ViewVC Help

View File | Revision Log | Show Annotations | Download File

/cvs/AnyEvent/lib/AnyEvent.pm

Comparing AnyEvent/lib/AnyEvent.pm (file contents):
Revision 1.28 by root, Sat Oct 27 15:10:09 2007 UTC vs.
Revision 1.249 by root, Mon Jul 20 06:00:42 2009 UTC

1	=head1 NAME	1	=head1 NAME
2		2
3	AnyEvent - provide framework for multiple event loops	3	AnyEvent - events independent of event loop implementation
4		4
5	Event, Coro, Glib, Tk, Perl - various supported event loops	5	EV, Event, Glib, Tk, Perl, Event::Lib, Qt and POE are various supported
		6	event loops.
6		7
7	=head1 SYNOPSIS	8	=head1 SYNOPSIS
8		9
9	use AnyEvent;	10	use AnyEvent;
10		11
		12	# file descriptor readable
11	my $w = AnyEvent->io (fh => $fh, poll => "r|w", cb => sub {	13	my $w = AnyEvent->io (fh => $fh, poll => "r", cb => sub { ... });
		14
		15	# one-shot or repeating timers
		16	my $w = AnyEvent->timer (after => $seconds, cb => sub { ... });
		17	my $w = AnyEvent->timer (after => $seconds, interval => $seconds, cb => ...
		18
		19	print AnyEvent->now; # prints current event loop time
		20	print AnyEvent->time; # think Time::HiRes::time or simply CORE::time.
		21
		22	# POSIX signal
		23	my $w = AnyEvent->signal (signal => "TERM", cb => sub { ... });
		24
		25	# child process exit
		26	my $w = AnyEvent->child (pid => $pid, cb => sub {
		27	my ($pid, $status) = @_;
12	...	28	...
13	});	29	});
14		30
15	my $w = AnyEvent->timer (after => $seconds, cb => sub {	31	# called when event loop idle (if applicable)
16	...	32	my $w = AnyEvent->idle (cb => sub { ... });
17	});
18		33
19	my $w = AnyEvent->condvar; # stores wether a condition was flagged	34	my $w = AnyEvent->condvar; # stores whether a condition was flagged
		35	$w->send; # wake up current and all future recv's
20	$w->wait; # enters "main loop" till $condvar gets ->broadcast	36	$w->recv; # enters "main loop" till $condvar gets ->send
21	$w->broadcast; # wake up current and all future wait's	37	# use a condvar in callback mode:
		38	$w->cb (sub { $_[0]->recv });
		39
		40	=head1 INTRODUCTION/TUTORIAL
		41
		42	This manpage is mainly a reference manual. If you are interested
		43	in a tutorial or some gentle introduction, have a look at the
		44	L<AnyEvent::Intro> manpage.
		45
		46	=head1 SUPPORT
		47
		48	There is a mailinglist for discussing all things AnyEvent, and an IRC
		49	channel, too.
		50
		51	See the AnyEvent project page at the B<Schmorpforge Ta-Sa Software
		52	Respository>, at L<http://anyevent.schmorp.de>, for more info.
		53
		54	=head1 WHY YOU SHOULD USE THIS MODULE (OR NOT)
		55
		56	Glib, POE, IO::Async, Event... CPAN offers event models by the dozen
		57	nowadays. So what is different about AnyEvent?
		58
		59	Executive Summary: AnyEvent is I<compatible>, AnyEvent is I<free of
		60	policy> and AnyEvent is I<small and efficient>.
		61
		62	First and foremost, I<AnyEvent is not an event model> itself, it only
		63	interfaces to whatever event model the main program happens to use, in a
		64	pragmatic way. For event models and certain classes of immortals alike,
		65	the statement "there can only be one" is a bitter reality: In general,
		66	only one event loop can be active at the same time in a process. AnyEvent
		67	cannot change this, but it can hide the differences between those event
		68	loops.
		69
		70	The goal of AnyEvent is to offer module authors the ability to do event
		71	programming (waiting for I/O or timer events) without subscribing to a
		72	religion, a way of living, and most importantly: without forcing your
		73	module users into the same thing by forcing them to use the same event
		74	model you use.
		75
		76	For modules like POE or IO::Async (which is a total misnomer as it is
		77	actually doing all I/O I<synchronously>...), using them in your module is
		78	like joining a cult: After you joined, you are dependent on them and you
		79	cannot use anything else, as they are simply incompatible to everything
		80	that isn't them. What's worse, all the potential users of your
		81	module are I<also> forced to use the same event loop you use.
		82
		83	AnyEvent is different: AnyEvent + POE works fine. AnyEvent + Glib works
		84	fine. AnyEvent + Tk works fine etc. etc. but none of these work together
		85	with the rest: POE + IO::Async? No go. Tk + Event? No go. Again: if
		86	your module uses one of those, every user of your module has to use it,
		87	too. But if your module uses AnyEvent, it works transparently with all
		88	event models it supports (including stuff like IO::Async, as long as those
		89	use one of the supported event loops. It is trivial to add new event loops
		90	to AnyEvent, too, so it is future-proof).
		91
		92	In addition to being free of having to use I<the one and only true event
		93	model>, AnyEvent also is free of bloat and policy: with POE or similar
		94	modules, you get an enormous amount of code and strict rules you have to
		95	follow. AnyEvent, on the other hand, is lean and up to the point, by only
		96	offering the functionality that is necessary, in as thin as a wrapper as
		97	technically possible.
		98
		99	Of course, AnyEvent comes with a big (and fully optional!) toolbox
		100	of useful functionality, such as an asynchronous DNS resolver, 100%
		101	non-blocking connects (even with TLS/SSL, IPv6 and on broken platforms
		102	such as Windows) and lots of real-world knowledge and workarounds for
		103	platform bugs and differences.
		104
		105	Now, if you I<do want> lots of policy (this can arguably be somewhat
		106	useful) and you want to force your users to use the one and only event
		107	model, you should I<not> use this module.
22		108
23	=head1 DESCRIPTION	109	=head1 DESCRIPTION
24		110
25	L<AnyEvent> provides an identical interface to multiple event loops. This	111	L<AnyEvent> provides an identical interface to multiple event loops. This
26	allows module authors to utilise an event loop without forcing module	112	allows module authors to utilise an event loop without forcing module
27	users to use the same event loop (as only a single event loop can coexist	113	users to use the same event loop (as only a single event loop can coexist
28	peacefully at any one time).	114	peacefully at any one time).
29		115
30	The interface itself is vaguely similar but not identical to the Event	116	The interface itself is vaguely similar, but not identical to the L<Event>
31	module.	117	module.
32		118
33	On the first call of any method, the module tries to detect the currently	119	During the first call of any watcher-creation method, the module tries
34	loaded event loop by probing wether any of the following modules is	120	to detect the currently loaded event loop by probing whether one of the
35	loaded: L<Coro::Event>, L<Event>, L<Glib>, L<Tk>. The first one found is	121	following modules is already loaded: L<EV>,
36	used. If none is found, the module tries to load these modules in the	122	L<Event>, L<Glib>, L<AnyEvent::Impl::Perl>, L<Tk>, L<Event::Lib>, L<Qt>,
37	order given. The first one that could be successfully loaded will be	123	L<POE>. The first one found is used. If none are found, the module tries
38	used. If still none could be found, AnyEvent will fall back to a pure-perl	124	to load these modules (excluding Tk, Event::Lib, Qt and POE as the pure perl
39	event loop, which is also not very efficient.	125	adaptor should always succeed) in the order given. The first one that can
		126	be successfully loaded will be used. If, after this, still none could be
		127	found, AnyEvent will fall back to a pure-perl event loop, which is not
		128	very efficient, but should work everywhere.
40		129
41	Because AnyEvent first checks for modules that are already loaded, loading	130	Because AnyEvent first checks for modules that are already loaded, loading
42	an Event model explicitly before first using AnyEvent will likely make	131	an event model explicitly before first using AnyEvent will likely make
43	that model the default. For example:	132	that model the default. For example:
44		133
45	use Tk;	134	use Tk;
46	use AnyEvent;	135	use AnyEvent;
47		136
48	# .. AnyEvent will likely default to Tk	137	# .. AnyEvent will likely default to Tk
49		138
		139	The I<likely> means that, if any module loads another event model and
		140	starts using it, all bets are off. Maybe you should tell their authors to
		141	use AnyEvent so their modules work together with others seamlessly...
		142
50	The pure-perl implementation of AnyEvent is called	143	The pure-perl implementation of AnyEvent is called
51	C<AnyEvent::Impl::Perl>. Like other event modules you can load it	144	C<AnyEvent::Impl::Perl>. Like other event modules you can load it
52	explicitly.	145	explicitly and enjoy the high availability of that event loop :)
53		146
54	=head1 WATCHERS	147	=head1 WATCHERS
55		148
56	AnyEvent has the central concept of a I<watcher>, which is an object that	149	AnyEvent has the central concept of a I<watcher>, which is an object that
57	stores relevant data for each kind of event you are waiting for, such as	150	stores relevant data for each kind of event you are waiting for, such as
58	the callback to call, the filehandle to watch, etc.	151	the callback to call, the file handle to watch, etc.
59		152
60	These watchers are normal Perl objects with normal Perl lifetime. After	153	These watchers are normal Perl objects with normal Perl lifetime. After
61	creating a watcher it will immediately "watch" for events and invoke	154	creating a watcher it will immediately "watch" for events and invoke the
		155	callback when the event occurs (of course, only when the event model
		156	is in control).
		157
		158	Note that B<callbacks must not permanently change global variables>
		159	potentially in use by the event loop (such as C<$_> or C<$[>) and that B<<
		160	callbacks must not C<die> >>. The former is good programming practise in
		161	Perl and the latter stems from the fact that exception handling differs
		162	widely between event loops.
		163
62	the callback. To disable the watcher you have to destroy it (e.g. by	164	To disable the watcher you have to destroy it (e.g. by setting the
63	setting the variable that stores it to C<undef> or otherwise deleting all	165	variable you store it in to C<undef> or otherwise deleting all references
64	references to it).	166	to it).
65		167
66	All watchers are created by calling a method on the C<AnyEvent> class.	168	All watchers are created by calling a method on the C<AnyEvent> class.
67		169
		170	Many watchers either are used with "recursion" (repeating timers for
		171	example), or need to refer to their watcher object in other ways.
		172
		173	An any way to achieve that is this pattern:
		174
		175	my $w; $w = AnyEvent->type (arg => value ..., cb => sub {
		176	# you can use $w here, for example to undef it
		177	undef $w;
		178	});
		179
		180	Note that C<my $w; $w => combination. This is necessary because in Perl,
		181	my variables are only visible after the statement in which they are
		182	declared.
		183
68	=head2 IO WATCHERS	184	=head2 I/O WATCHERS
69		185
70	You can create I/O watcher by calling the C<< AnyEvent->io >> method with	186	You can create an I/O watcher by calling the C<< AnyEvent->io >> method
71	the following mandatory arguments:	187	with the following mandatory key-value pairs as arguments:
72		188
73	C<fh> the Perl I<filehandle> (not filedescriptor) to watch for	189	C<fh> is the Perl I<file handle> (or a naked file descriptor) to watch
		190	for events (AnyEvent might or might not keep a reference to this file
		191	handle). Note that only file handles pointing to things for which
		192	non-blocking operation makes sense are allowed. This includes sockets,
		193	most character devices, pipes, fifos and so on, but not for example files
		194	or block devices.
		195
74	events. C<poll> must be a string that is either C<r> or C<w>, that creates	196	C<poll> must be a string that is either C<r> or C<w>, which creates a
75	a watcher waiting for "r"eadable or "w"ritable events. C<cb> teh callback	197	watcher waiting for "r"eadable or "w"ritable events, respectively.
76	to invoke everytime the filehandle becomes ready.
77		198
78	Only one io watcher per C<fh> and C<poll> combination is allowed (i.e. on	199	C<cb> is the callback to invoke each time the file handle becomes ready.
79	a socket you can have one r + one w, not any more (limitation comes from
80	Tk - if you are sure you are not using Tk this limitation is gone).
81		200
82	Filehandles will be kept alive, so as long as the watcher exists, the	201	Although the callback might get passed parameters, their value and
83	filehandle exists, too.	202	presence is undefined and you cannot rely on them. Portable AnyEvent
		203	callbacks cannot use arguments passed to I/O watcher callbacks.
84		204
85	Example:	205	The I/O watcher might use the underlying file descriptor or a copy of it.
		206	You must not close a file handle as long as any watcher is active on the
		207	underlying file descriptor.
86		208
		209	Some event loops issue spurious readyness notifications, so you should
		210	always use non-blocking calls when reading/writing from/to your file
		211	handles.
		212
87	# wait for readability of STDIN, then read a line and disable the watcher	213	Example: wait for readability of STDIN, then read a line and disable the
		214	watcher.
		215
88	my $w; $w = AnyEvent->io (fh => \*STDIN, poll => 'r', cb => sub {	216	my $w; $w = AnyEvent->io (fh => \*STDIN, poll => 'r', cb => sub {
89	chomp (my $input = <STDIN>);	217	chomp (my $input = <STDIN>);
90	warn "read: $input\n";	218	warn "read: $input\n";
91	undef $w;	219	undef $w;
92	});	220	});
…		…
94	=head2 TIME WATCHERS	222	=head2 TIME WATCHERS
95		223
96	You can create a time watcher by calling the C<< AnyEvent->timer >>	224	You can create a time watcher by calling the C<< AnyEvent->timer >>
97	method with the following mandatory arguments:	225	method with the following mandatory arguments:
98		226
99	C<after> after how many seconds (fractions are supported) should the timer	227	C<after> specifies after how many seconds (fractional values are
100	activate. C<cb> the callback to invoke.	228	supported) the callback should be invoked. C<cb> is the callback to invoke
		229	in that case.
101		230
102	The timer callback will be invoked at most once: if you want a repeating	231	Although the callback might get passed parameters, their value and
103	timer you have to create a new watcher (this is a limitation by both Tk	232	presence is undefined and you cannot rely on them. Portable AnyEvent
104	and Glib).	233	callbacks cannot use arguments passed to time watcher callbacks.
105		234
106	Example:	235	The callback will normally be invoked once only. If you specify another
		236	parameter, C<interval>, as a strictly positive number (> 0), then the
		237	callback will be invoked regularly at that interval (in fractional
		238	seconds) after the first invocation. If C<interval> is specified with a
		239	false value, then it is treated as if it were missing.
107		240
		241	The callback will be rescheduled before invoking the callback, but no
		242	attempt is done to avoid timer drift in most backends, so the interval is
		243	only approximate.
		244
108	# fire an event after 7.7 seconds	245	Example: fire an event after 7.7 seconds.
		246
109	my $w = AnyEvent->timer (after => 7.7, cb => sub {	247	my $w = AnyEvent->timer (after => 7.7, cb => sub {
110	warn "timeout\n";	248	warn "timeout\n";
111	});	249	});
112		250
113	# to cancel the timer:	251	# to cancel the timer:
114	undef $w	252	undef $w;
115		253
116	=head2 CONDITION WATCHERS	254	Example 2: fire an event after 0.5 seconds, then roughly every second.
117		255
118	Condition watchers can be created by calling the C<< AnyEvent->condvar >>	256	my $w = AnyEvent->timer (after => 0.5, interval => 1, cb => sub {
119	method without any arguments.	257	warn "timeout\n";
		258	};
120		259
121	A condition watcher watches for a condition - precisely that the C<<	260	=head3 TIMING ISSUES
122	->broadcast >> method has been called.
123		261
124	The watcher has only two methods:	262	There are two ways to handle timers: based on real time (relative, "fire
		263	in 10 seconds") and based on wallclock time (absolute, "fire at 12
		264	o'clock").
		265
		266	While most event loops expect timers to specified in a relative way, they
		267	use absolute time internally. This makes a difference when your clock
		268	"jumps", for example, when ntp decides to set your clock backwards from
		269	the wrong date of 2014-01-01 to 2008-01-01, a watcher that is supposed to
		270	fire "after" a second might actually take six years to finally fire.
		271
		272	AnyEvent cannot compensate for this. The only event loop that is conscious
		273	about these issues is L<EV>, which offers both relative (ev_timer, based
		274	on true relative time) and absolute (ev_periodic, based on wallclock time)
		275	timers.
		276
		277	AnyEvent always prefers relative timers, if available, matching the
		278	AnyEvent API.
		279
		280	AnyEvent has two additional methods that return the "current time":
125		281
126	=over 4	282	=over 4
127		283
128	=item $cv->wait	284	=item AnyEvent->time
129		285
130	Wait (blocking if necessary) until the C<< ->broadcast >> method has been	286	This returns the "current wallclock time" as a fractional number of
131	called on c<$cv>, while servicing other watchers normally.	287	seconds since the Epoch (the same thing as C<time> or C<Time::HiRes::time>
		288	return, and the result is guaranteed to be compatible with those).
132		289
133	Not all event models support a blocking wait - some die in that case, so	290	It progresses independently of any event loop processing, i.e. each call
134	if you are using this from a module, never require a blocking wait, but	291	will check the system clock, which usually gets updated frequently.
135	let the caller decide wether the call will block or not (for example,
136	by coupling condition variables with some kind of request results and
137	supporting callbacks so the caller knows that getting the result will not
138	block, while still suppporting blockign waits if the caller so desires).
139		292
140	You can only wait once on a condition - additional calls will return	293	=item AnyEvent->now
141	immediately.
142		294
143	=item $cv->broadcast	295	This also returns the "current wallclock time", but unlike C<time>, above,
		296	this value might change only once per event loop iteration, depending on
		297	the event loop (most return the same time as C<time>, above). This is the
		298	time that AnyEvent's timers get scheduled against.
144		299
145	Flag the condition as ready - a running C<< ->wait >> and all further	300	I<In almost all cases (in all cases if you don't care), this is the
146	calls to C<wait> will return after this method has been called. If nobody	301	function to call when you want to know the current time.>
147	is waiting the broadcast will be remembered..
148		302
149	Example:	303	This function is also often faster then C<< AnyEvent->time >>, and
		304	thus the preferred method if you want some timestamp (for example,
		305	L<AnyEvent::Handle> uses this to update it's activity timeouts).
		306
		307	The rest of this section is only of relevance if you try to be very exact
		308	with your timing, you can skip it without bad conscience.
		309
		310	For a practical example of when these times differ, consider L<Event::Lib>
		311	and L<EV> and the following set-up:
		312
		313	The event loop is running and has just invoked one of your callback at
		314	time=500 (assume no other callbacks delay processing). In your callback,
		315	you wait a second by executing C<sleep 1> (blocking the process for a
		316	second) and then (at time=501) you create a relative timer that fires
		317	after three seconds.
		318
		319	With L<Event::Lib>, C<< AnyEvent->time >> and C<< AnyEvent->now >> will
		320	both return C<501>, because that is the current time, and the timer will
		321	be scheduled to fire at time=504 (C<501> + C<3>).
		322
		323	With L<EV>, C<< AnyEvent->time >> returns C<501> (as that is the current
		324	time), but C<< AnyEvent->now >> returns C<500>, as that is the time the
		325	last event processing phase started. With L<EV>, your timer gets scheduled
		326	to run at time=503 (C<500> + C<3>).
		327
		328	In one sense, L<Event::Lib> is more exact, as it uses the current time
		329	regardless of any delays introduced by event processing. However, most
		330	callbacks do not expect large delays in processing, so this causes a
		331	higher drift (and a lot more system calls to get the current time).
		332
		333	In another sense, L<EV> is more exact, as your timer will be scheduled at
		334	the same time, regardless of how long event processing actually took.
		335
		336	In either case, if you care (and in most cases, you don't), then you
		337	can get whatever behaviour you want with any event loop, by taking the
		338	difference between C<< AnyEvent->time >> and C<< AnyEvent->now >> into
		339	account.
		340
		341	=item AnyEvent->now_update
		342
		343	Some event loops (such as L<EV> or L<AnyEvent::Impl::Perl>) cache
		344	the current time for each loop iteration (see the discussion of L<<
		345	AnyEvent->now >>, above).
		346
		347	When a callback runs for a long time (or when the process sleeps), then
		348	this "current" time will differ substantially from the real time, which
		349	might affect timers and time-outs.
		350
		351	When this is the case, you can call this method, which will update the
		352	event loop's idea of "current time".
		353
		354	Note that updating the time I<might> cause some events to be handled.
		355
		356	=back
		357
		358	=head2 SIGNAL WATCHERS
		359
		360	You can watch for signals using a signal watcher, C<signal> is the signal
		361	I<name> in uppercase and without any C<SIG> prefix, C<cb> is the Perl
		362	callback to be invoked whenever a signal occurs.
		363
		364	Although the callback might get passed parameters, their value and
		365	presence is undefined and you cannot rely on them. Portable AnyEvent
		366	callbacks cannot use arguments passed to signal watcher callbacks.
		367
		368	Multiple signal occurrences can be clumped together into one callback
		369	invocation, and callback invocation will be synchronous. Synchronous means
		370	that it might take a while until the signal gets handled by the process,
		371	but it is guaranteed not to interrupt any other callbacks.
		372
		373	The main advantage of using these watchers is that you can share a signal
		374	between multiple watchers, and AnyEvent will ensure that signals will not
		375	interrupt your program at bad times.
		376
		377	This watcher might use C<%SIG> (depending on the event loop used),
		378	so programs overwriting those signals directly will likely not work
		379	correctly.
		380
		381	Example: exit on SIGINT
		382
		383	my $w = AnyEvent->signal (signal => "INT", cb => sub { exit 1 });
		384
		385	=head3 Signal Races, Delays and Workarounds
		386
		387	Many event loops (e.g. Glib, Tk, Qt, IO::Async) do not support attaching
		388	callbacks to signals in a generic way, which is a pity, as you cannot do
		389	race-free signal handling in perl. AnyEvent will try to do it's best, but
		390	in some cases, signals will be delayed. The maximum time a signal might
		391	be delayed is specified in C<$AnyEvent::MAX_SIGNAL_LATENCY> (default: 10
		392	seconds). This variable can be changed only before the first signal
		393	watcher is created, and should be left alone otherwise. Higher values
		394	will cause fewer spurious wake-ups, which is better for power and CPU
		395	saving. All these problems can be avoided by installing the optional
		396	L<Async::Interrupt> module. This will not work with inherently broken
		397	event loops such as L<Event> or L<Event::Lib> (and not with L<POE>
		398	currently, as POE does it's own workaround with one-second latency). With
		399	those, you just have to suffer the delays.
		400
		401	=head2 CHILD PROCESS WATCHERS
		402
		403	You can also watch on a child process exit and catch its exit status.
		404
		405	The child process is specified by the C<pid> argument (if set to C<0>, it
		406	watches for any child process exit). The watcher will triggered only when
		407	the child process has finished and an exit status is available, not on
		408	any trace events (stopped/continued).
		409
		410	The callback will be called with the pid and exit status (as returned by
		411	waitpid), so unlike other watcher types, you I<can> rely on child watcher
		412	callback arguments.
		413
		414	This watcher type works by installing a signal handler for C<SIGCHLD>,
		415	and since it cannot be shared, nothing else should use SIGCHLD or reap
		416	random child processes (waiting for specific child processes, e.g. inside
		417	C<system>, is just fine).
		418
		419	There is a slight catch to child watchers, however: you usually start them
		420	I<after> the child process was created, and this means the process could
		421	have exited already (and no SIGCHLD will be sent anymore).
		422
		423	Not all event models handle this correctly (neither POE nor IO::Async do,
		424	see their AnyEvent::Impl manpages for details), but even for event models
		425	that I<do> handle this correctly, they usually need to be loaded before
		426	the process exits (i.e. before you fork in the first place). AnyEvent's
		427	pure perl event loop handles all cases correctly regardless of when you
		428	start the watcher.
		429
		430	This means you cannot create a child watcher as the very first
		431	thing in an AnyEvent program, you I<have> to create at least one
		432	watcher before you C<fork> the child (alternatively, you can call
		433	C<AnyEvent::detect>).
		434
		435	As most event loops do not support waiting for child events, they will be
		436	emulated by AnyEvent in most cases, in which the latency and race problems
		437	mentioned in the description of signal watchers apply.
		438
		439	Example: fork a process and wait for it
		440
		441	my $done = AnyEvent->condvar;
		442
		443	my $pid = fork or exit 5;
		444
		445	my $w = AnyEvent->child (
		446	pid => $pid,
		447	cb => sub {
		448	my ($pid, $status) = @_;
		449	warn "pid $pid exited with status $status";
		450	$done->send;
		451	},
		452	);
		453
		454	# do something else, then wait for process exit
		455	$done->recv;
		456
		457	=head2 IDLE WATCHERS
		458
		459	Sometimes there is a need to do something, but it is not so important
		460	to do it instantly, but only when there is nothing better to do. This
		461	"nothing better to do" is usually defined to be "no other events need
		462	attention by the event loop".
		463
		464	Idle watchers ideally get invoked when the event loop has nothing
		465	better to do, just before it would block the process to wait for new
		466	events. Instead of blocking, the idle watcher is invoked.
		467
		468	Most event loops unfortunately do not really support idle watchers (only
		469	EV, Event and Glib do it in a usable fashion) - for the rest, AnyEvent
		470	will simply call the callback "from time to time".
		471
		472	Example: read lines from STDIN, but only process them when the
		473	program is otherwise idle:
		474
		475	my @lines; # read data
		476	my $idle_w;
		477	my $io_w = AnyEvent->io (fh => \*STDIN, poll => 'r', cb => sub {
		478	push @lines, scalar <STDIN>;
		479
		480	# start an idle watcher, if not already done
		481	$idle_w ||= AnyEvent->idle (cb => sub {
		482	# handle only one line, when there are lines left
		483	if (my $line = shift @lines) {
		484	print "handled when idle: $line";
		485	} else {
		486	# otherwise disable the idle watcher again
		487	undef $idle_w;
		488	}
		489	});
		490	});
		491
		492	=head2 CONDITION VARIABLES
		493
		494	If you are familiar with some event loops you will know that all of them
		495	require you to run some blocking "loop", "run" or similar function that
		496	will actively watch for new events and call your callbacks.
		497
		498	AnyEvent is slightly different: it expects somebody else to run the event
		499	loop and will only block when necessary (usually when told by the user).
		500
		501	The instrument to do that is called a "condition variable", so called
		502	because they represent a condition that must become true.
		503
		504	Now is probably a good time to look at the examples further below.
		505
		506	Condition variables can be created by calling the C<< AnyEvent->condvar
		507	>> method, usually without arguments. The only argument pair allowed is
		508	C<cb>, which specifies a callback to be called when the condition variable
		509	becomes true, with the condition variable as the first argument (but not
		510	the results).
		511
		512	After creation, the condition variable is "false" until it becomes "true"
		513	by calling the C<send> method (or calling the condition variable as if it
		514	were a callback, read about the caveats in the description for the C<<
		515	->send >> method).
		516
		517	Condition variables are similar to callbacks, except that you can
		518	optionally wait for them. They can also be called merge points - points
		519	in time where multiple outstanding events have been processed. And yet
		520	another way to call them is transactions - each condition variable can be
		521	used to represent a transaction, which finishes at some point and delivers
		522	a result.
		523
		524	Condition variables are very useful to signal that something has finished,
		525	for example, if you write a module that does asynchronous http requests,
		526	then a condition variable would be the ideal candidate to signal the
		527	availability of results. The user can either act when the callback is
		528	called or can synchronously C<< ->recv >> for the results.
		529
		530	You can also use them to simulate traditional event loops - for example,
		531	you can block your main program until an event occurs - for example, you
		532	could C<< ->recv >> in your main program until the user clicks the Quit
		533	button of your app, which would C<< ->send >> the "quit" event.
		534
		535	Note that condition variables recurse into the event loop - if you have
		536	two pieces of code that call C<< ->recv >> in a round-robin fashion, you
		537	lose. Therefore, condition variables are good to export to your caller, but
		538	you should avoid making a blocking wait yourself, at least in callbacks,
		539	as this asks for trouble.
		540
		541	Condition variables are represented by hash refs in perl, and the keys
		542	used by AnyEvent itself are all named C<_ae_XXX> to make subclassing
		543	easy (it is often useful to build your own transaction class on top of
		544	AnyEvent). To subclass, use C<AnyEvent::CondVar> as base class and call
		545	it's C<new> method in your own C<new> method.
		546
		547	There are two "sides" to a condition variable - the "producer side" which
		548	eventually calls C<< -> send >>, and the "consumer side", which waits
		549	for the send to occur.
		550
		551	Example: wait for a timer.
150		552
151	# wait till the result is ready	553	# wait till the result is ready
152	my $result_ready = AnyEvent->condvar;	554	my $result_ready = AnyEvent->condvar;
153		555
154	# do something such as adding a timer	556	# do something such as adding a timer
155	# or socket watcher the calls $result_ready->broadcast	557	# or socket watcher the calls $result_ready->send
156	# when the "result" is ready.	558	# when the "result" is ready.
		559	# in this case, we simply use a timer:
		560	my $w = AnyEvent->timer (
		561	after => 1,
		562	cb => sub { $result_ready->send },
		563	);
157		564
		565	# this "blocks" (while handling events) till the callback
		566	# calls -<send
158	$result_ready->wait;	567	$result_ready->recv;
		568
		569	Example: wait for a timer, but take advantage of the fact that condition
		570	variables are also callable directly.
		571
		572	my $done = AnyEvent->condvar;
		573	my $delay = AnyEvent->timer (after => 5, cb => $done);
		574	$done->recv;
		575
		576	Example: Imagine an API that returns a condvar and doesn't support
		577	callbacks. This is how you make a synchronous call, for example from
		578	the main program:
		579
		580	use AnyEvent::CouchDB;
		581
		582	...
		583
		584	my @info = $couchdb->info->recv;
		585
		586	And this is how you would just set a callback to be called whenever the
		587	results are available:
		588
		589	$couchdb->info->cb (sub {
		590	my @info = $_[0]->recv;
		591	});
		592
		593	=head3 METHODS FOR PRODUCERS
		594
		595	These methods should only be used by the producing side, i.e. the
		596	code/module that eventually sends the signal. Note that it is also
		597	the producer side which creates the condvar in most cases, but it isn't
		598	uncommon for the consumer to create it as well.
		599
		600	=over 4
		601
		602	=item $cv->send (...)
		603
		604	Flag the condition as ready - a running C<< ->recv >> and all further
		605	calls to C<recv> will (eventually) return after this method has been
		606	called. If nobody is waiting the send will be remembered.
		607
		608	If a callback has been set on the condition variable, it is called
		609	immediately from within send.
		610
		611	Any arguments passed to the C<send> call will be returned by all
		612	future C<< ->recv >> calls.
		613
		614	Condition variables are overloaded so one can call them directly (as if
		615	they were a code reference). Calling them directly is the same as calling
		616	C<send>.
		617
		618	=item $cv->croak ($error)
		619
		620	Similar to send, but causes all call's to C<< ->recv >> to invoke
		621	C<Carp::croak> with the given error message/object/scalar.
		622
		623	This can be used to signal any errors to the condition variable
		624	user/consumer. Doing it this way instead of calling C<croak> directly
		625	delays the error detetcion, but has the overwhelmign advantage that it
		626	diagnoses the error at the place where the result is expected, and not
		627	deep in some event clalback without connection to the actual code causing
		628	the problem.
		629
		630	=item $cv->begin ([group callback])
		631
		632	=item $cv->end
		633
		634	These two methods can be used to combine many transactions/events into
		635	one. For example, a function that pings many hosts in parallel might want
		636	to use a condition variable for the whole process.
		637
		638	Every call to C<< ->begin >> will increment a counter, and every call to
		639	C<< ->end >> will decrement it. If the counter reaches C<0> in C<< ->end
		640	>>, the (last) callback passed to C<begin> will be executed. That callback
		641	is I<supposed> to call C<< ->send >>, but that is not required. If no
		642	callback was set, C<send> will be called without any arguments.
		643
		644	You can think of C<< $cv->send >> giving you an OR condition (one call
		645	sends), while C<< $cv->begin >> and C<< $cv->end >> giving you an AND
		646	condition (all C<begin> calls must be C<end>'ed before the condvar sends).
		647
		648	Let's start with a simple example: you have two I/O watchers (for example,
		649	STDOUT and STDERR for a program), and you want to wait for both streams to
		650	close before activating a condvar:
		651
		652	my $cv = AnyEvent->condvar;
		653
		654	$cv->begin; # first watcher
		655	my $w1 = AnyEvent->io (fh => $fh1, cb => sub {
		656	defined sysread $fh1, my $buf, 4096
		657	or $cv->end;
		658	});
		659
		660	$cv->begin; # second watcher
		661	my $w2 = AnyEvent->io (fh => $fh2, cb => sub {
		662	defined sysread $fh2, my $buf, 4096
		663	or $cv->end;
		664	});
		665
		666	$cv->recv;
		667
		668	This works because for every event source (EOF on file handle), there is
		669	one call to C<begin>, so the condvar waits for all calls to C<end> before
		670	sending.
		671
		672	The ping example mentioned above is slightly more complicated, as the
		673	there are results to be passwd back, and the number of tasks that are
		674	begung can potentially be zero:
		675
		676	my $cv = AnyEvent->condvar;
		677
		678	my %result;
		679	$cv->begin (sub { $cv->send (\%result) });
		680
		681	for my $host (@list_of_hosts) {
		682	$cv->begin;
		683	ping_host_then_call_callback $host, sub {
		684	$result{$host} = ...;
		685	$cv->end;
		686	};
		687	}
		688
		689	$cv->end;
		690
		691	This code fragment supposedly pings a number of hosts and calls
		692	C<send> after results for all then have have been gathered - in any
		693	order. To achieve this, the code issues a call to C<begin> when it starts
		694	each ping request and calls C<end> when it has received some result for
		695	it. Since C<begin> and C<end> only maintain a counter, the order in which
		696	results arrive is not relevant.
		697
		698	There is an additional bracketing call to C<begin> and C<end> outside the
		699	loop, which serves two important purposes: first, it sets the callback
		700	to be called once the counter reaches C<0>, and second, it ensures that
		701	C<send> is called even when C<no> hosts are being pinged (the loop
		702	doesn't execute once).
		703
		704	This is the general pattern when you "fan out" into multiple (but
		705	potentially none) subrequests: use an outer C<begin>/C<end> pair to set
		706	the callback and ensure C<end> is called at least once, and then, for each
		707	subrequest you start, call C<begin> and for each subrequest you finish,
		708	call C<end>.
159		709
160	=back	710	=back
161		711
162	=head2 SIGNAL WATCHERS	712	=head3 METHODS FOR CONSUMERS
163		713
164	You can listen for signals using a signal watcher, C<signal> is the signal	714	These methods should only be used by the consuming side, i.e. the
165	I<name> without any C<SIG> prefix. Multiple signals events can be clumped	715	code awaits the condition.
166	together into one callback invocation, and callback invocation might or
167	might not be asynchronous.
168
169	These watchers might use C<%SIG>, so programs overwriting those signals
170	directly will likely not work correctly.
171
172	Example: exit on SIGINT
173
174	my $w = AnyEvent->signal (signal => "INT", cb => sub { exit 1 });
175
176	=head2 CHILD PROCESS WATCHERS
177
178	You can also listen for the status of a child process specified by the
179	C<pid> argument. The watcher will only trigger once. This works by
180	installing a signal handler for C<SIGCHLD>.
181
182	Example: wait for pid 1333
183
184	my $w = AnyEvent->child (pid => 1333, cb => sub { warn "exit status $?" });
185
186	=head1 GLOBALS
187		716
188	=over 4	717	=over 4
189		718
		719	=item $cv->recv
		720
		721	Wait (blocking if necessary) until the C<< ->send >> or C<< ->croak
		722	>> methods have been called on c<$cv>, while servicing other watchers
		723	normally.
		724
		725	You can only wait once on a condition - additional calls are valid but
		726	will return immediately.
		727
		728	If an error condition has been set by calling C<< ->croak >>, then this
		729	function will call C<croak>.
		730
		731	In list context, all parameters passed to C<send> will be returned,
		732	in scalar context only the first one will be returned.
		733
		734	Note that doing a blocking wait in a callback is not supported by any
		735	event loop, that is, recursive invocation of a blocking C<< ->recv
		736	>> is not allowed, and the C<recv> call will C<croak> if such a
		737	condition is detected. This condition can be slightly loosened by using
		738	L<Coro::AnyEvent>, which allows you to do a blocking C<< ->recv >> from
		739	any thread that doesn't run the event loop itself.
		740
		741	Not all event models support a blocking wait - some die in that case
		742	(programs might want to do that to stay interactive), so I<if you are
		743	using this from a module, never require a blocking wait>. Instead, let the
		744	caller decide whether the call will block or not (for example, by coupling
		745	condition variables with some kind of request results and supporting
		746	callbacks so the caller knows that getting the result will not block,
		747	while still supporting blocking waits if the caller so desires).
		748
		749	You can ensure that C<< -recv >> never blocks by setting a callback and
		750	only calling C<< ->recv >> from within that callback (or at a later
		751	time). This will work even when the event loop does not support blocking
		752	waits otherwise.
		753
		754	=item $bool = $cv->ready
		755
		756	Returns true when the condition is "true", i.e. whether C<send> or
		757	C<croak> have been called.
		758
		759	=item $cb = $cv->cb ($cb->($cv))
		760
		761	This is a mutator function that returns the callback set and optionally
		762	replaces it before doing so.
		763
		764	The callback will be called when the condition becomes "true", i.e. when
		765	C<send> or C<croak> are called, with the only argument being the condition
		766	variable itself. Calling C<recv> inside the callback or at any later time
		767	is guaranteed not to block.
		768
		769	=back
		770
		771	=head1 SUPPORTED EVENT LOOPS/BACKENDS
		772
		773	The available backend classes are (every class has its own manpage):
		774
		775	=over 4
		776
		777	=item Backends that are autoprobed when no other event loop can be found.
		778
		779	EV is the preferred backend when no other event loop seems to be in
		780	use. If EV is not installed, then AnyEvent will try Event, and, failing
		781	that, will fall back to its own pure-perl implementation, which is
		782	available everywhere as it comes with AnyEvent itself.
		783
		784	AnyEvent::Impl::EV based on EV (interface to libev, best choice).
		785	AnyEvent::Impl::Event based on Event, very stable, few glitches.
		786	AnyEvent::Impl::Perl pure-perl implementation, fast and portable.
		787
		788	=item Backends that are transparently being picked up when they are used.
		789
		790	These will be used when they are currently loaded when the first watcher
		791	is created, in which case it is assumed that the application is using
		792	them. This means that AnyEvent will automatically pick the right backend
		793	when the main program loads an event module before anything starts to
		794	create watchers. Nothing special needs to be done by the main program.
		795
		796	AnyEvent::Impl::Glib based on Glib, slow but very stable.
		797	AnyEvent::Impl::Tk based on Tk, very broken.
		798	AnyEvent::Impl::EventLib based on Event::Lib, leaks memory and worse.
		799	AnyEvent::Impl::POE based on POE, very slow, some limitations.
		800
		801	=item Backends with special needs.
		802
		803	Qt requires the Qt::Application to be instantiated first, but will
		804	otherwise be picked up automatically. As long as the main program
		805	instantiates the application before any AnyEvent watchers are created,
		806	everything should just work.
		807
		808	AnyEvent::Impl::Qt based on Qt.
		809
		810	Support for IO::Async can only be partial, as it is too broken and
		811	architecturally limited to even support the AnyEvent API. It also
		812	is the only event loop that needs the loop to be set explicitly, so
		813	it can only be used by a main program knowing about AnyEvent. See
		814	L<AnyEvent::Impl::Async> for the gory details.
		815
		816	AnyEvent::Impl::IOAsync based on IO::Async, cannot be autoprobed.
		817
		818	=item Event loops that are indirectly supported via other backends.
		819
		820	Some event loops can be supported via other modules:
		821
		822	There is no direct support for WxWidgets (L<Wx>) or L<Prima>.
		823
		824	B<WxWidgets> has no support for watching file handles. However, you can
		825	use WxWidgets through the POE adaptor, as POE has a Wx backend that simply
		826	polls 20 times per second, which was considered to be too horrible to even
		827	consider for AnyEvent.
		828
		829	B<Prima> is not supported as nobody seems to be using it, but it has a POE
		830	backend, so it can be supported through POE.
		831
		832	AnyEvent knows about both L<Prima> and L<Wx>, however, and will try to
		833	load L<POE> when detecting them, in the hope that POE will pick them up,
		834	in which case everything will be automatic.
		835
		836	=back
		837
		838	=head1 GLOBAL VARIABLES AND FUNCTIONS
		839
		840	These are not normally required to use AnyEvent, but can be useful to
		841	write AnyEvent extension modules.
		842
		843	=over 4
		844
190	=item $AnyEvent::MODEL	845	=item $AnyEvent::MODEL
191		846
192	Contains C<undef> until the first watcher is being created. Then it	847	Contains C<undef> until the first watcher is being created, before the
		848	backend has been autodetected.
		849
193	contains the event model that is being used, which is the name of the	850	Afterwards it contains the event model that is being used, which is the
194	Perl class implementing the model. This class is usually one of the	851	name of the Perl class implementing the model. This class is usually one
195	C<AnyEvent::Impl:xxx> modules, but can be any other class in the case	852	of the C<AnyEvent::Impl:xxx> modules, but can be any other class in the
196	AnyEvent has been extended at runtime (e.g. in I<rxvt-unicode>).	853	case AnyEvent has been extended at runtime (e.g. in I<rxvt-unicode> it
197		854	will be C<urxvt::anyevent>).
198	The known classes so far are:
199
200	AnyEvent::Impl::Coro based on Coro::Event, best choice.
201	EV::AnyEvent based on EV (an interface to libevent)
202	AnyEvent::Impl::Event based on Event, also best choice :)
203	AnyEvent::Impl::Glib based on Glib, second-best choice.
204	AnyEvent::Impl::Tk based on Tk, very bad choice.
205	AnyEvent::Impl::Perl pure-perl implementation, inefficient.
206		855
207	=item AnyEvent::detect	856	=item AnyEvent::detect
208		857
209	Returns C<$AnyEvent::MODEL>, forcing autodetection of the event model if	858	Returns C<$AnyEvent::MODEL>, forcing autodetection of the event model
210	necessary. You should only call this function right before you would have	859	if necessary. You should only call this function right before you would
211	created an AnyEvent watcher anyway, that is, very late at runtime.	860	have created an AnyEvent watcher anyway, that is, as late as possible at
		861	runtime, and not e.g. while initialising of your module.
		862
		863	If you need to do some initialisation before AnyEvent watchers are
		864	created, use C<post_detect>.
		865
		866	=item $guard = AnyEvent::post_detect { BLOCK }
		867
		868	Arranges for the code block to be executed as soon as the event model is
		869	autodetected (or immediately if this has already happened).
		870
		871	The block will be executed I<after> the actual backend has been detected
		872	(C<$AnyEvent::MODEL> is set), but I<before> any watchers have been
		873	created, so it is possible to e.g. patch C<@AnyEvent::ISA> or do
		874	other initialisations - see the sources of L<AnyEvent::Strict> or
		875	L<AnyEvent::AIO> to see how this is used.
		876
		877	The most common usage is to create some global watchers, without forcing
		878	event module detection too early, for example, L<AnyEvent::AIO> creates
		879	and installs the global L<IO::AIO> watcher in a C<post_detect> block to
		880	avoid autodetecting the event module at load time.
		881
		882	If called in scalar or list context, then it creates and returns an object
		883	that automatically removes the callback again when it is destroyed. See
		884	L<Coro::BDB> for a case where this is useful.
		885
		886	=item @AnyEvent::post_detect
		887
		888	If there are any code references in this array (you can C<push> to it
		889	before or after loading AnyEvent), then they will called directly after
		890	the event loop has been chosen.
		891
		892	You should check C<$AnyEvent::MODEL> before adding to this array, though:
		893	if it is defined then the event loop has already been detected, and the
		894	array will be ignored.
		895
		896	Best use C<AnyEvent::post_detect { BLOCK }> when your application allows
		897	it,as it takes care of these details.
		898
		899	This variable is mainly useful for modules that can do something useful
		900	when AnyEvent is used and thus want to know when it is initialised, but do
		901	not need to even load it by default. This array provides the means to hook
		902	into AnyEvent passively, without loading it.
212		903
213	=back	904	=back
214		905
215	=head1 WHAT TO DO IN A MODULE	906	=head1 WHAT TO DO IN A MODULE
216		907
217	As a module author, you should "use AnyEvent" and call AnyEvent methods	908	As a module author, you should C<use AnyEvent> and call AnyEvent methods
218	freely, but you should not load a specific event module or rely on it.	909	freely, but you should not load a specific event module or rely on it.
219		910
220	Be careful when you create watchers in the module body - Anyevent will	911	Be careful when you create watchers in the module body - AnyEvent will
221	decide which event module to use as soon as the first method is called, so	912	decide which event module to use as soon as the first method is called, so
222	by calling AnyEvent in your module body you force the user of your module	913	by calling AnyEvent in your module body you force the user of your module
223	to load the event module first.	914	to load the event module first.
224		915
		916	Never call C<< ->recv >> on a condition variable unless you I<know> that
		917	the C<< ->send >> method has been called on it already. This is
		918	because it will stall the whole program, and the whole point of using
		919	events is to stay interactive.
		920
		921	It is fine, however, to call C<< ->recv >> when the user of your module
		922	requests it (i.e. if you create a http request object ad have a method
		923	called C<results> that returns the results, it should call C<< ->recv >>
		924	freely, as the user of your module knows what she is doing. always).
		925
225	=head1 WHAT TO DO IN THE MAIN PROGRAM	926	=head1 WHAT TO DO IN THE MAIN PROGRAM
226		927
227	There will always be a single main program - the only place that should	928	There will always be a single main program - the only place that should
228	dictate which event model to use.	929	dictate which event model to use.
229		930
230	If it doesn't care, it can just "use AnyEvent" and use it itself, or not	931	If it doesn't care, it can just "use AnyEvent" and use it itself, or not
231	do anything special and let AnyEvent decide which implementation to chose.	932	do anything special (it does not need to be event-based) and let AnyEvent
		933	decide which implementation to chose if some module relies on it.
232		934
233	If the main program relies on a specific event model (for example, in Gtk2	935	If the main program relies on a specific event model - for example, in
234	programs you have to rely on either Glib or Glib::Event), you should load	936	Gtk2 programs you have to rely on the Glib module - you should load the
235	it before loading AnyEvent or any module that uses it, generally, as early	937	event module before loading AnyEvent or any module that uses it: generally
236	as possible. The reason is that modules might create watchers when they	938	speaking, you should load it as early as possible. The reason is that
237	are loaded, and AnyEvent will decide on the event model to use as soon as	939	modules might create watchers when they are loaded, and AnyEvent will
238	it creates watchers, and it might chose the wrong one unless you load the	940	decide on the event model to use as soon as it creates watchers, and it
239	correct one yourself.	941	might chose the wrong one unless you load the correct one yourself.
240		942
241	You can chose to use a rather inefficient pure-perl implementation by	943	You can chose to use a pure-perl implementation by loading the
242	loading the C<AnyEvent::Impl::Perl> module, but letting AnyEvent chose is	944	C<AnyEvent::Impl::Perl> module, which gives you similar behaviour
243	generally better.	945	everywhere, but letting AnyEvent chose the model is generally better.
		946
		947	=head2 MAINLOOP EMULATION
		948
		949	Sometimes (often for short test scripts, or even standalone programs who
		950	only want to use AnyEvent), you do not want to run a specific event loop.
		951
		952	In that case, you can use a condition variable like this:
		953
		954	AnyEvent->condvar->recv;
		955
		956	This has the effect of entering the event loop and looping forever.
		957
		958	Note that usually your program has some exit condition, in which case
		959	it is better to use the "traditional" approach of storing a condition
		960	variable somewhere, waiting for it, and sending it when the program should
		961	exit cleanly.
		962
		963
		964	=head1 OTHER MODULES
		965
		966	The following is a non-exhaustive list of additional modules that use
		967	AnyEvent as a client and can therefore be mixed easily with other AnyEvent
		968	modules and other event loops in the same program. Some of the modules
		969	come with AnyEvent, most are available via CPAN.
		970
		971	=over 4
		972
		973	=item L<AnyEvent::Util>
		974
		975	Contains various utility functions that replace often-used but blocking
		976	functions such as C<inet_aton> by event-/callback-based versions.
		977
		978	=item L<AnyEvent::Socket>
		979
		980	Provides various utility functions for (internet protocol) sockets,
		981	addresses and name resolution. Also functions to create non-blocking tcp
		982	connections or tcp servers, with IPv6 and SRV record support and more.
		983
		984	=item L<AnyEvent::Handle>
		985
		986	Provide read and write buffers, manages watchers for reads and writes,
		987	supports raw and formatted I/O, I/O queued and fully transparent and
		988	non-blocking SSL/TLS (via L<AnyEvent::TLS>.
		989
		990	=item L<AnyEvent::DNS>
		991
		992	Provides rich asynchronous DNS resolver capabilities.
		993
		994	=item L<AnyEvent::HTTP>
		995
		996	A simple-to-use HTTP library that is capable of making a lot of concurrent
		997	HTTP requests.
		998
		999	=item L<AnyEvent::HTTPD>
		1000
		1001	Provides a simple web application server framework.
		1002
		1003	=item L<AnyEvent::FastPing>
		1004
		1005	The fastest ping in the west.
		1006
		1007	=item L<AnyEvent::DBI>
		1008
		1009	Executes L<DBI> requests asynchronously in a proxy process.
		1010
		1011	=item L<AnyEvent::AIO>
		1012
		1013	Truly asynchronous I/O, should be in the toolbox of every event
		1014	programmer. AnyEvent::AIO transparently fuses L<IO::AIO> and AnyEvent
		1015	together.
		1016
		1017	=item L<AnyEvent::BDB>
		1018
		1019	Truly asynchronous Berkeley DB access. AnyEvent::BDB transparently fuses
		1020	L<BDB> and AnyEvent together.
		1021
		1022	=item L<AnyEvent::GPSD>
		1023
		1024	A non-blocking interface to gpsd, a daemon delivering GPS information.
		1025
		1026	=item L<AnyEvent::IRC>
		1027
		1028	AnyEvent based IRC client module family (replacing the older Net::IRC3).
		1029
		1030	=item L<AnyEvent::XMPP>
		1031
		1032	AnyEvent based XMPP (Jabber protocol) module family (replacing the older
		1033	Net::XMPP2>.
		1034
		1035	=item L<AnyEvent::IGS>
		1036
		1037	A non-blocking interface to the Internet Go Server protocol (used by
		1038	L<App::IGS>).
		1039
		1040	=item L<Net::FCP>
		1041
		1042	AnyEvent-based implementation of the Freenet Client Protocol, birthplace
		1043	of AnyEvent.
		1044
		1045	=item L<Event::ExecFlow>
		1046
		1047	High level API for event-based execution flow control.
		1048
		1049	=item L<Coro>
		1050
		1051	Has special support for AnyEvent via L<Coro::AnyEvent>.
		1052
		1053	=back
244		1054
245	=cut	1055	=cut
246		1056
247	package AnyEvent;	1057	package AnyEvent;
248		1058
		1059	# basically a tuned-down version of common::sense
		1060	sub common_sense {
249	no warnings;	1061	# no warnings
250	use strict;	1062	${^WARNING_BITS} ^= ${^WARNING_BITS};
		1063	# use strict vars subs
		1064	$^H |= 0x00000600;
		1065	}
251		1066
		1067	BEGIN { AnyEvent::common_sense }
		1068
252	use Carp;	1069	use Carp ();
253		1070
254	our $VERSION = '2.55';	1071	our $VERSION = 4.85;
255	our $MODEL;	1072	our $MODEL;
256		1073
257	our $AUTOLOAD;	1074	our $AUTOLOAD;
258	our @ISA;	1075	our @ISA;
259		1076
260	our $verbose = $ENV{PERL_ANYEVENT_VERBOSE}*1;
261
262	our @REGISTRY;	1077	our @REGISTRY;
263		1078
		1079	our $WIN32;
		1080
		1081	our $VERBOSE;
		1082
		1083	BEGIN {
		1084	eval "sub WIN32(){ " . (($^O =~ /mswin32/i)*1) ." }";
		1085	eval "sub TAINT(){ " . (${^TAINT}*1) . " }";
		1086
		1087	delete @ENV{grep /^PERL_ANYEVENT_/, keys %ENV}
		1088	if ${^TAINT};
		1089
		1090	$VERBOSE = $ENV{PERL_ANYEVENT_VERBOSE}*1;
		1091
		1092	}
		1093
		1094	our $MAX_SIGNAL_LATENCY = 10;
		1095
		1096	our %PROTOCOL; # (ipv4|ipv6) => (1|2), higher numbers are preferred
		1097
		1098	{
		1099	my $idx;
		1100	$PROTOCOL{$_} = ++$idx
		1101	for reverse split /\s*,\s*/,
		1102	$ENV{PERL_ANYEVENT_PROTOCOLS} || "ipv4,ipv6";
		1103	}
		1104
264	my @models = (	1105	my @models = (
265	[Coro::Event:: => AnyEvent::Impl::Coro::],
266	[EV:: => EV::AnyEvent::],	1106	[EV:: => AnyEvent::Impl::EV::],
267	[Event:: => AnyEvent::Impl::Event::],	1107	[Event:: => AnyEvent::Impl::Event::],
268	[Glib:: => AnyEvent::Impl::Glib::],
269	[Tk:: => AnyEvent::Impl::Tk::],
270	[AnyEvent::Impl::Perl:: => AnyEvent::Impl::Perl::],	1108	[AnyEvent::Impl::Perl:: => AnyEvent::Impl::Perl::],
		1109	# everything below here will not be autoprobed
		1110	# as the pureperl backend should work everywhere
		1111	# and is usually faster
		1112	[Glib:: => AnyEvent::Impl::Glib::], # becomes extremely slow with many watchers
		1113	[Event::Lib:: => AnyEvent::Impl::EventLib::], # too buggy
		1114	[Tk:: => AnyEvent::Impl::Tk::], # crashes with many handles
		1115	[Qt:: => AnyEvent::Impl::Qt::], # requires special main program
		1116	[POE::Kernel:: => AnyEvent::Impl::POE::], # lasciate ogni speranza
		1117	[Wx:: => AnyEvent::Impl::POE::],
		1118	[Prima:: => AnyEvent::Impl::POE::],
		1119	# IO::Async is just too broken - we would need workarounds for its
		1120	# byzantine signal and broken child handling, among others.
		1121	# IO::Async is rather hard to detect, as it doesn't have any
		1122	# obvious default class.
		1123	# [IO::Async:: => AnyEvent::Impl::IOAsync::], # requires special main program
		1124	# [IO::Async::Loop:: => AnyEvent::Impl::IOAsync::], # requires special main program
		1125	# [IO::Async::Notifier:: => AnyEvent::Impl::IOAsync::], # requires special main program
271	);	1126	);
272		1127
273	our %method = map +($_ => 1), qw(io timer condvar broadcast wait signal one_event DESTROY);	1128	our %method = map +($_ => 1),
		1129	qw(io timer time now now_update signal child idle condvar one_event DESTROY);
		1130
		1131	our @post_detect;
		1132
		1133	sub post_detect(&) {
		1134	my ($cb) = @_;
		1135
		1136	if ($MODEL) {
		1137	$cb->();
		1138
		1139	1
		1140	} else {
		1141	push @post_detect, $cb;
		1142
		1143	defined wantarray
		1144	? bless \$cb, "AnyEvent::Util::postdetect"
		1145	: ()
		1146	}
		1147	}
		1148
		1149	sub AnyEvent::Util::postdetect::DESTROY {
		1150	@post_detect = grep $_ != ${$_[0]}, @post_detect;
		1151	}
274		1152
275	sub detect() {	1153	sub detect() {
276	unless ($MODEL) {	1154	unless ($MODEL) {
277	no strict 'refs';	1155	local $SIG{__DIE__};
		1156
		1157	if ($ENV{PERL_ANYEVENT_MODEL} =~ /^([a-zA-Z]+)$/) {
		1158	my $model = "AnyEvent::Impl::$1";
		1159	if (eval "require $model") {
		1160	$MODEL = $model;
		1161	warn "AnyEvent: loaded model '$model' (forced by \$ENV{PERL_ANYEVENT_MODEL}), using it.\n" if $VERBOSE >= 2;
		1162	} else {
		1163	warn "AnyEvent: unable to load model '$model' (from \$ENV{PERL_ANYEVENT_MODEL}):\n$@" if $VERBOSE;
		1164	}
		1165	}
278		1166
279	# check for already loaded models	1167	# check for already loaded models
		1168	unless ($MODEL) {
280	for (@REGISTRY, @models) {	1169	for (@REGISTRY, @models) {
281	my ($package, $model) = @$_;	1170	my ($package, $model) = @$_;
282	if (${"$package\::VERSION"} > 0) {	1171	if (${"$package\::VERSION"} > 0) {
283	if (eval "require $model") {	1172	if (eval "require $model") {
284	$MODEL = $model;	1173	$MODEL = $model;
285	warn "AnyEvent: found model '$model', using it.\n" if $verbose > 1;	1174	warn "AnyEvent: autodetected model '$model', using it.\n" if $VERBOSE >= 2;
286	last;	1175	last;
		1176	}
287	}	1177	}
288	}	1178	}
		1179
		1180	unless ($MODEL) {
		1181	# try to load a model
		1182
		1183	for (@REGISTRY, @models) {
		1184	my ($package, $model) = @$_;
		1185	if (eval "require $package"
		1186	and ${"$package\::VERSION"} > 0
		1187	and eval "require $model") {
		1188	$MODEL = $model;
		1189	warn "AnyEvent: autoprobed model '$model', using it.\n" if $VERBOSE >= 2;
		1190	last;
		1191	}
		1192	}
		1193
		1194	$MODEL
		1195	or die "No event module selected for AnyEvent and autodetect failed. Install any one of these modules: EV, Event or Glib.\n";
		1196	}
289	}	1197	}
290		1198
291	unless ($MODEL) {	1199	push @{"$MODEL\::ISA"}, "AnyEvent::Base";
292	# try to load a model
293
294	for (@REGISTRY, @models) {
295	my ($package, $model) = @$_;
296	if (eval "require $package"
297	and ${"$package\::VERSION"} > 0
298	and eval "require $model") {
299	$MODEL = $model;
300	warn "AnyEvent: autoprobed and loaded model '$model', using it.\n" if $verbose > 1;
301	last;
302	}
303	}
304
305	$MODEL
306	or die "No event module selected for AnyEvent and autodetect failed. Install any one of these modules: Event (or Coro+Event), Glib or Tk.";
307	}
308		1200
309	unshift @ISA, $MODEL;	1201	unshift @ISA, $MODEL;
310	push @{"$MODEL\::ISA"}, "AnyEvent::Base";	1202
		1203	require AnyEvent::Strict if $ENV{PERL_ANYEVENT_STRICT};
		1204
		1205	(shift @post_detect)->() while @post_detect;
311	}	1206	}
312		1207
313	$MODEL	1208	$MODEL
314	}	1209	}
315		1210
316	sub AUTOLOAD {	1211	sub AUTOLOAD {
317	(my $func = $AUTOLOAD) =~ s/.*://;	1212	(my $func = $AUTOLOAD) =~ s/.*://;
318		1213
319	$method{$func}	1214	$method{$func}
320	or croak "$func: not a valid method for AnyEvent objects";	1215	or Carp::croak "$func: not a valid method for AnyEvent objects";
321		1216
322	detect unless $MODEL;	1217	detect unless $MODEL;
323		1218
324	my $class = shift;	1219	my $class = shift;
325	$class->$func (@_);	1220	$class->$func (@_);
326	}	1221	}
327		1222
		1223	# utility function to dup a filehandle. this is used by many backends
		1224	# to support binding more than one watcher per filehandle (they usually
		1225	# allow only one watcher per fd, so we dup it to get a different one).
		1226	sub _dupfh($$;$$) {
		1227	my ($poll, $fh, $r, $w) = @_;
		1228
		1229	# cygwin requires the fh mode to be matching, unix doesn't
		1230	my ($rw, $mode) = $poll eq "r" ? ($r, "<&") : ($w, ">&");
		1231
		1232	open my $fh2, $mode, $fh
		1233	or die "AnyEvent->io: cannot dup() filehandle in mode '$poll': $!,";
		1234
		1235	# we assume CLOEXEC is already set by perl in all important cases
		1236
		1237	($fh2, $rw)
		1238	}
		1239
328	package AnyEvent::Base;	1240	package AnyEvent::Base;
329		1241
		1242	# default implementations for many methods
		1243
		1244	sub _time {
		1245	# probe for availability of Time::HiRes
		1246	if (eval "use Time::HiRes (); Time::HiRes::time (); 1") {
		1247	warn "AnyEvent: using Time::HiRes for sub-second timing accuracy.\n" if $VERBOSE >= 8;
		1248	*_time = \&Time::HiRes::time;
		1249	# if (eval "use POSIX (); (POSIX::times())...
		1250	} else {
		1251	warn "AnyEvent: using built-in time(), WARNING, no sub-second resolution!\n" if $VERBOSE;
		1252	*_time = sub { time }; # epic fail
		1253	}
		1254
		1255	&_time
		1256	}
		1257
		1258	sub time { _time }
		1259	sub now { _time }
		1260	sub now_update { }
		1261
330	# default implementation for ->condvar, ->wait, ->broadcast	1262	# default implementation for ->condvar
331		1263
332	sub condvar {	1264	sub condvar {
333	bless \my $flag, "AnyEvent::Base::CondVar"	1265	bless { @_ == 3 ? (_ae_cb => $_[2]) : () }, "AnyEvent::CondVar"
334	}
335
336	sub AnyEvent::Base::CondVar::broadcast {
337	${$_[0]}++;
338	}
339
340	sub AnyEvent::Base::CondVar::wait {
341	AnyEvent->one_event while !${$_[0]};
342	}	1266	}
343		1267
344	# default implementation for ->signal	1268	# default implementation for ->signal
345		1269
346	our %SIG_CB;	1270	our $HAVE_ASYNC_INTERRUPT;
		1271	our ($SIGPIPE_R, $SIGPIPE_W, %SIG_CB, %SIG_EV, $SIG_IO);
		1272	our (%SIG_ASY, %SIG_ASY_W);
		1273	our ($SIG_COUNT, $SIG_TW);
347		1274
		1275	sub _signal_exec {
		1276	$HAVE_ASYNC_INTERRUPT
		1277	? $SIGPIPE_R->drain
		1278	: sysread $SIGPIPE_R, my $dummy, 9;
		1279
		1280	while (%SIG_EV) {
		1281	for (keys %SIG_EV) {
		1282	delete $SIG_EV{$_};
		1283	$_->() for values %{ $SIG_CB{$_} || {} };
		1284	}
		1285	}
		1286	}
		1287
		1288	# install a dumym wakeupw atcher to reduce signal catching latency
		1289	sub _sig_add() {
		1290	unless ($SIG_COUNT++) {
		1291	# try to align timer on a full-second boundary, if possible
		1292	my $NOW = AnyEvent->now;
		1293
		1294	$SIG_TW = AnyEvent->timer (
		1295	after => $MAX_SIGNAL_LATENCY - ($NOW - int $NOW),
		1296	interval => $MAX_SIGNAL_LATENCY,
		1297	cb => sub { }, # just for the PERL_ASYNC_CHECK
		1298	);
		1299	}
		1300	}
		1301
		1302	sub _sig_del {
		1303	undef $SIG_TW
		1304	unless --$SIG_COUNT;
		1305	}
		1306
348	sub signal {	1307	sub _signal {
349	my (undef, %arg) = @_;	1308	my (undef, %arg) = @_;
350		1309
351	my $signal = uc $arg{signal}	1310	my $signal = uc $arg{signal}
352	or Carp::croak "required option 'signal' is missing";	1311	or Carp::croak "required option 'signal' is missing";
353		1312
354	$SIG_CB{$signal}{$arg{cb}} = $arg{cb};	1313	$SIG_CB{$signal}{$arg{cb}} = $arg{cb};
		1314
		1315	if ($HAVE_ASYNC_INTERRUPT) {
		1316	# async::interrupt
		1317
		1318	$SIG_ASY{$signal} ||= do {
		1319	my $asy = new Async::Interrupt
		1320	cb => sub { undef $SIG_EV{$signal} },
		1321	signal => $signal,
		1322	pipe => [$SIGPIPE_R->filenos],
		1323	;
		1324	$asy->pipe_autodrain (0);
		1325
		1326	$asy
		1327	};
		1328
		1329	} else {
		1330	# pure perl
		1331
355	$SIG{$signal} ||= sub {	1332	$SIG{$signal} ||= sub {
356	$_->() for values %{ $SIG_CB{$signal} || {} };	1333	local $!;
		1334	syswrite $SIGPIPE_W, "\x00", 1 unless %SIG_EV;
		1335	undef $SIG_EV{$signal};
		1336	};
		1337
		1338	# can't do signal processing without introducing races in pure perl,
		1339	# so limit the signal latency.
		1340	_sig_add;
357	};	1341	}
358		1342
359	bless [$signal, $arg{cb}], "AnyEvent::Base::Signal"	1343	bless [$signal, $arg{cb}], "AnyEvent::Base::signal"
360	}	1344	}
361		1345
		1346	sub signal {
		1347	# probe for availability of Async::Interrupt
		1348	if (!$ENV{PERL_ANYEVENT_AVOID_ASYNC_INTERRUPT} && eval "use Async::Interrupt 0.6 (); 1") {
		1349	warn "AnyEvent: using Async::Interrupt for race-free signal handling.\n" if $VERBOSE >= 8;
		1350
		1351	$HAVE_ASYNC_INTERRUPT = 1;
		1352	$SIGPIPE_R = new Async::Interrupt::EventPipe;
		1353	$SIG_IO = AnyEvent->io (fh => $SIGPIPE_R->fileno, poll => "r", cb => \&_signal_exec);
		1354
		1355	} else {
		1356	warn "AnyEvent: using emulated perl signal handling with latency timer.\n" if $VERBOSE >= 8;
		1357
		1358	require Fcntl;
		1359
		1360	if (AnyEvent::WIN32) {
		1361	require AnyEvent::Util;
		1362
		1363	($SIGPIPE_R, $SIGPIPE_W) = AnyEvent::Util::portable_pipe ();
		1364	AnyEvent::Util::fh_nonblocking ($SIGPIPE_R) if $SIGPIPE_R;
		1365	AnyEvent::Util::fh_nonblocking ($SIGPIPE_W) if $SIGPIPE_W; # just in case
		1366	} else {
		1367	pipe $SIGPIPE_R, $SIGPIPE_W;
		1368	fcntl $SIGPIPE_R, &Fcntl::F_SETFL, &Fcntl::O_NONBLOCK if $SIGPIPE_R;
		1369	fcntl $SIGPIPE_W, &Fcntl::F_SETFL, &Fcntl::O_NONBLOCK if $SIGPIPE_W; # just in case
		1370
		1371	# not strictly required, as $^F is normally 2, but let's make sure...
		1372	fcntl $SIGPIPE_R, &Fcntl::F_SETFD, &Fcntl::FD_CLOEXEC;
		1373	fcntl $SIGPIPE_W, &Fcntl::F_SETFD, &Fcntl::FD_CLOEXEC;
		1374	}
		1375
		1376	$SIGPIPE_R
		1377	or Carp::croak "AnyEvent: unable to create a signal reporting pipe: $!\n";
		1378
		1379	$SIG_IO = AnyEvent->io (fh => $SIGPIPE_R, poll => "r", cb => \&_signal_exec);
		1380	}
		1381
		1382	*signal = \&_signal;
		1383	&signal
		1384	}
		1385
362	sub AnyEvent::Base::Signal::DESTROY {	1386	sub AnyEvent::Base::signal::DESTROY {
363	my ($signal, $cb) = @{$_[0]};	1387	my ($signal, $cb) = @{$_[0]};
364		1388
		1389	_sig_del;
		1390
365	delete $SIG_CB{$signal}{$cb};	1391	delete $SIG_CB{$signal}{$cb};
366		1392
367	$SIG{$signal} = 'DEFAULT' unless keys %{ $SIG_CB{$signal} };	1393	$HAVE_ASYNC_INTERRUPT
		1394	? delete $SIG_ASY{$signal}
		1395	: # delete doesn't work with older perls - they then
		1396	# print weird messages, or just unconditionally exit
		1397	# instead of getting the default action.
		1398	undef $SIG{$signal}
		1399	unless keys %{ $SIG_CB{$signal} };
368	}	1400	}
369		1401
370	# default implementation for ->child	1402	# default implementation for ->child
371		1403
372	our %PID_CB;	1404	our %PID_CB;
373	our $CHLD_W;	1405	our $CHLD_W;
374	our $PID_IDLE;	1406	our $CHLD_DELAY_W;
375	our $WNOHANG;	1407	our $WNOHANG;
376		1408
377	sub _child_wait {	1409	sub _sigchld {
378	while (0 < (my $pid = waitpid -1, $WNOHANG)) {	1410	while (0 < (my $pid = waitpid -1, $WNOHANG)) {
		1411	$_->($pid, $?)
379	$_->() for values %{ (delete $PID_CB{$pid}) || {} };	1412	for values %{ $PID_CB{$pid} || {} },
		1413	values %{ $PID_CB{0} || {} };
380	}	1414	}
381
382	undef $PID_IDLE;
383	}	1415	}
384		1416
385	sub child {	1417	sub child {
386	my (undef, %arg) = @_;	1418	my (undef, %arg) = @_;
387		1419
388	my $pid = uc $arg{pid}	1420	defined (my $pid = $arg{pid} + 0)
389	or Carp::croak "required option 'pid' is missing";	1421	or Carp::croak "required option 'pid' is missing";
390		1422
391	$PID_CB{$pid}{$arg{cb}} = $arg{cb};	1423	$PID_CB{$pid}{$arg{cb}} = $arg{cb};
392		1424
393	unless ($WNOHANG) {	1425	# WNOHANG is almost cetrainly 1 everywhere
394	$WNOHANG = eval { require POSIX; &POSIX::WNOHANG } || 1;	1426	$WNOHANG ||= $^O =~ /^(?:openbsd|netbsd|linux|freebsd|cygwin|MSWin32)$/
		1427	? 1
		1428	: eval { local $SIG{__DIE__}; require POSIX; &POSIX::WNOHANG } || 1;
		1429
		1430	unless ($CHLD_W) {
		1431	$CHLD_W = AnyEvent->signal (signal => 'CHLD', cb => \&_sigchld);
		1432	# child could be a zombie already, so make at least one round
		1433	&_sigchld;
395	}	1434	}
396		1435
397	unless ($CHLD_W) {
398	$CHLD_W = AnyEvent->signal (signal => 'CHLD', cb => \&_child_wait);
399	# child could be a zombie already
400	$PID_IDLE ||= AnyEvent->timer (after => 0, cb => \&_child_wait);
401	}
402
403	bless [$pid, $arg{cb}], "AnyEvent::Base::Child"	1436	bless [$pid, $arg{cb}], "AnyEvent::Base::child"
404	}	1437	}
405		1438
406	sub AnyEvent::Base::Child::DESTROY {	1439	sub AnyEvent::Base::child::DESTROY {
407	my ($pid, $cb) = @{$_[0]};	1440	my ($pid, $cb) = @{$_[0]};
408		1441
409	delete $PID_CB{$pid}{$cb};	1442	delete $PID_CB{$pid}{$cb};
410	delete $PID_CB{$pid} unless keys %{ $PID_CB{$pid} };	1443	delete $PID_CB{$pid} unless keys %{ $PID_CB{$pid} };
411		1444
412	undef $CHLD_W unless keys %PID_CB;	1445	undef $CHLD_W unless keys %PID_CB;
413	}	1446	}
414		1447
		1448	# idle emulation is done by simply using a timer, regardless
		1449	# of whether the process is idle or not, and not letting
		1450	# the callback use more than 50% of the time.
		1451	sub idle {
		1452	my (undef, %arg) = @_;
		1453
		1454	my ($cb, $w, $rcb) = $arg{cb};
		1455
		1456	$rcb = sub {
		1457	if ($cb) {
		1458	$w = _time;
		1459	&$cb;
		1460	$w = _time - $w;
		1461
		1462	# never use more then 50% of the time for the idle watcher,
		1463	# within some limits
		1464	$w = 0.0001 if $w < 0.0001;
		1465	$w = 5 if $w > 5;
		1466
		1467	$w = AnyEvent->timer (after => $w, cb => $rcb);
		1468	} else {
		1469	# clean up...
		1470	undef $w;
		1471	undef $rcb;
		1472	}
		1473	};
		1474
		1475	$w = AnyEvent->timer (after => 0.05, cb => $rcb);
		1476
		1477	bless \\$cb, "AnyEvent::Base::idle"
		1478	}
		1479
		1480	sub AnyEvent::Base::idle::DESTROY {
		1481	undef $${$_[0]};
		1482	}
		1483
		1484	package AnyEvent::CondVar;
		1485
		1486	our @ISA = AnyEvent::CondVar::Base::;
		1487
		1488	package AnyEvent::CondVar::Base;
		1489
		1490	#use overload
		1491	# '&{}' => sub { my $self = shift; sub { $self->send (@_) } },
		1492	# fallback => 1;
		1493
		1494	# save 300+ kilobytes by dirtily hardcoding overloading
		1495	${"AnyEvent::CondVar::Base::OVERLOAD"}{dummy}++; # Register with magic by touching.
		1496	*{'AnyEvent::CondVar::Base::()'} = sub { }; # "Make it findable via fetchmethod."
		1497	*{'AnyEvent::CondVar::Base::(&{}'} = sub { my $self = shift; sub { $self->send (@_) } }; # &{}
		1498	${'AnyEvent::CondVar::Base::()'} = 1; # fallback
		1499
		1500	our $WAITING;
		1501
		1502	sub _send {
		1503	# nop
		1504	}
		1505
		1506	sub send {
		1507	my $cv = shift;
		1508	$cv->{_ae_sent} = [@_];
		1509	(delete $cv->{_ae_cb})->($cv) if $cv->{_ae_cb};
		1510	$cv->_send;
		1511	}
		1512
		1513	sub croak {
		1514	$_[0]{_ae_croak} = $_[1];
		1515	$_[0]->send;
		1516	}
		1517
		1518	sub ready {
		1519	$_[0]{_ae_sent}
		1520	}
		1521
		1522	sub _wait {
		1523	$WAITING
		1524	and !$_[0]{_ae_sent}
		1525	and Carp::croak "AnyEvent::CondVar: recursive blocking wait detected";
		1526
		1527	local $WAITING = 1;
		1528	AnyEvent->one_event while !$_[0]{_ae_sent};
		1529	}
		1530
		1531	sub recv {
		1532	$_[0]->_wait;
		1533
		1534	Carp::croak $_[0]{_ae_croak} if $_[0]{_ae_croak};
		1535	wantarray ? @{ $_[0]{_ae_sent} } : $_[0]{_ae_sent}[0]
		1536	}
		1537
		1538	sub cb {
		1539	$_[0]{_ae_cb} = $_[1] if @_ > 1;
		1540	$_[0]{_ae_cb}
		1541	}
		1542
		1543	sub begin {
		1544	++$_[0]{_ae_counter};
		1545	$_[0]{_ae_end_cb} = $_[1] if @_ > 1;
		1546	}
		1547
		1548	sub end {
		1549	return if --$_[0]{_ae_counter};
		1550	&{ $_[0]{_ae_end_cb} || sub { $_[0]->send } };
		1551	}
		1552
		1553	# undocumented/compatibility with pre-3.4
		1554	*broadcast = \&send;
		1555	*wait = \&_wait;
		1556
		1557	=head1 ERROR AND EXCEPTION HANDLING
		1558
		1559	In general, AnyEvent does not do any error handling - it relies on the
		1560	caller to do that if required. The L<AnyEvent::Strict> module (see also
		1561	the C<PERL_ANYEVENT_STRICT> environment variable, below) provides strict
		1562	checking of all AnyEvent methods, however, which is highly useful during
		1563	development.
		1564
		1565	As for exception handling (i.e. runtime errors and exceptions thrown while
		1566	executing a callback), this is not only highly event-loop specific, but
		1567	also not in any way wrapped by this module, as this is the job of the main
		1568	program.
		1569
		1570	The pure perl event loop simply re-throws the exception (usually
		1571	within C<< condvar->recv >>), the L<Event> and L<EV> modules call C<<
		1572	$Event/EV::DIED->() >>, L<Glib> uses C<< install_exception_handler >> and
		1573	so on.
		1574
		1575	=head1 ENVIRONMENT VARIABLES
		1576
		1577	The following environment variables are used by this module or its
		1578	submodules.
		1579
		1580	Note that AnyEvent will remove I<all> environment variables starting with
		1581	C<PERL_ANYEVENT_> from C<%ENV> when it is loaded while taint mode is
		1582	enabled.
		1583
		1584	=over 4
		1585
		1586	=item C<PERL_ANYEVENT_VERBOSE>
		1587
		1588	By default, AnyEvent will be completely silent except in fatal
		1589	conditions. You can set this environment variable to make AnyEvent more
		1590	talkative.
		1591
		1592	When set to C<1> or higher, causes AnyEvent to warn about unexpected
		1593	conditions, such as not being able to load the event model specified by
		1594	C<PERL_ANYEVENT_MODEL>.
		1595
		1596	When set to C<2> or higher, cause AnyEvent to report to STDERR which event
		1597	model it chooses.
		1598
		1599	When set to C<8> or higher, then AnyEvent will report extra information on
		1600	which optional modules it loads and how it implements certain features.
		1601
		1602	=item C<PERL_ANYEVENT_STRICT>
		1603
		1604	AnyEvent does not do much argument checking by default, as thorough
		1605	argument checking is very costly. Setting this variable to a true value
		1606	will cause AnyEvent to load C<AnyEvent::Strict> and then to thoroughly
		1607	check the arguments passed to most method calls. If it finds any problems,
		1608	it will croak.
		1609
		1610	In other words, enables "strict" mode.
		1611
		1612	Unlike C<use strict> (or it's modern cousin, C<< use L<common::sense>
		1613	>>, it is definitely recommended to keep it off in production. Keeping
		1614	C<PERL_ANYEVENT_STRICT=1> in your environment while developing programs
		1615	can be very useful, however.
		1616
		1617	=item C<PERL_ANYEVENT_MODEL>
		1618
		1619	This can be used to specify the event model to be used by AnyEvent, before
		1620	auto detection and -probing kicks in. It must be a string consisting
		1621	entirely of ASCII letters. The string C<AnyEvent::Impl::> gets prepended
		1622	and the resulting module name is loaded and if the load was successful,
		1623	used as event model. If it fails to load AnyEvent will proceed with
		1624	auto detection and -probing.
		1625
		1626	This functionality might change in future versions.
		1627
		1628	For example, to force the pure perl model (L<AnyEvent::Impl::Perl>) you
		1629	could start your program like this:
		1630
		1631	PERL_ANYEVENT_MODEL=Perl perl ...
		1632
		1633	=item C<PERL_ANYEVENT_PROTOCOLS>
		1634
		1635	Used by both L<AnyEvent::DNS> and L<AnyEvent::Socket> to determine preferences
		1636	for IPv4 or IPv6. The default is unspecified (and might change, or be the result
		1637	of auto probing).
		1638
		1639	Must be set to a comma-separated list of protocols or address families,
		1640	current supported: C<ipv4> and C<ipv6>. Only protocols mentioned will be
		1641	used, and preference will be given to protocols mentioned earlier in the
		1642	list.
		1643
		1644	This variable can effectively be used for denial-of-service attacks
		1645	against local programs (e.g. when setuid), although the impact is likely
		1646	small, as the program has to handle conenction and other failures anyways.
		1647
		1648	Examples: C<PERL_ANYEVENT_PROTOCOLS=ipv4,ipv6> - prefer IPv4 over IPv6,
		1649	but support both and try to use both. C<PERL_ANYEVENT_PROTOCOLS=ipv4>
		1650	- only support IPv4, never try to resolve or contact IPv6
		1651	addresses. C<PERL_ANYEVENT_PROTOCOLS=ipv6,ipv4> support either IPv4 or
		1652	IPv6, but prefer IPv6 over IPv4.
		1653
		1654	=item C<PERL_ANYEVENT_EDNS0>
		1655
		1656	Used by L<AnyEvent::DNS> to decide whether to use the EDNS0 extension
		1657	for DNS. This extension is generally useful to reduce DNS traffic, but
		1658	some (broken) firewalls drop such DNS packets, which is why it is off by
		1659	default.
		1660
		1661	Setting this variable to C<1> will cause L<AnyEvent::DNS> to announce
		1662	EDNS0 in its DNS requests.
		1663
		1664	=item C<PERL_ANYEVENT_MAX_FORKS>
		1665
		1666	The maximum number of child processes that C<AnyEvent::Util::fork_call>
		1667	will create in parallel.
		1668
		1669	=item C<PERL_ANYEVENT_MAX_OUTSTANDING_DNS>
		1670
		1671	The default value for the C<max_outstanding> parameter for the default DNS
		1672	resolver - this is the maximum number of parallel DNS requests that are
		1673	sent to the DNS server.
		1674
		1675	=item C<PERL_ANYEVENT_RESOLV_CONF>
		1676
		1677	The file to use instead of F</etc/resolv.conf> (or OS-specific
		1678	configuration) in the default resolver. When set to the empty string, no
		1679	default config will be used.
		1680
		1681	=item C<PERL_ANYEVENT_CA_FILE>, C<PERL_ANYEVENT_CA_PATH>.
		1682
		1683	When neither C<ca_file> nor C<ca_path> was specified during
		1684	L<AnyEvent::TLS> context creation, and either of these environment
		1685	variables exist, they will be used to specify CA certificate locations
		1686	instead of a system-dependent default.
		1687
		1688	=item C<PERL_ANYEVENT_AVOID_GUARD> and C<PERL_ANYEVENT_AVOID_ASYNC_INTERRUPT>
		1689
		1690	When these are set to C<1>, then the respective modules are not
		1691	loaded. Mostly good for testing AnyEvent itself.
		1692
		1693	=back
		1694
415	=head1 SUPPLYING YOUR OWN EVENT MODEL INTERFACE	1695	=head1 SUPPLYING YOUR OWN EVENT MODEL INTERFACE
		1696
		1697	This is an advanced topic that you do not normally need to use AnyEvent in
		1698	a module. This section is only of use to event loop authors who want to
		1699	provide AnyEvent compatibility.
416		1700
417	If you need to support another event library which isn't directly	1701	If you need to support another event library which isn't directly
418	supported by AnyEvent, you can supply your own interface to it by	1702	supported by AnyEvent, you can supply your own interface to it by
419	pushing, before the first watcher gets created, the package name of	1703	pushing, before the first watcher gets created, the package name of
420	the event module and the package name of the interface to use onto	1704	the event module and the package name of the interface to use onto
421	C<@AnyEvent::REGISTRY>. You can do that before and even without loading	1705	C<@AnyEvent::REGISTRY>. You can do that before and even without loading
422	AnyEvent.	1706	AnyEvent, so it is reasonably cheap.
423		1707
424	Example:	1708	Example:
425		1709
426	push @AnyEvent::REGISTRY, [urxvt => urxvt::anyevent::];	1710	push @AnyEvent::REGISTRY, [urxvt => urxvt::anyevent::];
427		1711
428	This tells AnyEvent to (literally) use the C<urxvt::anyevent::>	1712	This tells AnyEvent to (literally) use the C<urxvt::anyevent::>
429	package/class when it finds the C<urxvt> package/module is loaded. When	1713	package/class when it finds the C<urxvt> package/module is already loaded.
		1714
430	AnyEvent is loaded and asked to find a suitable event model, it will	1715	When AnyEvent is loaded and asked to find a suitable event model, it
431	first check for the presence of urxvt.	1716	will first check for the presence of urxvt by trying to C<use> the
		1717	C<urxvt::anyevent> module.
432		1718
433	The class should provide implementations for all watcher types (see	1719	The class should provide implementations for all watcher types. See
434	L<AnyEvent::Impl::Event> (source code), L<AnyEvent::Impl::Glib>	1720	L<AnyEvent::Impl::EV> (source code), L<AnyEvent::Impl::Glib> (Source code)
435	(Source code) and so on for actual examples, use C<perldoc -m	1721	and so on for actual examples. Use C<perldoc -m AnyEvent::Impl::Glib> to
436	AnyEvent::Impl::Glib> to see the sources).	1722	see the sources.
437		1723
		1724	If you don't provide C<signal> and C<child> watchers than AnyEvent will
		1725	provide suitable (hopefully) replacements.
		1726
438	The above isn't fictitious, the I<rxvt-unicode> (a.k.a. urxvt)	1727	The above example isn't fictitious, the I<rxvt-unicode> (a.k.a. urxvt)
439	uses the above line as-is. An interface isn't included in AnyEvent	1728	terminal emulator uses the above line as-is. An interface isn't included
440	because it doesn't make sense outside the embedded interpreter inside	1729	in AnyEvent because it doesn't make sense outside the embedded interpreter
441	I<rxvt-unicode>, and it is updated and maintained as part of the	1730	inside I<rxvt-unicode>, and it is updated and maintained as part of the
442	I<rxvt-unicode> distribution.	1731	I<rxvt-unicode> distribution.
443		1732
444	I<rxvt-unicode> also cheats a bit by not providing blocking access to	1733	I<rxvt-unicode> also cheats a bit by not providing blocking access to
445	condition variables: code blocking while waiting for a condition will	1734	condition variables: code blocking while waiting for a condition will
446	C<die>. This still works with most modules/usages, and blocking calls must	1735	C<die>. This still works with most modules/usages, and blocking calls must
447	not be in an interactive application, so it makes sense.	1736	not be done in an interactive application, so it makes sense.
448		1737
449	=head1 ENVIRONMENT VARIABLES
450
451	The following environment variables are used by this module:
452
453	C<PERL_ANYEVENT_VERBOSE> when set to C<2> or higher, reports which event
454	model gets used.
455
456	=head1 EXAMPLE	1738	=head1 EXAMPLE PROGRAM
457		1739
458	The following program uses an io watcher to read data from stdin, a timer	1740	The following program uses an I/O watcher to read data from STDIN, a timer
459	to display a message once per second, and a condvar to exit the program	1741	to display a message once per second, and a condition variable to quit the
460	when the user enters quit:	1742	program when the user enters quit:
461		1743
462	use AnyEvent;	1744	use AnyEvent;
463		1745
464	my $cv = AnyEvent->condvar;	1746	my $cv = AnyEvent->condvar;
465		1747
466	my $io_watcher = AnyEvent->io (fh => \*STDIN, poll => 'r', cb => sub {	1748	my $io_watcher = AnyEvent->io (
		1749	fh => \*STDIN,
		1750	poll => 'r',
		1751	cb => sub {
467	warn "io event <$_[0]>\n"; # will always output <r>	1752	warn "io event <$_[0]>\n"; # will always output <r>
468	chomp (my $input = <STDIN>); # read a line	1753	chomp (my $input = <STDIN>); # read a line
469	warn "read: $input\n"; # output what has been read	1754	warn "read: $input\n"; # output what has been read
470	$cv->broadcast if $input =~ /^q/i; # quit program if /^q/i	1755	$cv->send if $input =~ /^q/i; # quit program if /^q/i
		1756	},
471	});	1757	);
472		1758
473	my $time_watcher; # can only be used once	1759	my $time_watcher; # can only be used once
474		1760
475	sub new_timer {	1761	sub new_timer {
476	$timer = AnyEvent->timer (after => 1, cb => sub {	1762	$timer = AnyEvent->timer (after => 1, cb => sub {
…		…
479	});	1765	});
480	}	1766	}
481		1767
482	new_timer; # create first timer	1768	new_timer; # create first timer
483		1769
484	$cv->wait; # wait until user enters /^q/i	1770	$cv->recv; # wait until user enters /^q/i
485		1771
486	=head1 REAL-WORLD EXAMPLE	1772	=head1 REAL-WORLD EXAMPLE
487		1773
488	Consider the L<Net::FCP> module. It features (among others) the following	1774	Consider the L<Net::FCP> module. It features (among others) the following
489	API calls, which are to freenet what HTTP GET requests are to http:	1775	API calls, which are to freenet what HTTP GET requests are to http:
…		…
539	syswrite $txn->{fh}, $txn->{request}	1825	syswrite $txn->{fh}, $txn->{request}
540	or die "connection or write error";	1826	or die "connection or write error";
541	$txn->{w} = AnyEvent->io (fh => $txn->{fh}, poll => 'r', cb => sub { $txn->fh_ready_r });	1827	$txn->{w} = AnyEvent->io (fh => $txn->{fh}, poll => 'r', cb => sub { $txn->fh_ready_r });
542		1828
543	Again, C<fh_ready_r> waits till all data has arrived, and then stores the	1829	Again, C<fh_ready_r> waits till all data has arrived, and then stores the
544	result and signals any possible waiters that the request ahs finished:	1830	result and signals any possible waiters that the request has finished:
545		1831
546	sysread $txn->{fh}, $txn->{buf}, length $txn->{$buf};	1832	sysread $txn->{fh}, $txn->{buf}, length $txn->{$buf};
547		1833
548	if (end-of-file or data complete) {	1834	if (end-of-file or data complete) {
549	$txn->{result} = $txn->{buf};	1835	$txn->{result} = $txn->{buf};
550	$txn->{finished}->broadcast;	1836	$txn->{finished}->send;
551	$txb->{cb}->($txn) of $txn->{cb}; # also call callback	1837	$txb->{cb}->($txn) of $txn->{cb}; # also call callback
552	}	1838	}
553		1839
554	The C<result> method, finally, just waits for the finished signal (if the	1840	The C<result> method, finally, just waits for the finished signal (if the
555	request was already finished, it doesn't wait, of course, and returns the	1841	request was already finished, it doesn't wait, of course, and returns the
556	data:	1842	data:
557		1843
558	$txn->{finished}->wait;	1844	$txn->{finished}->recv;
559	return $txn->{result};	1845	return $txn->{result};
560		1846
561	The actual code goes further and collects all errors (C<die>s, exceptions)	1847	The actual code goes further and collects all errors (C<die>s, exceptions)
562	that occured during request processing. The C<result> method detects	1848	that occurred during request processing. The C<result> method detects
563	wether an exception as thrown (it is stored inside the $txn object)	1849	whether an exception as thrown (it is stored inside the $txn object)
564	and just throws the exception, which means connection errors and other	1850	and just throws the exception, which means connection errors and other
565	problems get reported tot he code that tries to use the result, not in a	1851	problems get reported tot he code that tries to use the result, not in a
566	random callback.	1852	random callback.
567		1853
568	All of this enables the following usage styles:	1854	All of this enables the following usage styles:
569		1855
570	1. Blocking:	1856	1. Blocking:
571		1857
572	my $data = $fcp->client_get ($url);	1858	my $data = $fcp->client_get ($url);
573		1859
574	2. Blocking, but parallelizing:	1860	2. Blocking, but running in parallel:
575		1861
576	my @datas = map $_->result,	1862	my @datas = map $_->result,
577	map $fcp->txn_client_get ($_),	1863	map $fcp->txn_client_get ($_),
578	@urls;	1864	@urls;
579		1865
580	Both blocking examples work without the module user having to know	1866	Both blocking examples work without the module user having to know
581	anything about events.	1867	anything about events.
582		1868
583	3a. Event-based in a main program, using any support Event module:	1869	3a. Event-based in a main program, using any supported event module:
584		1870
585	use Event;	1871	use EV;
586		1872
587	$fcp->txn_client_get ($url)->cb (sub {	1873	$fcp->txn_client_get ($url)->cb (sub {
588	my $txn = shift;	1874	my $txn = shift;
589	my $data = $txn->result;	1875	my $data = $txn->result;
590	...	1876	...
591	});	1877	});
592		1878
593	Event::loop;	1879	EV::loop;
594		1880
595	3b. The module user could use AnyEvent, too:	1881	3b. The module user could use AnyEvent, too:
596		1882
597	use AnyEvent;	1883	use AnyEvent;
598		1884
599	my $quit = AnyEvent->condvar;	1885	my $quit = AnyEvent->condvar;
600		1886
601	$fcp->txn_client_get ($url)->cb (sub {	1887	$fcp->txn_client_get ($url)->cb (sub {
602	...	1888	...
603	$quit->broadcast;	1889	$quit->send;
604	});	1890	});
605		1891
606	$quit->wait;	1892	$quit->recv;
		1893
		1894
		1895	=head1 BENCHMARKS
		1896
		1897	To give you an idea of the performance and overheads that AnyEvent adds
		1898	over the event loops themselves and to give you an impression of the speed
		1899	of various event loops I prepared some benchmarks.
		1900
		1901	=head2 BENCHMARKING ANYEVENT OVERHEAD
		1902
		1903	Here is a benchmark of various supported event models used natively and
		1904	through AnyEvent. The benchmark creates a lot of timers (with a zero
		1905	timeout) and I/O watchers (watching STDOUT, a pty, to become writable,
		1906	which it is), lets them fire exactly once and destroys them again.
		1907
		1908	Source code for this benchmark is found as F<eg/bench> in the AnyEvent
		1909	distribution.
		1910
		1911	=head3 Explanation of the columns
		1912
		1913	I<watcher> is the number of event watchers created/destroyed. Since
		1914	different event models feature vastly different performances, each event
		1915	loop was given a number of watchers so that overall runtime is acceptable
		1916	and similar between tested event loop (and keep them from crashing): Glib
		1917	would probably take thousands of years if asked to process the same number
		1918	of watchers as EV in this benchmark.
		1919
		1920	I<bytes> is the number of bytes (as measured by the resident set size,
		1921	RSS) consumed by each watcher. This method of measuring captures both C
		1922	and Perl-based overheads.
		1923
		1924	I<create> is the time, in microseconds (millionths of seconds), that it
		1925	takes to create a single watcher. The callback is a closure shared between
		1926	all watchers, to avoid adding memory overhead. That means closure creation
		1927	and memory usage is not included in the figures.
		1928
		1929	I<invoke> is the time, in microseconds, used to invoke a simple
		1930	callback. The callback simply counts down a Perl variable and after it was
		1931	invoked "watcher" times, it would C<< ->send >> a condvar once to
		1932	signal the end of this phase.
		1933
		1934	I<destroy> is the time, in microseconds, that it takes to destroy a single
		1935	watcher.
		1936
		1937	=head3 Results
		1938
		1939	name watchers bytes create invoke destroy comment
		1940	EV/EV 400000 224 0.47 0.35 0.27 EV native interface
		1941	EV/Any 100000 224 2.88 0.34 0.27 EV + AnyEvent watchers
		1942	CoroEV/Any 100000 224 2.85 0.35 0.28 coroutines + Coro::Signal
		1943	Perl/Any 100000 452 4.13 0.73 0.95 pure perl implementation
		1944	Event/Event 16000 517 32.20 31.80 0.81 Event native interface
		1945	Event/Any 16000 590 35.85 31.55 1.06 Event + AnyEvent watchers
		1946	IOAsync/Any 16000 989 38.10 32.77 11.13 via IO::Async::Loop::IO_Poll
		1947	IOAsync/Any 16000 990 37.59 29.50 10.61 via IO::Async::Loop::Epoll
		1948	Glib/Any 16000 1357 102.33 12.31 51.00 quadratic behaviour
		1949	Tk/Any 2000 1860 27.20 66.31 14.00 SEGV with >> 2000 watchers
		1950	POE/Event 2000 6328 109.99 751.67 14.02 via POE::Loop::Event
		1951	POE/Select 2000 6027 94.54 809.13 579.80 via POE::Loop::Select
		1952
		1953	=head3 Discussion
		1954
		1955	The benchmark does I<not> measure scalability of the event loop very
		1956	well. For example, a select-based event loop (such as the pure perl one)
		1957	can never compete with an event loop that uses epoll when the number of
		1958	file descriptors grows high. In this benchmark, all events become ready at
		1959	the same time, so select/poll-based implementations get an unnatural speed
		1960	boost.
		1961
		1962	Also, note that the number of watchers usually has a nonlinear effect on
		1963	overall speed, that is, creating twice as many watchers doesn't take twice
		1964	the time - usually it takes longer. This puts event loops tested with a
		1965	higher number of watchers at a disadvantage.
		1966
		1967	To put the range of results into perspective, consider that on the
		1968	benchmark machine, handling an event takes roughly 1600 CPU cycles with
		1969	EV, 3100 CPU cycles with AnyEvent's pure perl loop and almost 3000000 CPU
		1970	cycles with POE.
		1971
		1972	C<EV> is the sole leader regarding speed and memory use, which are both
		1973	maximal/minimal, respectively. Even when going through AnyEvent, it uses
		1974	far less memory than any other event loop and is still faster than Event
		1975	natively.
		1976
		1977	The pure perl implementation is hit in a few sweet spots (both the
		1978	constant timeout and the use of a single fd hit optimisations in the perl
		1979	interpreter and the backend itself). Nevertheless this shows that it
		1980	adds very little overhead in itself. Like any select-based backend its
		1981	performance becomes really bad with lots of file descriptors (and few of
		1982	them active), of course, but this was not subject of this benchmark.
		1983
		1984	The C<Event> module has a relatively high setup and callback invocation
		1985	cost, but overall scores in on the third place.
		1986
		1987	C<IO::Async> performs admirably well, about on par with C<Event>, even
		1988	when using its pure perl backend.
		1989
		1990	C<Glib>'s memory usage is quite a bit higher, but it features a
		1991	faster callback invocation and overall ends up in the same class as
		1992	C<Event>. However, Glib scales extremely badly, doubling the number of
		1993	watchers increases the processing time by more than a factor of four,
		1994	making it completely unusable when using larger numbers of watchers
		1995	(note that only a single file descriptor was used in the benchmark, so
		1996	inefficiencies of C<poll> do not account for this).
		1997
		1998	The C<Tk> adaptor works relatively well. The fact that it crashes with
		1999	more than 2000 watchers is a big setback, however, as correctness takes
		2000	precedence over speed. Nevertheless, its performance is surprising, as the
		2001	file descriptor is dup()ed for each watcher. This shows that the dup()
		2002	employed by some adaptors is not a big performance issue (it does incur a
		2003	hidden memory cost inside the kernel which is not reflected in the figures
		2004	above).
		2005
		2006	C<POE>, regardless of underlying event loop (whether using its pure perl
		2007	select-based backend or the Event module, the POE-EV backend couldn't
		2008	be tested because it wasn't working) shows abysmal performance and
		2009	memory usage with AnyEvent: Watchers use almost 30 times as much memory
		2010	as EV watchers, and 10 times as much memory as Event (the high memory
		2011	requirements are caused by requiring a session for each watcher). Watcher
		2012	invocation speed is almost 900 times slower than with AnyEvent's pure perl
		2013	implementation.
		2014
		2015	The design of the POE adaptor class in AnyEvent can not really account
		2016	for the performance issues, though, as session creation overhead is
		2017	small compared to execution of the state machine, which is coded pretty
		2018	optimally within L<AnyEvent::Impl::POE> (and while everybody agrees that
		2019	using multiple sessions is not a good approach, especially regarding
		2020	memory usage, even the author of POE could not come up with a faster
		2021	design).
		2022
		2023	=head3 Summary
		2024
		2025	=over 4
		2026
		2027	=item * Using EV through AnyEvent is faster than any other event loop
		2028	(even when used without AnyEvent), but most event loops have acceptable
		2029	performance with or without AnyEvent.
		2030
		2031	=item * The overhead AnyEvent adds is usually much smaller than the overhead of
		2032	the actual event loop, only with extremely fast event loops such as EV
		2033	adds AnyEvent significant overhead.
		2034
		2035	=item * You should avoid POE like the plague if you want performance or
		2036	reasonable memory usage.
		2037
		2038	=back
		2039
		2040	=head2 BENCHMARKING THE LARGE SERVER CASE
		2041
		2042	This benchmark actually benchmarks the event loop itself. It works by
		2043	creating a number of "servers": each server consists of a socket pair, a
		2044	timeout watcher that gets reset on activity (but never fires), and an I/O
		2045	watcher waiting for input on one side of the socket. Each time the socket
		2046	watcher reads a byte it will write that byte to a random other "server".
		2047
		2048	The effect is that there will be a lot of I/O watchers, only part of which
		2049	are active at any one point (so there is a constant number of active
		2050	fds for each loop iteration, but which fds these are is random). The
		2051	timeout is reset each time something is read because that reflects how
		2052	most timeouts work (and puts extra pressure on the event loops).
		2053
		2054	In this benchmark, we use 10000 socket pairs (20000 sockets), of which 100
		2055	(1%) are active. This mirrors the activity of large servers with many
		2056	connections, most of which are idle at any one point in time.
		2057
		2058	Source code for this benchmark is found as F<eg/bench2> in the AnyEvent
		2059	distribution.
		2060
		2061	=head3 Explanation of the columns
		2062
		2063	I<sockets> is the number of sockets, and twice the number of "servers" (as
		2064	each server has a read and write socket end).
		2065
		2066	I<create> is the time it takes to create a socket pair (which is
		2067	nontrivial) and two watchers: an I/O watcher and a timeout watcher.
		2068
		2069	I<request>, the most important value, is the time it takes to handle a
		2070	single "request", that is, reading the token from the pipe and forwarding
		2071	it to another server. This includes deleting the old timeout and creating
		2072	a new one that moves the timeout into the future.
		2073
		2074	=head3 Results
		2075
		2076	name sockets create request
		2077	EV 20000 69.01 11.16
		2078	Perl 20000 73.32 35.87
		2079	IOAsync 20000 157.00 98.14 epoll
		2080	IOAsync 20000 159.31 616.06 poll
		2081	Event 20000 212.62 257.32
		2082	Glib 20000 651.16 1896.30
		2083	POE 20000 349.67 12317.24 uses POE::Loop::Event
		2084
		2085	=head3 Discussion
		2086
		2087	This benchmark I<does> measure scalability and overall performance of the
		2088	particular event loop.
		2089
		2090	EV is again fastest. Since it is using epoll on my system, the setup time
		2091	is relatively high, though.
		2092
		2093	Perl surprisingly comes second. It is much faster than the C-based event
		2094	loops Event and Glib.
		2095
		2096	IO::Async performs very well when using its epoll backend, and still quite
		2097	good compared to Glib when using its pure perl backend.
		2098
		2099	Event suffers from high setup time as well (look at its code and you will
		2100	understand why). Callback invocation also has a high overhead compared to
		2101	the C<< $_->() for .. >>-style loop that the Perl event loop uses. Event
		2102	uses select or poll in basically all documented configurations.
		2103
		2104	Glib is hit hard by its quadratic behaviour w.r.t. many watchers. It
		2105	clearly fails to perform with many filehandles or in busy servers.
		2106
		2107	POE is still completely out of the picture, taking over 1000 times as long
		2108	as EV, and over 100 times as long as the Perl implementation, even though
		2109	it uses a C-based event loop in this case.
		2110
		2111	=head3 Summary
		2112
		2113	=over 4
		2114
		2115	=item * The pure perl implementation performs extremely well.
		2116
		2117	=item * Avoid Glib or POE in large projects where performance matters.
		2118
		2119	=back
		2120
		2121	=head2 BENCHMARKING SMALL SERVERS
		2122
		2123	While event loops should scale (and select-based ones do not...) even to
		2124	large servers, most programs we (or I :) actually write have only a few
		2125	I/O watchers.
		2126
		2127	In this benchmark, I use the same benchmark program as in the large server
		2128	case, but it uses only eight "servers", of which three are active at any
		2129	one time. This should reflect performance for a small server relatively
		2130	well.
		2131
		2132	The columns are identical to the previous table.
		2133
		2134	=head3 Results
		2135
		2136	name sockets create request
		2137	EV 16 20.00 6.54
		2138	Perl 16 25.75 12.62
		2139	Event 16 81.27 35.86
		2140	Glib 16 32.63 15.48
		2141	POE 16 261.87 276.28 uses POE::Loop::Event
		2142
		2143	=head3 Discussion
		2144
		2145	The benchmark tries to test the performance of a typical small
		2146	server. While knowing how various event loops perform is interesting, keep
		2147	in mind that their overhead in this case is usually not as important, due
		2148	to the small absolute number of watchers (that is, you need efficiency and
		2149	speed most when you have lots of watchers, not when you only have a few of
		2150	them).
		2151
		2152	EV is again fastest.
		2153
		2154	Perl again comes second. It is noticeably faster than the C-based event
		2155	loops Event and Glib, although the difference is too small to really
		2156	matter.
		2157
		2158	POE also performs much better in this case, but is is still far behind the
		2159	others.
		2160
		2161	=head3 Summary
		2162
		2163	=over 4
		2164
		2165	=item * C-based event loops perform very well with small number of
		2166	watchers, as the management overhead dominates.
		2167
		2168	=back
		2169
		2170	=head2 THE IO::Lambda BENCHMARK
		2171
		2172	Recently I was told about the benchmark in the IO::Lambda manpage, which
		2173	could be misinterpreted to make AnyEvent look bad. In fact, the benchmark
		2174	simply compares IO::Lambda with POE, and IO::Lambda looks better (which
		2175	shouldn't come as a surprise to anybody). As such, the benchmark is
		2176	fine, and mostly shows that the AnyEvent backend from IO::Lambda isn't
		2177	very optimal. But how would AnyEvent compare when used without the extra
		2178	baggage? To explore this, I wrote the equivalent benchmark for AnyEvent.
		2179
		2180	The benchmark itself creates an echo-server, and then, for 500 times,
		2181	connects to the echo server, sends a line, waits for the reply, and then
		2182	creates the next connection. This is a rather bad benchmark, as it doesn't
		2183	test the efficiency of the framework or much non-blocking I/O, but it is a
		2184	benchmark nevertheless.
		2185
		2186	name runtime
		2187	Lambda/select 0.330 sec
		2188	+ optimized 0.122 sec
		2189	Lambda/AnyEvent 0.327 sec
		2190	+ optimized 0.138 sec
		2191	Raw sockets/select 0.077 sec
		2192	POE/select, components 0.662 sec
		2193	POE/select, raw sockets 0.226 sec
		2194	POE/select, optimized 0.404 sec
		2195
		2196	AnyEvent/select/nb 0.085 sec
		2197	AnyEvent/EV/nb 0.068 sec
		2198	+state machine 0.134 sec
		2199
		2200	The benchmark is also a bit unfair (my fault): the IO::Lambda/POE
		2201	benchmarks actually make blocking connects and use 100% blocking I/O,
		2202	defeating the purpose of an event-based solution. All of the newly
		2203	written AnyEvent benchmarks use 100% non-blocking connects (using
		2204	AnyEvent::Socket::tcp_connect and the asynchronous pure perl DNS
		2205	resolver), so AnyEvent is at a disadvantage here, as non-blocking connects
		2206	generally require a lot more bookkeeping and event handling than blocking
		2207	connects (which involve a single syscall only).
		2208
		2209	The last AnyEvent benchmark additionally uses L<AnyEvent::Handle>, which
		2210	offers similar expressive power as POE and IO::Lambda, using conventional
		2211	Perl syntax. This means that both the echo server and the client are 100%
		2212	non-blocking, further placing it at a disadvantage.
		2213
		2214	As you can see, the AnyEvent + EV combination even beats the
		2215	hand-optimised "raw sockets benchmark", while AnyEvent + its pure perl
		2216	backend easily beats IO::Lambda and POE.
		2217
		2218	And even the 100% non-blocking version written using the high-level (and
		2219	slow :) L<AnyEvent::Handle> abstraction beats both POE and IO::Lambda by a
		2220	large margin, even though it does all of DNS, tcp-connect and socket I/O
		2221	in a non-blocking way.
		2222
		2223	The two AnyEvent benchmarks programs can be found as F<eg/ae0.pl> and
		2224	F<eg/ae2.pl> in the AnyEvent distribution, the remaining benchmarks are
		2225	part of the IO::lambda distribution and were used without any changes.
		2226
		2227
		2228	=head1 SIGNALS
		2229
		2230	AnyEvent currently installs handlers for these signals:
		2231
		2232	=over 4
		2233
		2234	=item SIGCHLD
		2235
		2236	A handler for C<SIGCHLD> is installed by AnyEvent's child watcher
		2237	emulation for event loops that do not support them natively. Also, some
		2238	event loops install a similar handler.
		2239
		2240	Additionally, when AnyEvent is loaded and SIGCHLD is set to IGNORE, then
		2241	AnyEvent will reset it to default, to avoid losing child exit statuses.
		2242
		2243	=item SIGPIPE
		2244
		2245	A no-op handler is installed for C<SIGPIPE> when C<$SIG{PIPE}> is C<undef>
		2246	when AnyEvent gets loaded.
		2247
		2248	The rationale for this is that AnyEvent users usually do not really depend
		2249	on SIGPIPE delivery (which is purely an optimisation for shell use, or
		2250	badly-written programs), but C<SIGPIPE> can cause spurious and rare
		2251	program exits as a lot of people do not expect C<SIGPIPE> when writing to
		2252	some random socket.
		2253
		2254	The rationale for installing a no-op handler as opposed to ignoring it is
		2255	that this way, the handler will be restored to defaults on exec.
		2256
		2257	Feel free to install your own handler, or reset it to defaults.
		2258
		2259	=back
		2260
		2261	=cut
		2262
		2263	undef $SIG{CHLD}
		2264	if $SIG{CHLD} eq 'IGNORE';
		2265
		2266	$SIG{PIPE} = sub { }
		2267	unless defined $SIG{PIPE};
		2268
		2269	=head1 RECOMMENDED/OPTIONAL MODULES
		2270
		2271	One of AnyEvent's main goals is to be 100% Pure-Perl(tm): only perl (and
		2272	it's built-in modules) are required to use it.
		2273
		2274	That does not mean that AnyEvent won't take advantage of some additional
		2275	modules if they are installed.
		2276
		2277	This section epxlains which additional modules will be used, and how they
		2278	affect AnyEvent's operetion.
		2279
		2280	=over 4
		2281
		2282	=item L<Async::Interrupt>
		2283
		2284	This slightly arcane module is used to implement fast signal handling: To
		2285	my knowledge, there is no way to do completely race-free and quick
		2286	signal handling in pure perl. To ensure that signals still get
		2287	delivered, AnyEvent will start an interval timer to wake up perl (and
		2288	catch the signals) with some delay (default is 10 seconds, look for
		2289	C<$AnyEvent::MAX_SIGNAL_LATENCY>).
		2290
		2291	If this module is available, then it will be used to implement signal
		2292	catching, which means that signals will not be delayed, and the event loop
		2293	will not be interrupted regularly, which is more efficient (And good for
		2294	battery life on laptops).
		2295
		2296	This affects not just the pure-perl event loop, but also other event loops
		2297	that have no signal handling on their own (e.g. Glib, Tk, Qt).
		2298
		2299	Some event loops (POE, Event, Event::Lib) offer signal watchers natively,
		2300	and either employ their own workarounds (POE) or use AnyEvent's workaround
		2301	(using C<$AnyEvent::MAX_SIGNAL_LATENCY>). Installing L<Async::Interrupt>
		2302	does nothing for those backends.
		2303
		2304	=item L<EV>
		2305
		2306	This module isn't really "optional", as it is simply one of the backend
		2307	event loops that AnyEvent can use. However, it is simply the best event
		2308	loop available in terms of features, speed and stability: It supports
		2309	the AnyEvent API optimally, implements all the watcher types in XS, does
		2310	automatic timer adjustments even when no monotonic clock is available,
		2311	can take avdantage of advanced kernel interfaces such as C<epoll> and
		2312	C<kqueue>, and is the fastest backend I<by far>. You can even embed
		2313	L<Glib>/L<Gtk2> in it (or vice versa, see L<EV::Glib> and L<Glib::EV>).
		2314
		2315	=item L<Guard>
		2316
		2317	The guard module, when used, will be used to implement
		2318	C<AnyEvent::Util::guard>. This speeds up guards considerably (and uses a
		2319	lot less memory), but otherwise doesn't affect guard operation much. It is
		2320	purely used for performance.
		2321
		2322	=item L<JSON> and L<JSON::XS>
		2323
		2324	This module is required when you want to read or write JSON data via
		2325	L<AnyEvent::Handle>. It is also written in pure-perl, but can take
		2326	advantage of the ultra-high-speed L<JSON::XS> module when it is installed.
		2327
		2328	In fact, L<AnyEvent::Handle> will use L<JSON::XS> by default if it is
		2329	installed.
		2330
		2331	=item L<Net::SSLeay>
		2332
		2333	Implementing TLS/SSL in Perl is certainly interesting, but not very
		2334	worthwhile: If this module is installed, then L<AnyEvent::Handle> (with
		2335	the help of L<AnyEvent::TLS>), gains the ability to do TLS/SSL.
		2336
		2337	=item L<Time::HiRes>
		2338
		2339	This module is part of perl since release 5.008. It will be used when the
		2340	chosen event library does not come with a timing source on it's own. The
		2341	pure-perl event loop (L<AnyEvent::Impl::Perl>) will additionally use it to
		2342	try to use a monotonic clock for timing stability.
		2343
		2344	=back
		2345
		2346
		2347	=head1 FORK
		2348
		2349	Most event libraries are not fork-safe. The ones who are usually are
		2350	because they rely on inefficient but fork-safe C<select> or C<poll>
		2351	calls. Only L<EV> is fully fork-aware.
		2352
		2353	If you have to fork, you must either do so I<before> creating your first
		2354	watcher OR you must not use AnyEvent at all in the child OR you must do
		2355	something completely out of the scope of AnyEvent.
		2356
		2357
		2358	=head1 SECURITY CONSIDERATIONS
		2359
		2360	AnyEvent can be forced to load any event model via
		2361	$ENV{PERL_ANYEVENT_MODEL}. While this cannot (to my knowledge) be used to
		2362	execute arbitrary code or directly gain access, it can easily be used to
		2363	make the program hang or malfunction in subtle ways, as AnyEvent watchers
		2364	will not be active when the program uses a different event model than
		2365	specified in the variable.
		2366
		2367	You can make AnyEvent completely ignore this variable by deleting it
		2368	before the first watcher gets created, e.g. with a C<BEGIN> block:
		2369
		2370	BEGIN { delete $ENV{PERL_ANYEVENT_MODEL} }
		2371
		2372	use AnyEvent;
		2373
		2374	Similar considerations apply to $ENV{PERL_ANYEVENT_VERBOSE}, as that can
		2375	be used to probe what backend is used and gain other information (which is
		2376	probably even less useful to an attacker than PERL_ANYEVENT_MODEL), and
		2377	$ENV{PERL_ANYEVENT_STRICT}.
		2378
		2379	Note that AnyEvent will remove I<all> environment variables starting with
		2380	C<PERL_ANYEVENT_> from C<%ENV> when it is loaded while taint mode is
		2381	enabled.
		2382
		2383
		2384	=head1 BUGS
		2385
		2386	Perl 5.8 has numerous memleaks that sometimes hit this module and are hard
		2387	to work around. If you suffer from memleaks, first upgrade to Perl 5.10
		2388	and check wether the leaks still show up. (Perl 5.10.0 has other annoying
		2389	memleaks, such as leaking on C<map> and C<grep> but it is usually not as
		2390	pronounced).
		2391
607		2392
608	=head1 SEE ALSO	2393	=head1 SEE ALSO
609		2394
610	Event modules: L<Coro::Event>, L<Coro>, L<Event>, L<Glib::Event>, L<Glib>.	2395	Utility functions: L<AnyEvent::Util>.
611		2396
612	Implementations: L<AnyEvent::Impl::Coro>, L<AnyEvent::Impl::Event>, L<AnyEvent::Impl::Glib>, L<AnyEvent::Impl::Tk>.	2397	Event modules: L<EV>, L<EV::Glib>, L<Glib::EV>, L<Event>, L<Glib::Event>,
		2398	L<Glib>, L<Tk>, L<Event::Lib>, L<Qt>, L<POE>.
613		2399
614	Nontrivial usage example: L<Net::FCP>.	2400	Implementations: L<AnyEvent::Impl::EV>, L<AnyEvent::Impl::Event>,
		2401	L<AnyEvent::Impl::Glib>, L<AnyEvent::Impl::Tk>, L<AnyEvent::Impl::Perl>,
		2402	L<AnyEvent::Impl::EventLib>, L<AnyEvent::Impl::Qt>,
		2403	L<AnyEvent::Impl::POE>, L<AnyEvent::Impl::IOAsync>.
615		2404
616	=head1	2405	Non-blocking file handles, sockets, TCP clients and
		2406	servers: L<AnyEvent::Handle>, L<AnyEvent::Socket>, L<AnyEvent::TLS>.
		2407
		2408	Asynchronous DNS: L<AnyEvent::DNS>.
		2409
		2410	Coroutine support: L<Coro>, L<Coro::AnyEvent>, L<Coro::EV>,
		2411	L<Coro::Event>,
		2412
		2413	Nontrivial usage examples: L<AnyEvent::GPSD>, L<AnyEvent::XMPP>,
		2414	L<AnyEvent::HTTP>.
		2415
		2416
		2417	=head1 AUTHOR
		2418
		2419	Marc Lehmann <schmorp@schmorp.de>
		2420	http://home.schmorp.de/
617		2421
618	=cut	2422	=cut
619		2423
620	1	2424	1
621		2425

Diff Legend

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