$) {
		1538	my ($poll, $fh, $r, $w) = @_;
		1539
		1540	# cygwin requires the fh mode to be matching, unix doesn't
		1541	my ($rw, $mode) = $poll eq "r" ? ($r, "<&") : ($w, ">&");
		1542
		1543	open my $fh2, $mode, $fh
		1544	or die "AnyEvent->io: cannot dup() filehandle in mode '$poll': $!,";
		1545
		1546	# we assume CLOEXEC is already set by perl in all important cases
		1547
		1548	($fh2, $rw)
		1549	}
		1550
		1551	=head1 SIMPLIFIED AE API
		1552
		1553	Starting with version 5.0, AnyEvent officially supports a second, much
		1554	simpler, API that is designed to reduce the calling, typing and memory
		1555	overhead by using function call syntax and a fixed number of parameters.
		1556
		1557	See the L<AE> manpage for details.
		1558
		1559	=cut
		1560
		1561	package AE;
		1562
		1563	our $VERSION = $AnyEvent::VERSION;
		1564
		1565	sub _reset() {
		1566	eval q{
		1567	# fall back to the main API by default - backends and AnyEvent::Base
		1568	# implementations can overwrite these.
		1569
		1570	sub io($$$) {
		1571	AnyEvent->io (fh => $_[0], poll => $_[1] ? "w" : "r", cb => $_[2])
		1572	}
		1573
		1574	sub timer($$$) {
		1575	AnyEvent->timer (after => $_[0], interval => $_[1], cb => $_[2])
		1576	}
		1577
		1578	sub signal($$) {
		1579	AnyEvent->signal (signal => $_[0], cb => $_[1])
		1580	}
		1581
		1582	sub child($$) {
		1583	AnyEvent->child (pid => $_[0], cb => $_[1])
		1584	}
		1585
		1586	sub idle($) {
		1587	AnyEvent->idle (cb => $_[0]);
		1588	}
		1589
		1590	sub cv(;&) {
		1591	AnyEvent->condvar (@_ ? (cb => $_[0]) : ())
		1592	}
		1593
		1594	sub now() {
		1595	AnyEvent->now
		1596	}
		1597
		1598	sub now_update() {
		1599	AnyEvent->now_update
		1600	}
		1601
		1602	sub time() {
		1603	AnyEvent->time
		1604	}
		1605
		1606	*postpone = \&AnyEvent::postpone;
		1607	*log = \&AnyEvent::log;
		1608	};
		1609	die if $@;
		1610	}
		1611
		1612	BEGIN { _reset }
		1613
		1614	package AnyEvent::Base;
		1615
		1616	# default implementations for many methods
		1617
		1618	sub time {
		1619	eval q{ # poor man's autoloading {}
		1620	# probe for availability of Time::HiRes
		1621	if (eval "use Time::HiRes (); Time::HiRes::time (); 1") {
		1622	*time = sub { Time::HiRes::time () };
		1623	*AE::time = \& Time::HiRes::time ;
		1624	*now = \&time;
		1625	AnyEvent::log 8 => "using Time::HiRes for sub-second timing accuracy.";
		1626	# if (eval "use POSIX (); (POSIX::times())...
		1627	} else {
		1628	*time = sub { CORE::time };
		1629	*AE::time = sub (){ CORE::time };
		1630	*now = \&time;
		1631	AnyEvent::log 3 => "Using built-in time(), no sub-second resolution!";
		1632	}
		1633	};
		1634	die if $@;
		1635
		1636	&time
		1637	}
		1638
		1639	*now = \&time;
		1640	sub now_update { }
		1641
		1642	sub _poll {
		1643	Carp::croak "$AnyEvent::MODEL does not support blocking waits. Caught";
		1644	}
		1645
		1646	# default implementation for ->condvar
		1647	# in fact, the default should not be overwritten
		1648
		1649	sub condvar {
		1650	eval q{ # poor man's autoloading {}
		1651	*condvar = sub {
		1652	bless { @_ == 3 ? (_ae_cb => $_[2]) : () }, "AnyEvent::CondVar"
		1653	};
		1654
		1655	*AE::cv = sub (;&) {
		1656	bless { @_ ? (_ae_cb => shift) : () }, "AnyEvent::CondVar"
		1657	};
		1658	};
		1659	die if $@;
		1660
		1661	&condvar
		1662	}
		1663
		1664	# default implementation for ->signal
		1665
		1666	our $HAVE_ASYNC_INTERRUPT;
		1667
		1668	sub _have_async_interrupt() {
		1669	$HAVE_ASYNC_INTERRUPT = 1*(!$ENV{PERL_ANYEVENT_AVOID_ASYNC_INTERRUPT}
		1670	&& eval "use Async::Interrupt 1.02 (); 1")
		1671	unless defined $HAVE_ASYNC_INTERRUPT;
		1672
		1673	$HAVE_ASYNC_INTERRUPT
		1674	}
		1675
		1676	our ($SIGPIPE_R, $SIGPIPE_W, %SIG_CB, %SIG_EV, $SIG_IO);
		1677	our (%SIG_ASY, %SIG_ASY_W);
		1678	our ($SIG_COUNT, $SIG_TW);
		1679
		1680	# install a dummy wakeup watcher to reduce signal catching latency
		1681	# used by Impls
		1682	sub _sig_add() {
		1683	unless ($SIG_COUNT++) {
		1684	# try to align timer on a full-second boundary, if possible
		1685	my $NOW = AE::now;
		1686
		1687	$SIG_TW = AE::timer
		1688	$MAX_SIGNAL_LATENCY - ($NOW - int $NOW),
		1689	$MAX_SIGNAL_LATENCY,
		1690	sub { } # just for the PERL_ASYNC_CHECK
		1691	;
		1692	}
		1693	}
		1694
		1695	sub _sig_del {
		1696	undef $SIG_TW
		1697	unless --$SIG_COUNT;
		1698	}
		1699
		1700	our $_sig_name_init; $_sig_name_init = sub {
		1701	eval q{ # poor man's autoloading {}
		1702	undef $_sig_name_init;
		1703
		1704	if (_have_async_interrupt) {
		1705	*sig2num = \&Async::Interrupt::sig2num;
		1706	*sig2name = \&Async::Interrupt::sig2name;
		1707	} else {
		1708	require Config;
		1709
		1710	my %signame2num;
		1711	@signame2num{ split ' ', $Config::Config{sig_name} }
		1712	= split ' ', $Config::Config{sig_num};
		1713
		1714	my @signum2name;
		1715	@signum2name[values %signame2num] = keys %signame2num;
		1716
		1717	*sig2num = sub($) {
		1718	$_[0] > 0 ? shift : $signame2num{+shift}
		1719	};
		1720	*sig2name = sub ($) {
		1721	$_[0] > 0 ? $signum2name[+shift] : shift
		1722	};
		1723	}
		1724	};
		1725	die if $@;
		1726	};
		1727
		1728	sub sig2num ($) { &$_sig_name_init; &sig2num }
		1729	sub sig2name($) { &$_sig_name_init; &sig2name }
		1730
		1731	sub signal {
		1732	eval q{ # poor man's autoloading {}
		1733	# probe for availability of Async::Interrupt
		1734	if (_have_async_interrupt) {
		1735	AnyEvent::log 8 => "Using Async::Interrupt for race-free signal handling.";
		1736
		1737	$SIGPIPE_R = new Async::Interrupt::EventPipe;
		1738	$SIG_IO = AE::io $SIGPIPE_R->fileno, 0, \&_signal_exec;
		1739
		1740	} else {
		1741	AnyEvent::log 8 => "Using emulated perl signal handling with latency timer.";
		1742
		1743	if (AnyEvent::WIN32) {
		1744	require AnyEvent::Util;
		1745
		1746	($SIGPIPE_R, $SIGPIPE_W) = AnyEvent::Util::portable_pipe ();
		1747	AnyEvent::Util::fh_nonblocking ($SIGPIPE_R, 1) if $SIGPIPE_R;
		1748	AnyEvent::Util::fh_nonblocking ($SIGPIPE_W, 1) if $SIGPIPE_W; # just in case
		1749	} else {
		1750	pipe $SIGPIPE_R, $SIGPIPE_W;
		1751	fcntl $SIGPIPE_R, AnyEvent::F_SETFL, AnyEvent::O_NONBLOCK if $SIGPIPE_R;
		1752	fcntl $SIGPIPE_W, AnyEvent::F_SETFL, AnyEvent::O_NONBLOCK if $SIGPIPE_W; # just in case
		1753
		1754	# not strictly required, as $^F is normally 2, but let's make sure...
		1755	fcntl $SIGPIPE_R, AnyEvent::F_SETFD, AnyEvent::FD_CLOEXEC;
		1756	fcntl $SIGPIPE_W, AnyEvent::F_SETFD, AnyEvent::FD_CLOEXEC;
		1757	}
		1758
		1759	$SIGPIPE_R
		1760	or Carp::croak "AnyEvent: unable to create a signal reporting pipe: $!\n";
		1761
		1762	$SIG_IO = AE::io $SIGPIPE_R, 0, \&_signal_exec;
		1763	}
		1764
		1765	*signal = $HAVE_ASYNC_INTERRUPT
		1766	? sub {
		1767	my (undef, %arg) = @_;
		1768
		1769	# async::interrupt
		1770	my $signal = sig2num $arg{signal};
		1771	$SIG_CB{$signal}{$arg{cb}} = $arg{cb};
		1772
		1773	$SIG_ASY{$signal} ||= new Async::Interrupt
		1774	cb => sub { undef $SIG_EV{$signal} },
		1775	signal => $signal,
		1776	pipe => [$SIGPIPE_R->filenos],
		1777	pipe_autodrain => 0,
364	last;	1778	;
		1779
		1780	bless [$signal, $arg{cb}], "AnyEvent::Base::signal"
		1781	}
		1782	: sub {
		1783	my (undef, %arg) = @_;
		1784
		1785	# pure perl
		1786	my $signal = sig2name $arg{signal};
		1787	$SIG_CB{$signal}{$arg{cb}} = $arg{cb};
		1788
		1789	$SIG{$signal} ||= sub {
		1790	local $!;
		1791	syswrite $SIGPIPE_W, "\x00", 1 unless %SIG_EV;
		1792	undef $SIG_EV{$signal};
		1793	};
		1794
		1795	# can't do signal processing without introducing races in pure perl,
		1796	# so limit the signal latency.
		1797	_sig_add;
		1798
		1799	bless [$signal, $arg{cb}], "AnyEvent::Base::signal"
		1800	}
		1801	;
		1802
		1803	*AnyEvent::Base::signal::DESTROY = sub {
		1804	my ($signal, $cb) = @{$_[0]};
		1805
		1806	_sig_del;
		1807
		1808	delete $SIG_CB{$signal}{$cb};
		1809
		1810	$HAVE_ASYNC_INTERRUPT
		1811	? delete $SIG_ASY{$signal}
		1812	: # delete doesn't work with older perls - they then
		1813	# print weird messages, or just unconditionally exit
		1814	# instead of getting the default action.
		1815	undef $SIG{$signal}
		1816	unless keys %{ $SIG_CB{$signal} };
		1817	};
		1818
		1819	*_signal_exec = sub {
		1820	$HAVE_ASYNC_INTERRUPT
		1821	? $SIGPIPE_R->drain
		1822	: sysread $SIGPIPE_R, (my $dummy), 9;
		1823
		1824	while (%SIG_EV) {
		1825	for (keys %SIG_EV) {
		1826	delete $SIG_EV{$_};
		1827	&$_ for values %{ $SIG_CB{$_} || {} };
365	}	1828	}
366	}	1829	}
367
368	$MODEL
369	or die "No event module selected for AnyEvent and autodetect failed. Install any one of these modules: EV (or Coro+EV), Event (or Coro+Event), Glib or Tk.";
370	}	1830	};
371
372	unshift @ISA, $MODEL;
373	push @{"$MODEL\::ISA"}, "AnyEvent::Base";
374	}
375
376	$MODEL
377	}
378
379	sub AUTOLOAD {
380	(my $func = $AUTOLOAD) =~ s/.*://;
381
382	$method{$func}
383	or croak "$func: not a valid method for AnyEvent objects";
384
385	detect unless $MODEL;
386
387	my $class = shift;
388	$class->$func (@_);
389	}
390
391	package AnyEvent::Base;
392
393	# default implementation for ->condvar, ->wait, ->broadcast
394
395	sub condvar {
396	bless \my $flag, "AnyEvent::Base::CondVar"
397	}
398
399	sub AnyEvent::Base::CondVar::broadcast {
400	${$_[0]}++;
401	}
402
403	sub AnyEvent::Base::CondVar::wait {
404	AnyEvent->one_event while !${$_[0]};
405	}
406
407	# default implementation for ->signal
408
409	our %SIG_CB;
410
411	sub signal {
412	my (undef, %arg) = @_;
413
414	my $signal = uc $arg{signal}
415	or Carp::croak "required option 'signal' is missing";
416
417	$SIG_CB{$signal}{$arg{cb}} = $arg{cb};
418	$SIG{$signal} ||= sub {
419	$_->() for values %{ $SIG_CB{$signal} || {} };
420	};	1831	};
		1832	die if $@;
421		1833
422	bless [$signal, $arg{cb}], "AnyEvent::Base::Signal"	1834	&signal
423	}
424
425	sub AnyEvent::Base::Signal::DESTROY {
426	my ($signal, $cb) = @{$_[0]};
427
428	delete $SIG_CB{$signal}{$cb};
429
430	$SIG{$signal} = 'DEFAULT' unless keys %{ $SIG_CB{$signal} };
431	}	1835	}
432		1836
433	# default implementation for ->child	1837	# default implementation for ->child
434		1838
435	our %PID_CB;	1839	our %PID_CB;
436	our $CHLD_W;	1840	our $CHLD_W;
437	our $CHLD_DELAY_W;	1841	our $CHLD_DELAY_W;
438	our $PID_IDLE;
439	our $WNOHANG;
440		1842
441	sub _child_wait {	1843	# used by many Impl's
442	while (0 < (my $pid = waitpid -1, $WNOHANG)) {	1844	sub _emit_childstatus($$) {
		1845	my (undef, $rpid, $rstatus) = @_;
		1846
		1847	$_->($rpid, $rstatus)
443	$_->($pid, $?) for (values %{ $PID_CB{$pid} || {} }),	1848	for values %{ $PID_CB{$rpid} || {} },
444	(values %{ $PID_CB{0} || {} });	1849	values %{ $PID_CB{0} || {} };
		1850	}
		1851
		1852	sub child {
		1853	eval q{ # poor man's autoloading {}
		1854	*_sigchld = sub {
		1855	my $pid;
		1856
		1857	AnyEvent->_emit_childstatus ($pid, $?)
		1858	while ($pid = waitpid -1, WNOHANG) > 0;
		1859	};
		1860
		1861	*child = sub {
		1862	my (undef, %arg) = @_;
		1863
		1864	my $pid = $arg{pid};
		1865	my $cb = $arg{cb};
		1866
		1867	$PID_CB{$pid}{$cb+0} = $cb;
		1868
		1869	unless ($CHLD_W) {
		1870	$CHLD_W = AE::signal CHLD => \&_sigchld;
		1871	# child could be a zombie already, so make at least one round
		1872	&_sigchld;
		1873	}
		1874
		1875	bless [$pid, $cb+0], "AnyEvent::Base::child"
		1876	};
		1877
		1878	*AnyEvent::Base::child::DESTROY = sub {
		1879	my ($pid, $icb) = @{$_[0]};
		1880
		1881	delete $PID_CB{$pid}{$icb};
		1882	delete $PID_CB{$pid} unless keys %{ $PID_CB{$pid} };
		1883
		1884	undef $CHLD_W unless keys %PID_CB;
		1885	};
		1886	};
		1887	die if $@;
		1888
		1889	&child
		1890	}
		1891
		1892	# idle emulation is done by simply using a timer, regardless
		1893	# of whether the process is idle or not, and not letting
		1894	# the callback use more than 50% of the time.
		1895	sub idle {
		1896	eval q{ # poor man's autoloading {}
		1897	*idle = sub {
		1898	my (undef, %arg) = @_;
		1899
		1900	my ($cb, $w, $rcb) = $arg{cb};
		1901
		1902	$rcb = sub {
		1903	if ($cb) {
		1904	$w = AE::time;
		1905	&$cb;
		1906	$w = AE::time - $w;
		1907
		1908	# never use more then 50% of the time for the idle watcher,
		1909	# within some limits
		1910	$w = 0.0001 if $w < 0.0001;
		1911	$w = 5 if $w > 5;
		1912
		1913	$w = AE::timer $w, 0, $rcb;
		1914	} else {
		1915	# clean up...
		1916	undef $w;
		1917	undef $rcb;
		1918	}
		1919	};
		1920
		1921	$w = AE::timer 0.05, 0, $rcb;
		1922
		1923	bless \\$cb, "AnyEvent::Base::idle"
		1924	};
		1925
		1926	*AnyEvent::Base::idle::DESTROY = sub {
		1927	undef $${$_[0]};
		1928	};
		1929	};
		1930	die if $@;
		1931
		1932	&idle
		1933	}
		1934
		1935	package AnyEvent::CondVar;
		1936
		1937	our @ISA = AnyEvent::CondVar::Base::;
		1938
		1939	# only to be used for subclassing
		1940	sub new {
		1941	my $class = shift;
		1942	bless AnyEvent->condvar (@_), $class
		1943	}
		1944
		1945	package AnyEvent::CondVar::Base;
		1946
		1947	#use overload
		1948	# '&{}' => sub { my $self = shift; sub { $self->send (@_) } },
		1949	# fallback => 1;
		1950
		1951	# save 300+ kilobytes by dirtily hardcoding overloading
		1952	${"AnyEvent::CondVar::Base::OVERLOAD"}{dummy}++; # Register with magic by touching.
		1953	*{'AnyEvent::CondVar::Base::()'} = sub { }; # "Make it findable via fetchmethod."
		1954	*{'AnyEvent::CondVar::Base::(&{}'} = sub { my $self = shift; sub { $self->send (@_) } }; # &{}
		1955	${'AnyEvent::CondVar::Base::()'} = 1; # fallback
		1956
		1957	our $WAITING;
		1958
		1959	sub _send {
		1960	# nop
		1961	}
		1962
		1963	sub _wait {
		1964	AnyEvent->_poll until $_[0]{_ae_sent};
		1965	}
		1966
		1967	sub send {
		1968	my $cv = shift;
		1969	$cv->{_ae_sent} = [@_];
		1970	(delete $cv->{_ae_cb})->($cv) if $cv->{_ae_cb};
		1971	$cv->_send;
		1972	}
		1973
		1974	sub croak {
		1975	$_[0]{_ae_croak} = $_[1];
		1976	$_[0]->send;
		1977	}
		1978
		1979	sub ready {
		1980	$_[0]{_ae_sent}
		1981	}
		1982
		1983	sub recv {
		1984	unless ($_[0]{_ae_sent}) {
		1985	$WAITING
		1986	and Carp::croak "AnyEvent::CondVar: recursive blocking wait attempted";
		1987
		1988	local $WAITING = 1;
		1989	$_[0]->_wait;
445	}	1990	}
446		1991
447	undef $PID_IDLE;	1992	$_[0]{_ae_croak}
448	}	1993	and Carp::croak $_[0]{_ae_croak};
449		1994
450	sub _sigchld {	1995	wantarray
451	# make sure we deliver these changes "synchronous" with the event loop.	1996	? @{ $_[0]{_ae_sent} }
452	$CHLD_DELAY_W ||= AnyEvent->timer (after => 0, cb => sub {	1997	: $_[0]{_ae_sent}[0]
453	undef $CHLD_DELAY_W;
454	&_child_wait;
455	});
456	}	1998	}
457		1999
458	sub child {	2000	sub cb {
459	my (undef, %arg) = @_;	2001	my $cv = shift;
460		2002
461	defined (my $pid = $arg{pid} + 0)	2003	@_
462	or Carp::croak "required option 'pid' is missing";	2004	and $cv->{_ae_cb} = shift
		2005	and $cv->{_ae_sent}
		2006	and (delete $cv->{_ae_cb})->($cv);
463		2007
464	$PID_CB{$pid}{$arg{cb}} = $arg{cb};	2008	$cv->{_ae_cb}
465
466	unless ($WNOHANG) {
467	$WNOHANG = eval { require POSIX; &POSIX::WNOHANG } || 1;
468	}
469
470	unless ($CHLD_W) {
471	$CHLD_W = AnyEvent->signal (signal => 'CHLD', cb => \&_sigchld);
472	# child could be a zombie already, so make at least one round
473	&_sigchld;
474	}
475
476	bless [$pid, $arg{cb}], "AnyEvent::Base::Child"
477	}	2009	}
478		2010
479	sub AnyEvent::Base::Child::DESTROY {	2011	sub begin {
480	my ($pid, $cb) = @{$_[0]};	2012	++$_[0]{_ae_counter};
481		2013	$_[0]{_ae_end_cb} = $_[1] if @_ > 1;
482	delete $PID_CB{$pid}{$cb};
483	delete $PID_CB{$pid} unless keys %{ $PID_CB{$pid} };
484
485	undef $CHLD_W unless keys %PID_CB;
486	}	2014	}
		2015
		2016	sub end {
		2017	return if --$_[0]{_ae_counter};
		2018	&{ $_[0]{_ae_end_cb} || sub { $_[0]->send } };
		2019	}
		2020
		2021	# undocumented/compatibility with pre-3.4
		2022	*broadcast = \&send;
		2023	*wait = \&recv;
		2024
		2025	=head1 ERROR AND EXCEPTION HANDLING
		2026
		2027	In general, AnyEvent does not do any error handling - it relies on the
		2028	caller to do that if required. The L<AnyEvent::Strict> module (see also
		2029	the C<PERL_ANYEVENT_STRICT> environment variable, below) provides strict
		2030	checking of all AnyEvent methods, however, which is highly useful during
		2031	development.
		2032
		2033	As for exception handling (i.e. runtime errors and exceptions thrown while
		2034	executing a callback), this is not only highly event-loop specific, but
		2035	also not in any way wrapped by this module, as this is the job of the main
		2036	program.
		2037
		2038	The pure perl event loop simply re-throws the exception (usually
		2039	within C<< condvar->recv >>), the L<Event> and L<EV> modules call C<<
		2040	$Event/EV::DIED->() >>, L<Glib> uses C<< install_exception_handler >> and
		2041	so on.
		2042
		2043	=head1 ENVIRONMENT VARIABLES
		2044
		2045	AnyEvent supports a number of environment variables that tune the
		2046	runtime behaviour. They are usually evaluated when AnyEvent is
		2047	loaded, initialised, or a submodule that uses them is loaded. Many of
		2048	them also cause AnyEvent to load additional modules - for example,
		2049	C<PERL_ANYEVENT_DEBUG_WRAP> causes the L<AnyEvent::Debug> module to be
		2050	loaded.
		2051
		2052	All the environment variables documented here start with
		2053	C<PERL_ANYEVENT_>, which is what AnyEvent considers its own
		2054	namespace. Other modules are encouraged (but by no means required) to use
		2055	C<PERL_ANYEVENT_SUBMODULE> if they have registered the AnyEvent::Submodule
		2056	namespace on CPAN, for any submodule. For example, L<AnyEvent::HTTP> could
		2057	be expected to use C<PERL_ANYEVENT_HTTP_PROXY> (it should not access env
		2058	variables starting with C<AE_>, see below).
		2059
		2060	All variables can also be set via the C<AE_> prefix, that is, instead
		2061	of setting C<PERL_ANYEVENT_VERBOSE> you can also set C<AE_VERBOSE>. In
		2062	case there is a clash btween anyevent and another program that uses
		2063	C<AE_something> you can set the corresponding C<PERL_ANYEVENT_something>
		2064	variable to the empty string, as those variables take precedence.
		2065
		2066	When AnyEvent is first loaded, it copies all C<AE_xxx> env variables
		2067	to their C<PERL_ANYEVENT_xxx> counterpart unless that variable already
		2068	exists. If taint mode is on, then AnyEvent will remove I<all> environment
		2069	variables starting with C<PERL_ANYEVENT_> from C<%ENV> (or replace them
		2070	with C<undef> or the empty string, if the corresaponding C<AE_> variable
		2071	is set).
		2072
		2073	The exact algorithm is currently:
		2074
		2075	1. if taint mode enabled, delete all PERL_ANYEVENT_xyz variables from %ENV
		2076	2. copy over AE_xyz to PERL_ANYEVENT_xyz unless the latter alraedy exists
		2077	3. if taint mode enabled, set all PERL_ANYEVENT_xyz variables to undef.
		2078
		2079	This ensures that child processes will not see the C<AE_> variables.
		2080
		2081	The following environment variables are currently known to AnyEvent:
		2082
		2083	=over 4
		2084
		2085	=item C<PERL_ANYEVENT_VERBOSE>
		2086
		2087	By default, AnyEvent will log messages with loglevel C<4> (C<error>) or
		2088	higher (see L<AnyEvent::Log>). You can set this environment variable to a
		2089	numerical loglevel to make AnyEvent more (or less) talkative.
		2090
		2091	If you want to do more than just set the global logging level
		2092	you should have a look at C<PERL_ANYEVENT_LOG>, which allows much more
		2093	complex specifications.
		2094
		2095	When set to C<0> (C<off>), then no messages whatsoever will be logged with
		2096	everything else at defaults.
		2097
		2098	When set to C<5> or higher (C<warn>), AnyEvent warns about unexpected
		2099	conditions, such as not being able to load the event model specified by
		2100	C<PERL_ANYEVENT_MODEL>, or a guard callback throwing an exception - this
		2101	is the minimum recommended level for use during development.
		2102
		2103	When set to C<7> or higher (info), AnyEvent reports which event model it
		2104	chooses.
		2105
		2106	When set to C<8> or higher (debug), then AnyEvent will report extra
		2107	information on which optional modules it loads and how it implements
		2108	certain features.
		2109
		2110	=item C<PERL_ANYEVENT_LOG>
		2111
		2112	Accepts rather complex logging specifications. For example, you could log
		2113	all C<debug> messages of some module to stderr, warnings and above to
		2114	stderr, and errors and above to syslog, with:
		2115
		2116	PERL_ANYEVENT_LOG=Some::Module=debug,+log:filter=warn,+%syslog:%syslog=error,syslog
		2117
		2118	For the rather extensive details, see L<AnyEvent::Log>.
		2119
		2120	This variable is evaluated when AnyEvent (or L<AnyEvent::Log>) is loaded,
		2121	so will take effect even before AnyEvent has initialised itself.
		2122
		2123	Note that specifying this environment variable causes the L<AnyEvent::Log>
		2124	module to be loaded, while C<PERL_ANYEVENT_VERBOSE> does not, so only
		2125	using the latter saves a few hundred kB of memory unless a module
		2126	explicitly needs the extra features of AnyEvent::Log.
		2127
		2128	=item C<PERL_ANYEVENT_STRICT>
		2129
		2130	AnyEvent does not do much argument checking by default, as thorough
		2131	argument checking is very costly. Setting this variable to a true value
		2132	will cause AnyEvent to load C<AnyEvent::Strict> and then to thoroughly
		2133	check the arguments passed to most method calls. If it finds any problems,
		2134	it will croak.
		2135
		2136	In other words, enables "strict" mode.
		2137
		2138	Unlike C<use strict> (or its modern cousin, C<< use L<common::sense>
		2139	>>, it is definitely recommended to keep it off in production. Keeping
		2140	C<PERL_ANYEVENT_STRICT=1> in your environment while developing programs
		2141	can be very useful, however.
		2142
		2143	=item C<PERL_ANYEVENT_DEBUG_SHELL>
		2144
		2145	If this env variable is nonempty, then its contents will be interpreted by
		2146	C<AnyEvent::Socket::parse_hostport> and C<AnyEvent::Debug::shell> (after
		2147	replacing every occurance of C<$$> by the process pid). The shell object
		2148	is saved in C<$AnyEvent::Debug::SHELL>.
		2149
		2150	This happens when the first watcher is created.
		2151
		2152	For example, to bind a debug shell on a unix domain socket in
		2153	F<< /tmp/debug<pid>.sock >>, you could use this:
		2154
		2155	PERL_ANYEVENT_DEBUG_SHELL=/tmp/debug\$\$.sock perlprog
		2156	# connect with e.g.: socat readline /tmp/debug123.sock
		2157
		2158	Or to bind to tcp port 4545 on localhost:
		2159
		2160	PERL_ANYEVENT_DEBUG_SHELL=127.0.0.1:4545 perlprog
		2161	# connect with e.g.: telnet localhost 4545
		2162
		2163	Note that creating sockets in F</tmp> or on localhost is very unsafe on
		2164	multiuser systems.
		2165
		2166	=item C<PERL_ANYEVENT_DEBUG_WRAP>
		2167
		2168	Can be set to C<0>, C<1> or C<2> and enables wrapping of all watchers for
		2169	debugging purposes. See C<AnyEvent::Debug::wrap> for details.
		2170
		2171	=item C<PERL_ANYEVENT_MODEL>
		2172
		2173	This can be used to specify the event model to be used by AnyEvent, before
		2174	auto detection and -probing kicks in.
		2175
		2176	It normally is a string consisting entirely of ASCII letters (e.g. C<EV>
		2177	or C<IOAsync>). The string C<AnyEvent::Impl::> gets prepended and the
		2178	resulting module name is loaded and - if the load was successful - used as
		2179	event model backend. If it fails to load then AnyEvent will proceed with
		2180	auto detection and -probing.
		2181
		2182	If the string ends with C<::> instead (e.g. C<AnyEvent::Impl::EV::>) then
		2183	nothing gets prepended and the module name is used as-is (hint: C<::> at
		2184	the end of a string designates a module name and quotes it appropriately).
		2185
		2186	For example, to force the pure perl model (L<AnyEvent::Loop::Perl>) you
		2187	could start your program like this:
		2188
		2189	PERL_ANYEVENT_MODEL=Perl perl ...
		2190
		2191	=item C<PERL_ANYEVENT_IO_MODEL>
		2192
		2193	The current file I/O model - see L<AnyEvent::IO> for more info.
		2194
		2195	At the moment, only C<Perl> (small, pure-perl, synchronous) and
		2196	C<IOAIO> (truly asynchronous) are supported. The default is C<IOAIO> if
		2197	L<AnyEvent::AIO> can be loaded, otherwise it is C<Perl>.
		2198
		2199	=item C<PERL_ANYEVENT_PROTOCOLS>
		2200
		2201	Used by both L<AnyEvent::DNS> and L<AnyEvent::Socket> to determine preferences
		2202	for IPv4 or IPv6. The default is unspecified (and might change, or be the result
		2203	of auto probing).
		2204
		2205	Must be set to a comma-separated list of protocols or address families,
		2206	current supported: C<ipv4> and C<ipv6>. Only protocols mentioned will be
		2207	used, and preference will be given to protocols mentioned earlier in the
		2208	list.
		2209
		2210	This variable can effectively be used for denial-of-service attacks
		2211	against local programs (e.g. when setuid), although the impact is likely
		2212	small, as the program has to handle conenction and other failures anyways.
		2213
		2214	Examples: C<PERL_ANYEVENT_PROTOCOLS=ipv4,ipv6> - prefer IPv4 over IPv6,
		2215	but support both and try to use both. C<PERL_ANYEVENT_PROTOCOLS=ipv4>
		2216	- only support IPv4, never try to resolve or contact IPv6
		2217	addresses. C<PERL_ANYEVENT_PROTOCOLS=ipv6,ipv4> support either IPv4 or
		2218	IPv6, but prefer IPv6 over IPv4.
		2219
		2220	=item C<PERL_ANYEVENT_HOSTS>
		2221
		2222	This variable, if specified, overrides the F</etc/hosts> file used by
		2223	L<AnyEvent::Socket>C<::resolve_sockaddr>, i.e. hosts aliases will be read
		2224	from that file instead.
		2225
		2226	=item C<PERL_ANYEVENT_EDNS0>
		2227
		2228	Used by L<AnyEvent::DNS> to decide whether to use the EDNS0 extension for
		2229	DNS. This extension is generally useful to reduce DNS traffic, especially
		2230	when DNSSEC is involved, but some (broken) firewalls drop such DNS
		2231	packets, which is why it is off by default.
		2232
		2233	Setting this variable to C<1> will cause L<AnyEvent::DNS> to announce
		2234	EDNS0 in its DNS requests.
		2235
		2236	=item C<PERL_ANYEVENT_MAX_FORKS>
		2237
		2238	The maximum number of child processes that C<AnyEvent::Util::fork_call>
		2239	will create in parallel.
		2240
		2241	=item C<PERL_ANYEVENT_MAX_OUTSTANDING_DNS>
		2242
		2243	The default value for the C<max_outstanding> parameter for the default DNS
		2244	resolver - this is the maximum number of parallel DNS requests that are
		2245	sent to the DNS server.
		2246
		2247	=item C<PERL_ANYEVENT_MAX_SIGNAL_LATENCY>
		2248
		2249	Perl has inherently racy signal handling (you can basically choose between
		2250	losing signals and memory corruption) - pure perl event loops (including
		2251	C<AnyEvent::Loop>, when C<Async::Interrupt> isn't available) therefore
		2252	have to poll regularly to avoid losing signals.
		2253
		2254	Some event loops are racy, but don't poll regularly, and some event loops
		2255	are written in C but are still racy. For those event loops, AnyEvent
		2256	installs a timer that regularly wakes up the event loop.
		2257
		2258	By default, the interval for this timer is C<10> seconds, but you can
		2259	override this delay with this environment variable (or by setting
		2260	the C<$AnyEvent::MAX_SIGNAL_LATENCY> variable before creating signal
		2261	watchers).
		2262
		2263	Lower values increase CPU (and energy) usage, higher values can introduce
		2264	long delays when reaping children or waiting for signals.
		2265
		2266	The L<AnyEvent::Async> module, if available, will be used to avoid this
		2267	polling (with most event loops).
		2268
		2269	=item C<PERL_ANYEVENT_RESOLV_CONF>
		2270
		2271	The absolute path to a F<resolv.conf>-style file to use instead of
		2272	F</etc/resolv.conf> (or the OS-specific configuration) in the default
		2273	resolver, or the empty string to select the default configuration.
		2274
		2275	=item C<PERL_ANYEVENT_CA_FILE>, C<PERL_ANYEVENT_CA_PATH>.
		2276
		2277	When neither C<ca_file> nor C<ca_path> was specified during
		2278	L<AnyEvent::TLS> context creation, and either of these environment
		2279	variables are nonempty, they will be used to specify CA certificate
		2280	locations instead of a system-dependent default.
		2281
		2282	=item C<PERL_ANYEVENT_AVOID_GUARD> and C<PERL_ANYEVENT_AVOID_ASYNC_INTERRUPT>
		2283
		2284	When these are set to C<1>, then the respective modules are not
		2285	loaded. Mostly good for testing AnyEvent itself.
		2286
		2287	=back
487		2288
488	=head1 SUPPLYING YOUR OWN EVENT MODEL INTERFACE	2289	=head1 SUPPLYING YOUR OWN EVENT MODEL INTERFACE
		2290
		2291	This is an advanced topic that you do not normally need to use AnyEvent in
		2292	a module. This section is only of use to event loop authors who want to
		2293	provide AnyEvent compatibility.
489		2294
490	If you need to support another event library which isn't directly	2295	If you need to support another event library which isn't directly
491	supported by AnyEvent, you can supply your own interface to it by	2296	supported by AnyEvent, you can supply your own interface to it by
492	pushing, before the first watcher gets created, the package name of	2297	pushing, before the first watcher gets created, the package name of
493	the event module and the package name of the interface to use onto	2298	the event module and the package name of the interface to use onto
494	C<@AnyEvent::REGISTRY>. You can do that before and even without loading	2299	C<@AnyEvent::REGISTRY>. You can do that before and even without loading
495	AnyEvent.	2300	AnyEvent, so it is reasonably cheap.
496		2301
497	Example:	2302	Example:
498		2303
499	push @AnyEvent::REGISTRY, [urxvt => urxvt::anyevent::];	2304	push @AnyEvent::REGISTRY, [urxvt => urxvt::anyevent::];
500		2305
501	This tells AnyEvent to (literally) use the C<urxvt::anyevent::>	2306	This tells AnyEvent to (literally) use the C<urxvt::anyevent::>
502	package/class when it finds the C<urxvt> package/module is loaded. When	2307	package/class when it finds the C<urxvt> package/module is already loaded.
		2308
503	AnyEvent is loaded and asked to find a suitable event model, it will	2309	When AnyEvent is loaded and asked to find a suitable event model, it
504	first check for the presence of urxvt.	2310	will first check for the presence of urxvt by trying to C<use> the
		2311	C<urxvt::anyevent> module.
505		2312
506	The class should provide implementations for all watcher types (see	2313	The class should provide implementations for all watcher types. See
507	L<AnyEvent::Impl::Event> (source code), L<AnyEvent::Impl::Glib>	2314	L<AnyEvent::Impl::EV> (source code), L<AnyEvent::Impl::Glib> (Source code)
508	(Source code) and so on for actual examples, use C<perldoc -m	2315	and so on for actual examples. Use C<perldoc -m AnyEvent::Impl::Glib> to
509	AnyEvent::Impl::Glib> to see the sources).	2316	see the sources.
510		2317
		2318	If you don't provide C<signal> and C<child> watchers than AnyEvent will
		2319	provide suitable (hopefully) replacements.
		2320
511	The above isn't fictitious, the I<rxvt-unicode> (a.k.a. urxvt)	2321	The above example isn't fictitious, the I<rxvt-unicode> (a.k.a. urxvt)
512	uses the above line as-is. An interface isn't included in AnyEvent	2322	terminal emulator uses the above line as-is. An interface isn't included
513	because it doesn't make sense outside the embedded interpreter inside	2323	in AnyEvent because it doesn't make sense outside the embedded interpreter
514	I<rxvt-unicode>, and it is updated and maintained as part of the	2324	inside I<rxvt-unicode>, and it is updated and maintained as part of the
515	I<rxvt-unicode> distribution.	2325	I<rxvt-unicode> distribution.
516		2326
517	I<rxvt-unicode> also cheats a bit by not providing blocking access to	2327	I<rxvt-unicode> also cheats a bit by not providing blocking access to
518	condition variables: code blocking while waiting for a condition will	2328	condition variables: code blocking while waiting for a condition will
519	C<die>. This still works with most modules/usages, and blocking calls must	2329	C<die>. This still works with most modules/usages, and blocking calls must
520	not be in an interactive application, so it makes sense.	2330	not be done in an interactive application, so it makes sense.
521		2331
522	=head1 ENVIRONMENT VARIABLES
523
524	The following environment variables are used by this module:
525
526	C<PERL_ANYEVENT_VERBOSE> when set to C<2> or higher, reports which event
527	model gets used.
528
529	=head1 EXAMPLE	2332	=head1 EXAMPLE PROGRAM
530		2333
531	The following program uses an io watcher to read data from stdin, a timer	2334	The following program uses an I/O watcher to read data from STDIN, a timer
532	to display a message once per second, and a condvar to exit the program	2335	to display a message once per second, and a condition variable to quit the
533	when the user enters quit:	2336	program when the user enters quit:
534		2337
535	use AnyEvent;	2338	use AnyEvent;
536		2339
537	my $cv = AnyEvent->condvar;	2340	my $cv = AnyEvent->condvar;
538		2341
539	my $io_watcher = AnyEvent->io (fh => \*STDIN, poll => 'r', cb => sub {	2342	my $io_watcher = AnyEvent->io (
		2343	fh => \*STDIN,
		2344	poll => 'r',
		2345	cb => sub {
540	warn "io event <$_[0]>\n"; # will always output <r>	2346	warn "io event <$_[0]>\n"; # will always output <r>
541	chomp (my $input = <STDIN>); # read a line	2347	chomp (my $input = <STDIN>); # read a line
542	warn "read: $input\n"; # output what has been read	2348	warn "read: $input\n"; # output what has been read
543	$cv->broadcast if $input =~ /^q/i; # quit program if /^q/i	2349	$cv->send if $input =~ /^q/i; # quit program if /^q/i
		2350	},
		2351	);
		2352
		2353	my $time_watcher = AnyEvent->timer (after => 1, interval => 1, cb => sub {
		2354	warn "timeout\n"; # print 'timeout' at most every second
544	});	2355	});
545		2356
546	my $time_watcher; # can only be used once
547
548	sub new_timer {
549	$timer = AnyEvent->timer (after => 1, cb => sub {
550	warn "timeout\n"; # print 'timeout' about every second
551	&new_timer; # and restart the time
552	});
553	}
554
555	new_timer; # create first timer
556
557	$cv->wait; # wait until user enters /^q/i	2357	$cv->recv; # wait until user enters /^q/i
558		2358
559	=head1 REAL-WORLD EXAMPLE	2359	=head1 REAL-WORLD EXAMPLE
560		2360
561	Consider the L<Net::FCP> module. It features (among others) the following	2361	Consider the L<Net::FCP> module. It features (among others) the following
562	API calls, which are to freenet what HTTP GET requests are to http:	2362	API calls, which are to freenet what HTTP GET requests are to http:
…		…
612	syswrite $txn->{fh}, $txn->{request}	2412	syswrite $txn->{fh}, $txn->{request}
613	or die "connection or write error";	2413	or die "connection or write error";
614	$txn->{w} = AnyEvent->io (fh => $txn->{fh}, poll => 'r', cb => sub { $txn->fh_ready_r });	2414	$txn->{w} = AnyEvent->io (fh => $txn->{fh}, poll => 'r', cb => sub { $txn->fh_ready_r });
615		2415
616	Again, C<fh_ready_r> waits till all data has arrived, and then stores the	2416	Again, C<fh_ready_r> waits till all data has arrived, and then stores the
617	result and signals any possible waiters that the request ahs finished:	2417	result and signals any possible waiters that the request has finished:
618		2418
619	sysread $txn->{fh}, $txn->{buf}, length $txn->{$buf};	2419	sysread $txn->{fh}, $txn->{buf}, length $txn->{$buf};
620		2420
621	if (end-of-file or data complete) {	2421	if (end-of-file or data complete) {
622	$txn->{result} = $txn->{buf};	2422	$txn->{result} = $txn->{buf};
623	$txn->{finished}->broadcast;	2423	$txn->{finished}->send;
624	$txb->{cb}->($txn) of $txn->{cb}; # also call callback	2424	$txb->{cb}->($txn) of $txn->{cb}; # also call callback
625	}	2425	}
626		2426
627	The C<result> method, finally, just waits for the finished signal (if the	2427	The C<result> method, finally, just waits for the finished signal (if the
628	request was already finished, it doesn't wait, of course, and returns the	2428	request was already finished, it doesn't wait, of course, and returns the
629	data:	2429	data:
630		2430
631	$txn->{finished}->wait;	2431	$txn->{finished}->recv;
632	return $txn->{result};	2432	return $txn->{result};
633		2433
634	The actual code goes further and collects all errors (C<die>s, exceptions)	2434	The actual code goes further and collects all errors (C<die>s, exceptions)
635	that occured during request processing. The C<result> method detects	2435	that occurred during request processing. The C<result> method detects
636	wether an exception as thrown (it is stored inside the $txn object)	2436	whether an exception as thrown (it is stored inside the $txn object)
637	and just throws the exception, which means connection errors and other	2437	and just throws the exception, which means connection errors and other
638	problems get reported tot he code that tries to use the result, not in a	2438	problems get reported to the code that tries to use the result, not in a
639	random callback.	2439	random callback.
640		2440
641	All of this enables the following usage styles:	2441	All of this enables the following usage styles:
642		2442
643	1. Blocking:	2443	1. Blocking:
644		2444
645	my $data = $fcp->client_get ($url);	2445	my $data = $fcp->client_get ($url);
646		2446
647	2. Blocking, but parallelizing:	2447	2. Blocking, but running in parallel:
648		2448
649	my @datas = map $_->result,	2449	my @datas = map $_->result,
650	map $fcp->txn_client_get ($_),	2450	map $fcp->txn_client_get ($_),
651	@urls;	2451	@urls;
652		2452
653	Both blocking examples work without the module user having to know	2453	Both blocking examples work without the module user having to know
654	anything about events.	2454	anything about events.
655		2455
656	3a. Event-based in a main program, using any support Event module:	2456	3a. Event-based in a main program, using any supported event module:
657		2457
658	use Event;	2458	use EV;
659		2459
660	$fcp->txn_client_get ($url)->cb (sub {	2460	$fcp->txn_client_get ($url)->cb (sub {
661	my $txn = shift;	2461	my $txn = shift;
662	my $data = $txn->result;	2462	my $data = $txn->result;
663	...	2463	...
664	});	2464	});
665		2465
666	Event::loop;	2466	EV::loop;
667		2467
668	3b. The module user could use AnyEvent, too:	2468	3b. The module user could use AnyEvent, too:
669		2469
670	use AnyEvent;	2470	use AnyEvent;
671		2471
672	my $quit = AnyEvent->condvar;	2472	my $quit = AnyEvent->condvar;
673		2473
674	$fcp->txn_client_get ($url)->cb (sub {	2474	$fcp->txn_client_get ($url)->cb (sub {
675	...	2475	...
676	$quit->broadcast;	2476	$quit->send;
677	});	2477	});
678		2478
679	$quit->wait;	2479	$quit->recv;
		2480
		2481
		2482	=head1 BENCHMARKS
		2483
		2484	To give you an idea of the performance and overheads that AnyEvent adds
		2485	over the event loops themselves and to give you an impression of the speed
		2486	of various event loops I prepared some benchmarks.
		2487
		2488	=head2 BENCHMARKING ANYEVENT OVERHEAD
		2489
		2490	Here is a benchmark of various supported event models used natively and
		2491	through AnyEvent. The benchmark creates a lot of timers (with a zero
		2492	timeout) and I/O watchers (watching STDOUT, a pty, to become writable,
		2493	which it is), lets them fire exactly once and destroys them again.
		2494
		2495	Source code for this benchmark is found as F<eg/bench> in the AnyEvent
		2496	distribution. It uses the L<AE> interface, which makes a real difference
		2497	for the EV and Perl backends only.
		2498
		2499	=head3 Explanation of the columns
		2500
		2501	I<watcher> is the number of event watchers created/destroyed. Since
		2502	different event models feature vastly different performances, each event
		2503	loop was given a number of watchers so that overall runtime is acceptable
		2504	and similar between tested event loop (and keep them from crashing): Glib
		2505	would probably take thousands of years if asked to process the same number
		2506	of watchers as EV in this benchmark.
		2507
		2508	I<bytes> is the number of bytes (as measured by the resident set size,
		2509	RSS) consumed by each watcher. This method of measuring captures both C
		2510	and Perl-based overheads.
		2511
		2512	I<create> is the time, in microseconds (millionths of seconds), that it
		2513	takes to create a single watcher. The callback is a closure shared between
		2514	all watchers, to avoid adding memory overhead. That means closure creation
		2515	and memory usage is not included in the figures.
		2516
		2517	I<invoke> is the time, in microseconds, used to invoke a simple
		2518	callback. The callback simply counts down a Perl variable and after it was
		2519	invoked "watcher" times, it would C<< ->send >> a condvar once to
		2520	signal the end of this phase.
		2521
		2522	I<destroy> is the time, in microseconds, that it takes to destroy a single
		2523	watcher.
		2524
		2525	=head3 Results
		2526
		2527	name watchers bytes create invoke destroy comment
		2528	EV/EV 100000 223 0.47 0.43 0.27 EV native interface
		2529	EV/Any 100000 223 0.48 0.42 0.26 EV + AnyEvent watchers
		2530	Coro::EV/Any 100000 223 0.47 0.42 0.26 coroutines + Coro::Signal
		2531	Perl/Any 100000 431 2.70 0.74 0.92 pure perl implementation
		2532	Event/Event 16000 516 31.16 31.84 0.82 Event native interface
		2533	Event/Any 16000 1203 42.61 34.79 1.80 Event + AnyEvent watchers
		2534	IOAsync/Any 16000 1911 41.92 27.45 16.81 via IO::Async::Loop::IO_Poll
		2535	IOAsync/Any 16000 1726 40.69 26.37 15.25 via IO::Async::Loop::Epoll
		2536	Glib/Any 16000 1118 89.00 12.57 51.17 quadratic behaviour
		2537	Tk/Any 2000 1346 20.96 10.75 8.00 SEGV with >> 2000 watchers
		2538	POE/Any 2000 6951 108.97 795.32 14.24 via POE::Loop::Event
		2539	POE/Any 2000 6648 94.79 774.40 575.51 via POE::Loop::Select
		2540
		2541	=head3 Discussion
		2542
		2543	The benchmark does I<not> measure scalability of the event loop very
		2544	well. For example, a select-based event loop (such as the pure perl one)
		2545	can never compete with an event loop that uses epoll when the number of
		2546	file descriptors grows high. In this benchmark, all events become ready at
		2547	the same time, so select/poll-based implementations get an unnatural speed
		2548	boost.
		2549
		2550	Also, note that the number of watchers usually has a nonlinear effect on
		2551	overall speed, that is, creating twice as many watchers doesn't take twice
		2552	the time - usually it takes longer. This puts event loops tested with a
		2553	higher number of watchers at a disadvantage.
		2554
		2555	To put the range of results into perspective, consider that on the
		2556	benchmark machine, handling an event takes roughly 1600 CPU cycles with
		2557	EV, 3100 CPU cycles with AnyEvent's pure perl loop and almost 3000000 CPU
		2558	cycles with POE.
		2559
		2560	C<EV> is the sole leader regarding speed and memory use, which are both
		2561	maximal/minimal, respectively. When using the L<AE> API there is zero
		2562	overhead (when going through the AnyEvent API create is about 5-6 times
		2563	slower, with other times being equal, so still uses far less memory than
		2564	any other event loop and is still faster than Event natively).
		2565
		2566	The pure perl implementation is hit in a few sweet spots (both the
		2567	constant timeout and the use of a single fd hit optimisations in the perl
		2568	interpreter and the backend itself). Nevertheless this shows that it
		2569	adds very little overhead in itself. Like any select-based backend its
		2570	performance becomes really bad with lots of file descriptors (and few of
		2571	them active), of course, but this was not subject of this benchmark.
		2572
		2573	The C<Event> module has a relatively high setup and callback invocation
		2574	cost, but overall scores in on the third place.
		2575
		2576	C<IO::Async> performs admirably well, about on par with C<Event>, even
		2577	when using its pure perl backend.
		2578
		2579	C<Glib>'s memory usage is quite a bit higher, but it features a
		2580	faster callback invocation and overall ends up in the same class as
		2581	C<Event>. However, Glib scales extremely badly, doubling the number of
		2582	watchers increases the processing time by more than a factor of four,
		2583	making it completely unusable when using larger numbers of watchers
		2584	(note that only a single file descriptor was used in the benchmark, so
		2585	inefficiencies of C<poll> do not account for this).
		2586
		2587	The C<Tk> adaptor works relatively well. The fact that it crashes with
		2588	more than 2000 watchers is a big setback, however, as correctness takes
		2589	precedence over speed. Nevertheless, its performance is surprising, as the
		2590	file descriptor is dup()ed for each watcher. This shows that the dup()
		2591	employed by some adaptors is not a big performance issue (it does incur a
		2592	hidden memory cost inside the kernel which is not reflected in the figures
		2593	above).
		2594
		2595	C<POE>, regardless of underlying event loop (whether using its pure perl
		2596	select-based backend or the Event module, the POE-EV backend couldn't
		2597	be tested because it wasn't working) shows abysmal performance and
		2598	memory usage with AnyEvent: Watchers use almost 30 times as much memory
		2599	as EV watchers, and 10 times as much memory as Event (the high memory
		2600	requirements are caused by requiring a session for each watcher). Watcher
		2601	invocation speed is almost 900 times slower than with AnyEvent's pure perl
		2602	implementation.
		2603
		2604	The design of the POE adaptor class in AnyEvent can not really account
		2605	for the performance issues, though, as session creation overhead is
		2606	small compared to execution of the state machine, which is coded pretty
		2607	optimally within L<AnyEvent::Impl::POE> (and while everybody agrees that
		2608	using multiple sessions is not a good approach, especially regarding
		2609	memory usage, even the author of POE could not come up with a faster
		2610	design).
		2611
		2612	=head3 Summary
		2613
		2614	=over 4
		2615
		2616	=item * Using EV through AnyEvent is faster than any other event loop
		2617	(even when used without AnyEvent), but most event loops have acceptable
		2618	performance with or without AnyEvent.
		2619
		2620	=item * The overhead AnyEvent adds is usually much smaller than the overhead of
		2621	the actual event loop, only with extremely fast event loops such as EV
		2622	does AnyEvent add significant overhead.
		2623
		2624	=item * You should avoid POE like the plague if you want performance or
		2625	reasonable memory usage.
		2626
		2627	=back
		2628
		2629	=head2 BENCHMARKING THE LARGE SERVER CASE
		2630
		2631	This benchmark actually benchmarks the event loop itself. It works by
		2632	creating a number of "servers": each server consists of a socket pair, a
		2633	timeout watcher that gets reset on activity (but never fires), and an I/O
		2634	watcher waiting for input on one side of the socket. Each time the socket
		2635	watcher reads a byte it will write that byte to a random other "server".
		2636
		2637	The effect is that there will be a lot of I/O watchers, only part of which
		2638	are active at any one point (so there is a constant number of active
		2639	fds for each loop iteration, but which fds these are is random). The
		2640	timeout is reset each time something is read because that reflects how
		2641	most timeouts work (and puts extra pressure on the event loops).
		2642
		2643	In this benchmark, we use 10000 socket pairs (20000 sockets), of which 100
		2644	(1%) are active. This mirrors the activity of large servers with many
		2645	connections, most of which are idle at any one point in time.
		2646
		2647	Source code for this benchmark is found as F<eg/bench2> in the AnyEvent
		2648	distribution. It uses the L<AE> interface, which makes a real difference
		2649	for the EV and Perl backends only.
		2650
		2651	=head3 Explanation of the columns
		2652
		2653	I<sockets> is the number of sockets, and twice the number of "servers" (as
		2654	each server has a read and write socket end).
		2655
		2656	I<create> is the time it takes to create a socket pair (which is
		2657	nontrivial) and two watchers: an I/O watcher and a timeout watcher.
		2658
		2659	I<request>, the most important value, is the time it takes to handle a
		2660	single "request", that is, reading the token from the pipe and forwarding
		2661	it to another server. This includes deleting the old timeout and creating
		2662	a new one that moves the timeout into the future.
		2663
		2664	=head3 Results
		2665
		2666	name sockets create request
		2667	EV 20000 62.66 7.99
		2668	Perl 20000 68.32 32.64
		2669	IOAsync 20000 174.06 101.15 epoll
		2670	IOAsync 20000 174.67 610.84 poll
		2671	Event 20000 202.69 242.91
		2672	Glib 20000 557.01 1689.52
		2673	POE 20000 341.54 12086.32 uses POE::Loop::Event
		2674
		2675	=head3 Discussion
		2676
		2677	This benchmark I<does> measure scalability and overall performance of the
		2678	particular event loop.
		2679
		2680	EV is again fastest. Since it is using epoll on my system, the setup time
		2681	is relatively high, though.
		2682
		2683	Perl surprisingly comes second. It is much faster than the C-based event
		2684	loops Event and Glib.
		2685
		2686	IO::Async performs very well when using its epoll backend, and still quite
		2687	good compared to Glib when using its pure perl backend.
		2688
		2689	Event suffers from high setup time as well (look at its code and you will
		2690	understand why). Callback invocation also has a high overhead compared to
		2691	the C<< $_->() for .. >>-style loop that the Perl event loop uses. Event
		2692	uses select or poll in basically all documented configurations.
		2693
		2694	Glib is hit hard by its quadratic behaviour w.r.t. many watchers. It
		2695	clearly fails to perform with many filehandles or in busy servers.
		2696
		2697	POE is still completely out of the picture, taking over 1000 times as long
		2698	as EV, and over 100 times as long as the Perl implementation, even though
		2699	it uses a C-based event loop in this case.
		2700
		2701	=head3 Summary
		2702
		2703	=over 4
		2704
		2705	=item * The pure perl implementation performs extremely well.
		2706
		2707	=item * Avoid Glib or POE in large projects where performance matters.
		2708
		2709	=back
		2710
		2711	=head2 BENCHMARKING SMALL SERVERS
		2712
		2713	While event loops should scale (and select-based ones do not...) even to
		2714	large servers, most programs we (or I :) actually write have only a few
		2715	I/O watchers.
		2716
		2717	In this benchmark, I use the same benchmark program as in the large server
		2718	case, but it uses only eight "servers", of which three are active at any
		2719	one time. This should reflect performance for a small server relatively
		2720	well.
		2721
		2722	The columns are identical to the previous table.
		2723
		2724	=head3 Results
		2725
		2726	name sockets create request
		2727	EV 16 20.00 6.54
		2728	Perl 16 25.75 12.62
		2729	Event 16 81.27 35.86
		2730	Glib 16 32.63 15.48
		2731	POE 16 261.87 276.28 uses POE::Loop::Event
		2732
		2733	=head3 Discussion
		2734
		2735	The benchmark tries to test the performance of a typical small
		2736	server. While knowing how various event loops perform is interesting, keep
		2737	in mind that their overhead in this case is usually not as important, due
		2738	to the small absolute number of watchers (that is, you need efficiency and
		2739	speed most when you have lots of watchers, not when you only have a few of
		2740	them).
		2741
		2742	EV is again fastest.
		2743
		2744	Perl again comes second. It is noticeably faster than the C-based event
		2745	loops Event and Glib, although the difference is too small to really
		2746	matter.
		2747
		2748	POE also performs much better in this case, but is is still far behind the
		2749	others.
		2750
		2751	=head3 Summary
		2752
		2753	=over 4
		2754
		2755	=item * C-based event loops perform very well with small number of
		2756	watchers, as the management overhead dominates.
		2757
		2758	=back
		2759
		2760	=head2 THE IO::Lambda BENCHMARK
		2761
		2762	Recently I was told about the benchmark in the IO::Lambda manpage, which
		2763	could be misinterpreted to make AnyEvent look bad. In fact, the benchmark
		2764	simply compares IO::Lambda with POE, and IO::Lambda looks better (which
		2765	shouldn't come as a surprise to anybody). As such, the benchmark is
		2766	fine, and mostly shows that the AnyEvent backend from IO::Lambda isn't
		2767	very optimal. But how would AnyEvent compare when used without the extra
		2768	baggage? To explore this, I wrote the equivalent benchmark for AnyEvent.
		2769
		2770	The benchmark itself creates an echo-server, and then, for 500 times,
		2771	connects to the echo server, sends a line, waits for the reply, and then
		2772	creates the next connection. This is a rather bad benchmark, as it doesn't
		2773	test the efficiency of the framework or much non-blocking I/O, but it is a
		2774	benchmark nevertheless.
		2775
		2776	name runtime
		2777	Lambda/select 0.330 sec
		2778	+ optimized 0.122 sec
		2779	Lambda/AnyEvent 0.327 sec
		2780	+ optimized 0.138 sec
		2781	Raw sockets/select 0.077 sec
		2782	POE/select, components 0.662 sec
		2783	POE/select, raw sockets 0.226 sec
		2784	POE/select, optimized 0.404 sec
		2785
		2786	AnyEvent/select/nb 0.085 sec
		2787	AnyEvent/EV/nb 0.068 sec
		2788	+state machine 0.134 sec
		2789
		2790	The benchmark is also a bit unfair (my fault): the IO::Lambda/POE
		2791	benchmarks actually make blocking connects and use 100% blocking I/O,
		2792	defeating the purpose of an event-based solution. All of the newly
		2793	written AnyEvent benchmarks use 100% non-blocking connects (using
		2794	AnyEvent::Socket::tcp_connect and the asynchronous pure perl DNS
		2795	resolver), so AnyEvent is at a disadvantage here, as non-blocking connects
		2796	generally require a lot more bookkeeping and event handling than blocking
		2797	connects (which involve a single syscall only).
		2798
		2799	The last AnyEvent benchmark additionally uses L<AnyEvent::Handle>, which
		2800	offers similar expressive power as POE and IO::Lambda, using conventional
		2801	Perl syntax. This means that both the echo server and the client are 100%
		2802	non-blocking, further placing it at a disadvantage.
		2803
		2804	As you can see, the AnyEvent + EV combination even beats the
		2805	hand-optimised "raw sockets benchmark", while AnyEvent + its pure perl
		2806	backend easily beats IO::Lambda and POE.
		2807
		2808	And even the 100% non-blocking version written using the high-level (and
		2809	slow :) L<AnyEvent::Handle> abstraction beats both POE and IO::Lambda
		2810	higher level ("unoptimised") abstractions by a large margin, even though
		2811	it does all of DNS, tcp-connect and socket I/O in a non-blocking way.
		2812
		2813	The two AnyEvent benchmarks programs can be found as F<eg/ae0.pl> and
		2814	F<eg/ae2.pl> in the AnyEvent distribution, the remaining benchmarks are
		2815	part of the IO::Lambda distribution and were used without any changes.
		2816
		2817
		2818	=head1 SIGNALS
		2819
		2820	AnyEvent currently installs handlers for these signals:
		2821
		2822	=over 4
		2823
		2824	=item SIGCHLD
		2825
		2826	A handler for C<SIGCHLD> is installed by AnyEvent's child watcher
		2827	emulation for event loops that do not support them natively. Also, some
		2828	event loops install a similar handler.
		2829
		2830	Additionally, when AnyEvent is loaded and SIGCHLD is set to IGNORE, then
		2831	AnyEvent will reset it to default, to avoid losing child exit statuses.
		2832
		2833	=item SIGPIPE
		2834
		2835	A no-op handler is installed for C<SIGPIPE> when C<$SIG{PIPE}> is C<undef>
		2836	when AnyEvent gets loaded.
		2837
		2838	The rationale for this is that AnyEvent users usually do not really depend
		2839	on SIGPIPE delivery (which is purely an optimisation for shell use, or
		2840	badly-written programs), but C<SIGPIPE> can cause spurious and rare
		2841	program exits as a lot of people do not expect C<SIGPIPE> when writing to
		2842	some random socket.
		2843
		2844	The rationale for installing a no-op handler as opposed to ignoring it is
		2845	that this way, the handler will be restored to defaults on exec.
		2846
		2847	Feel free to install your own handler, or reset it to defaults.
		2848
		2849	=back
		2850
		2851	=cut
		2852
		2853	undef $SIG{CHLD}
		2854	if $SIG{CHLD} eq 'IGNORE';
		2855
		2856	$SIG{PIPE} = sub { }
		2857	unless defined $SIG{PIPE};
		2858
		2859	=head1 RECOMMENDED/OPTIONAL MODULES
		2860
		2861	One of AnyEvent's main goals is to be 100% Pure-Perl(tm): only perl (and
		2862	its built-in modules) are required to use it.
		2863
		2864	That does not mean that AnyEvent won't take advantage of some additional
		2865	modules if they are installed.
		2866
		2867	This section explains which additional modules will be used, and how they
		2868	affect AnyEvent's operation.
		2869
		2870	=over 4
		2871
		2872	=item L<Async::Interrupt>
		2873
		2874	This slightly arcane module is used to implement fast signal handling: To
		2875	my knowledge, there is no way to do completely race-free and quick
		2876	signal handling in pure perl. To ensure that signals still get
		2877	delivered, AnyEvent will start an interval timer to wake up perl (and
		2878	catch the signals) with some delay (default is 10 seconds, look for
		2879	C<$AnyEvent::MAX_SIGNAL_LATENCY>).
		2880
		2881	If this module is available, then it will be used to implement signal
		2882	catching, which means that signals will not be delayed, and the event loop
		2883	will not be interrupted regularly, which is more efficient (and good for
		2884	battery life on laptops).
		2885
		2886	This affects not just the pure-perl event loop, but also other event loops
		2887	that have no signal handling on their own (e.g. Glib, Tk, Qt).
		2888
		2889	Some event loops (POE, Event, Event::Lib) offer signal watchers natively,
		2890	and either employ their own workarounds (POE) or use AnyEvent's workaround
		2891	(using C<$AnyEvent::MAX_SIGNAL_LATENCY>). Installing L<Async::Interrupt>
		2892	does nothing for those backends.
		2893
		2894	=item L<EV>
		2895
		2896	This module isn't really "optional", as it is simply one of the backend
		2897	event loops that AnyEvent can use. However, it is simply the best event
		2898	loop available in terms of features, speed and stability: It supports
		2899	the AnyEvent API optimally, implements all the watcher types in XS, does
		2900	automatic timer adjustments even when no monotonic clock is available,
		2901	can take avdantage of advanced kernel interfaces such as C<epoll> and
		2902	C<kqueue>, and is the fastest backend I<by far>. You can even embed
		2903	L<Glib>/L<Gtk2> in it (or vice versa, see L<EV::Glib> and L<Glib::EV>).
		2904
		2905	If you only use backends that rely on another event loop (e.g. C<Tk>),
		2906	then this module will do nothing for you.
		2907
		2908	=item L<Guard>
		2909
		2910	The guard module, when used, will be used to implement
		2911	C<AnyEvent::Util::guard>. This speeds up guards considerably (and uses a
		2912	lot less memory), but otherwise doesn't affect guard operation much. It is
		2913	purely used for performance.
		2914
		2915	=item L<JSON> and L<JSON::XS>
		2916
		2917	One of these modules is required when you want to read or write JSON data
		2918	via L<AnyEvent::Handle>. L<JSON> is also written in pure-perl, but can take
		2919	advantage of the ultra-high-speed L<JSON::XS> module when it is installed.
		2920
		2921	=item L<Net::SSLeay>
		2922
		2923	Implementing TLS/SSL in Perl is certainly interesting, but not very
		2924	worthwhile: If this module is installed, then L<AnyEvent::Handle> (with
		2925	the help of L<AnyEvent::TLS>), gains the ability to do TLS/SSL.
		2926
		2927	=item L<Time::HiRes>
		2928
		2929	This module is part of perl since release 5.008. It will be used when the
		2930	chosen event library does not come with a timing source of its own. The
		2931	pure-perl event loop (L<AnyEvent::Loop>) will additionally load it to
		2932	try to use a monotonic clock for timing stability.
		2933
		2934	=item L<AnyEvent::AIO> (and L<IO::AIO>)
		2935
		2936	The default implementation of L<AnyEvent::IO> is to do I/O synchronously,
		2937	stopping programs while they access the disk, which is fine for a lot of
		2938	programs.
		2939
		2940	Installing AnyEvent::AIO (and its IO::AIO dependency) makes it switch to
		2941	a true asynchronous implementation, so event processing can continue even
		2942	while waiting for disk I/O.
		2943
		2944	=back
		2945
		2946
		2947	=head1 FORK
		2948
		2949	Most event libraries are not fork-safe. The ones who are usually are
		2950	because they rely on inefficient but fork-safe C<select> or C<poll> calls
		2951	- higher performance APIs such as BSD's kqueue or the dreaded Linux epoll
		2952	are usually badly thought-out hacks that are incompatible with fork in
		2953	one way or another. Only L<EV> is fully fork-aware and ensures that you
		2954	continue event-processing in both parent and child (or both, if you know
		2955	what you are doing).
		2956
		2957	This means that, in general, you cannot fork and do event processing in
		2958	the child if the event library was initialised before the fork (which
		2959	usually happens when the first AnyEvent watcher is created, or the library
		2960	is loaded).
		2961
		2962	If you have to fork, you must either do so I<before> creating your first
		2963	watcher OR you must not use AnyEvent at all in the child OR you must do
		2964	something completely out of the scope of AnyEvent.
		2965
		2966	The problem of doing event processing in the parent I<and> the child
		2967	is much more complicated: even for backends that I<are> fork-aware or
		2968	fork-safe, their behaviour is not usually what you want: fork clones all
		2969	watchers, that means all timers, I/O watchers etc. are active in both
		2970	parent and child, which is almost never what you want. USing C<exec>
		2971	to start worker children from some kind of manage rprocess is usually
		2972	preferred, because it is much easier and cleaner, at the expense of having
		2973	to have another binary.
		2974
		2975
		2976	=head1 SECURITY CONSIDERATIONS
		2977
		2978	AnyEvent can be forced to load any event model via
		2979	$ENV{PERL_ANYEVENT_MODEL}. While this cannot (to my knowledge) be used to
		2980	execute arbitrary code or directly gain access, it can easily be used to
		2981	make the program hang or malfunction in subtle ways, as AnyEvent watchers
		2982	will not be active when the program uses a different event model than
		2983	specified in the variable.
		2984
		2985	You can make AnyEvent completely ignore this variable by deleting it
		2986	before the first watcher gets created, e.g. with a C<BEGIN> block:
		2987
		2988	BEGIN { delete $ENV{PERL_ANYEVENT_MODEL} }
		2989
		2990	use AnyEvent;
		2991
		2992	Similar considerations apply to $ENV{PERL_ANYEVENT_VERBOSE}, as that can
		2993	be used to probe what backend is used and gain other information (which is
		2994	probably even less useful to an attacker than PERL_ANYEVENT_MODEL), and
		2995	$ENV{PERL_ANYEVENT_STRICT}.
		2996
		2997	Note that AnyEvent will remove I<all> environment variables starting with
		2998	C<PERL_ANYEVENT_> from C<%ENV> when it is loaded while taint mode is
		2999	enabled.
		3000
		3001
		3002	=head1 BUGS
		3003
		3004	Perl 5.8 has numerous memleaks that sometimes hit this module and are hard
		3005	to work around. If you suffer from memleaks, first upgrade to Perl 5.10
		3006	and check wether the leaks still show up. (Perl 5.10.0 has other annoying
		3007	memleaks, such as leaking on C<map> and C<grep> but it is usually not as
		3008	pronounced).
		3009
680		3010
681	=head1 SEE ALSO	3011	=head1 SEE ALSO
682		3012
683	Event modules: L<Coro::Event>, L<Coro>, L<Event>, L<Glib::Event>, L<Glib>.	3013	Tutorial/Introduction: L<AnyEvent::Intro>.
684		3014
685	Implementations: L<AnyEvent::Impl::Coro>, L<AnyEvent::Impl::Event>, L<AnyEvent::Impl::Glib>, L<AnyEvent::Impl::Tk>.	3015	FAQ: L<AnyEvent::FAQ>.
686		3016
687	Nontrivial usage example: L<Net::FCP>.	3017	Utility functions: L<AnyEvent::Util> (misc. grab-bag), L<AnyEvent::Log>
		3018	(simply logging).
688		3019
689	=head1	3020	Development/Debugging: L<AnyEvent::Strict> (stricter checking),
		3021	L<AnyEvent::Debug> (interactive shell, watcher tracing).
		3022
		3023	Supported event modules: L<AnyEvent::Loop>, L<EV>, L<EV::Glib>,
		3024	L<Glib::EV>, L<Event>, L<Glib::Event>, L<Glib>, L<Tk>, L<Event::Lib>,
		3025	L<Qt>, L<POE>, L<FLTK>.
		3026
		3027	Implementations: L<AnyEvent::Impl::EV>, L<AnyEvent::Impl::Event>,
		3028	L<AnyEvent::Impl::Glib>, L<AnyEvent::Impl::Tk>, L<AnyEvent::Impl::Perl>,
		3029	L<AnyEvent::Impl::EventLib>, L<AnyEvent::Impl::Qt>,
		3030	L<AnyEvent::Impl::POE>, L<AnyEvent::Impl::IOAsync>, L<Anyevent::Impl::Irssi>,
		3031	L<AnyEvent::Impl::FLTK>.
		3032
		3033	Non-blocking handles, pipes, stream sockets, TCP clients and
		3034	servers: L<AnyEvent::Handle>, L<AnyEvent::Socket>, L<AnyEvent::TLS>.
		3035
		3036	Asynchronous File I/O: L<AnyEvent::IO>.
		3037
		3038	Asynchronous DNS: L<AnyEvent::DNS>.
		3039
		3040	Thread support: L<Coro>, L<Coro::AnyEvent>, L<Coro::EV>, L<Coro::Event>.
		3041
		3042	Nontrivial usage examples: L<AnyEvent::GPSD>, L<AnyEvent::IRC>,
		3043	L<AnyEvent::HTTP>.
		3044
		3045
		3046	=head1 AUTHOR
		3047
		3048	Marc Lehmann <schmorp@schmorp.de>
		3049	http://anyevent.schmorp.de
690		3050
691	=cut	3051	=cut
692		3052
693	1	3053	1
694		3054

Diff Legend

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