ViewVC Help

View File | Revision Log | Show Annotations | Download File

/cvs/AnyEvent/lib/AnyEvent.pm

Comparing AnyEvent/lib/AnyEvent.pm (file contents):
Revision 1.1 by root, Wed Apr 27 01:26:44 2005 UTC vs.
Revision 1.134 by root, Sun May 25 04:44:04 2008 UTC

1	=head1 NAME	1	=head1 => NAME
2		2
3	AnyEvent - ???	3	AnyEvent - provide framework for multiple event loops
		4
		5	EV, Event, Glib, Tk, Perl, Event::Lib, Qt, POE - various supported event loops
4		6
5	=head1 SYNOPSIS	7	=head1 SYNOPSIS
6		8
		9	use AnyEvent;
		10
		11	my $w = AnyEvent->io (fh => $fh, poll => "r|w", cb => sub {
		12	...
		13	});
		14
		15	my $w = AnyEvent->timer (after => $seconds, cb => sub {
		16	...
		17	});
		18
		19	my $w = AnyEvent->condvar; # stores whether a condition was flagged
		20	$w->send; # wake up current and all future recv's
		21	$w->recv; # enters "main loop" till $condvar gets ->send
		22
		23	=head1 WHY YOU SHOULD USE THIS MODULE (OR NOT)
		24
		25	Glib, POE, IO::Async, Event... CPAN offers event models by the dozen
		26	nowadays. So what is different about AnyEvent?
		27
		28	Executive Summary: AnyEvent is I<compatible>, AnyEvent is I<free of
		29	policy> and AnyEvent is I<small and efficient>.
		30
		31	First and foremost, I<AnyEvent is not an event model> itself, it only
		32	interfaces to whatever event model the main program happens to use in a
		33	pragmatic way. For event models and certain classes of immortals alike,
		34	the statement "there can only be one" is a bitter reality: In general,
		35	only one event loop can be active at the same time in a process. AnyEvent
		36	helps hiding the differences between those event loops.
		37
		38	The goal of AnyEvent is to offer module authors the ability to do event
		39	programming (waiting for I/O or timer events) without subscribing to a
		40	religion, a way of living, and most importantly: without forcing your
		41	module users into the same thing by forcing them to use the same event
		42	model you use.
		43
		44	For modules like POE or IO::Async (which is a total misnomer as it is
		45	actually doing all I/O I<synchronously>...), using them in your module is
		46	like joining a cult: After you joined, you are dependent on them and you
		47	cannot use anything else, as it is simply incompatible to everything that
		48	isn't itself. What's worse, all the potential users of your module are
		49	I<also> forced to use the same event loop you use.
		50
		51	AnyEvent is different: AnyEvent + POE works fine. AnyEvent + Glib works
		52	fine. AnyEvent + Tk works fine etc. etc. but none of these work together
		53	with the rest: POE + IO::Async? no go. Tk + Event? no go. Again: if
		54	your module uses one of those, every user of your module has to use it,
		55	too. But if your module uses AnyEvent, it works transparently with all
		56	event models it supports (including stuff like POE and IO::Async, as long
		57	as those use one of the supported event loops. It is trivial to add new
		58	event loops to AnyEvent, too, so it is future-proof).
		59
		60	In addition to being free of having to use I<the one and only true event
		61	model>, AnyEvent also is free of bloat and policy: with POE or similar
		62	modules, you get an enormous amount of code and strict rules you have to
		63	follow. AnyEvent, on the other hand, is lean and up to the point, by only
		64	offering the functionality that is necessary, in as thin as a wrapper as
		65	technically possible.
		66
		67	Of course, if you want lots of policy (this can arguably be somewhat
		68	useful) and you want to force your users to use the one and only event
		69	model, you should I<not> use this module.
		70
7	=head1 DESCRIPTION	71	=head1 DESCRIPTION
8		72
		73	L<AnyEvent> provides an identical interface to multiple event loops. This
		74	allows module authors to utilise an event loop without forcing module
		75	users to use the same event loop (as only a single event loop can coexist
		76	peacefully at any one time).
		77
		78	The interface itself is vaguely similar, but not identical to the L<Event>
		79	module.
		80
		81	During the first call of any watcher-creation method, the module tries
		82	to detect the currently loaded event loop by probing whether one of the
		83	following modules is already loaded: L<EV>,
		84	L<Event>, L<Glib>, L<AnyEvent::Impl::Perl>, L<Tk>, L<Event::Lib>, L<Qt>,
		85	L<POE>. The first one found is used. If none are found, the module tries
		86	to load these modules (excluding Tk, Event::Lib, Qt and POE as the pure perl
		87	adaptor should always succeed) in the order given. The first one that can
		88	be successfully loaded will be used. If, after this, still none could be
		89	found, AnyEvent will fall back to a pure-perl event loop, which is not
		90	very efficient, but should work everywhere.
		91
		92	Because AnyEvent first checks for modules that are already loaded, loading
		93	an event model explicitly before first using AnyEvent will likely make
		94	that model the default. For example:
		95
		96	use Tk;
		97	use AnyEvent;
		98
		99	# .. AnyEvent will likely default to Tk
		100
		101	The I<likely> means that, if any module loads another event model and
		102	starts using it, all bets are off. Maybe you should tell their authors to
		103	use AnyEvent so their modules work together with others seamlessly...
		104
		105	The pure-perl implementation of AnyEvent is called
		106	C<AnyEvent::Impl::Perl>. Like other event modules you can load it
		107	explicitly.
		108
		109	=head1 WATCHERS
		110
		111	AnyEvent has the central concept of a I<watcher>, which is an object that
		112	stores relevant data for each kind of event you are waiting for, such as
		113	the callback to call, the file handle to watch, etc.
		114
		115	These watchers are normal Perl objects with normal Perl lifetime. After
		116	creating a watcher it will immediately "watch" for events and invoke the
		117	callback when the event occurs (of course, only when the event model
		118	is in control).
		119
		120	To disable the watcher you have to destroy it (e.g. by setting the
		121	variable you store it in to C<undef> or otherwise deleting all references
		122	to it).
		123
		124	All watchers are created by calling a method on the C<AnyEvent> class.
		125
		126	Many watchers either are used with "recursion" (repeating timers for
		127	example), or need to refer to their watcher object in other ways.
		128
		129	An any way to achieve that is this pattern:
		130
		131	my $w; $w = AnyEvent->type (arg => value ..., cb => sub {
		132	# you can use $w here, for example to undef it
		133	undef $w;
		134	});
		135
		136	Note that C<my $w; $w => combination. This is necessary because in Perl,
		137	my variables are only visible after the statement in which they are
		138	declared.
		139
		140	=head2 I/O WATCHERS
		141
		142	You can create an I/O watcher by calling the C<< AnyEvent->io >> method
		143	with the following mandatory key-value pairs as arguments:
		144
		145	C<fh> the Perl I<file handle> (I<not> file descriptor) to watch
		146	for events. C<poll> must be a string that is either C<r> or C<w>,
		147	which creates a watcher waiting for "r"eadable or "w"ritable events,
		148	respectively. C<cb> is the callback to invoke each time the file handle
		149	becomes ready.
		150
		151	Although the callback might get passed parameters, their value and
		152	presence is undefined and you cannot rely on them. Portable AnyEvent
		153	callbacks cannot use arguments passed to I/O watcher callbacks.
		154
		155	The I/O watcher might use the underlying file descriptor or a copy of it.
		156	You must not close a file handle as long as any watcher is active on the
		157	underlying file descriptor.
		158
		159	Some event loops issue spurious readyness notifications, so you should
		160	always use non-blocking calls when reading/writing from/to your file
		161	handles.
		162
		163	Example:
		164
		165	# wait for readability of STDIN, then read a line and disable the watcher
		166	my $w; $w = AnyEvent->io (fh => \*STDIN, poll => 'r', cb => sub {
		167	chomp (my $input = <STDIN>);
		168	warn "read: $input\n";
		169	undef $w;
		170	});
		171
		172	=head2 TIME WATCHERS
		173
		174	You can create a time watcher by calling the C<< AnyEvent->timer >>
		175	method with the following mandatory arguments:
		176
		177	C<after> specifies after how many seconds (fractional values are
		178	supported) the callback should be invoked. C<cb> is the callback to invoke
		179	in that case.
		180
		181	Although the callback might get passed parameters, their value and
		182	presence is undefined and you cannot rely on them. Portable AnyEvent
		183	callbacks cannot use arguments passed to time watcher callbacks.
		184
		185	The timer callback will be invoked at most once: if you want a repeating
		186	timer you have to create a new watcher (this is a limitation by both Tk
		187	and Glib).
		188
		189	Example:
		190
		191	# fire an event after 7.7 seconds
		192	my $w = AnyEvent->timer (after => 7.7, cb => sub {
		193	warn "timeout\n";
		194	});
		195
		196	# to cancel the timer:
		197	undef $w;
		198
		199	Example 2:
		200
		201	# fire an event after 0.5 seconds, then roughly every second
		202	my $w;
		203
		204	my $cb = sub {
		205	# cancel the old timer while creating a new one
		206	$w = AnyEvent->timer (after => 1, cb => $cb);
		207	};
		208
		209	# start the "loop" by creating the first watcher
		210	$w = AnyEvent->timer (after => 0.5, cb => $cb);
		211
		212	=head3 TIMING ISSUES
		213
		214	There are two ways to handle timers: based on real time (relative, "fire
		215	in 10 seconds") and based on wallclock time (absolute, "fire at 12
		216	o'clock").
		217
		218	While most event loops expect timers to specified in a relative way, they
		219	use absolute time internally. This makes a difference when your clock
		220	"jumps", for example, when ntp decides to set your clock backwards from
		221	the wrong date of 2014-01-01 to 2008-01-01, a watcher that is supposed to
		222	fire "after" a second might actually take six years to finally fire.
		223
		224	AnyEvent cannot compensate for this. The only event loop that is conscious
		225	about these issues is L<EV>, which offers both relative (ev_timer, based
		226	on true relative time) and absolute (ev_periodic, based on wallclock time)
		227	timers.
		228
		229	AnyEvent always prefers relative timers, if available, matching the
		230	AnyEvent API.
		231
		232	=head2 SIGNAL WATCHERS
		233
		234	You can watch for signals using a signal watcher, C<signal> is the signal
		235	I<name> without any C<SIG> prefix, C<cb> is the Perl callback to
		236	be invoked whenever a signal occurs.
		237
		238	Although the callback might get passed parameters, their value and
		239	presence is undefined and you cannot rely on them. Portable AnyEvent
		240	callbacks cannot use arguments passed to signal watcher callbacks.
		241
		242	Multiple signal occurrences can be clumped together into one callback
		243	invocation, and callback invocation will be synchronous. Synchronous means
		244	that it might take a while until the signal gets handled by the process,
		245	but it is guaranteed not to interrupt any other callbacks.
		246
		247	The main advantage of using these watchers is that you can share a signal
		248	between multiple watchers.
		249
		250	This watcher might use C<%SIG>, so programs overwriting those signals
		251	directly will likely not work correctly.
		252
		253	Example: exit on SIGINT
		254
		255	my $w = AnyEvent->signal (signal => "INT", cb => sub { exit 1 });
		256
		257	=head2 CHILD PROCESS WATCHERS
		258
		259	You can also watch on a child process exit and catch its exit status.
		260
		261	The child process is specified by the C<pid> argument (if set to C<0>, it
		262	watches for any child process exit). The watcher will trigger as often
		263	as status change for the child are received. This works by installing a
		264	signal handler for C<SIGCHLD>. The callback will be called with the pid
		265	and exit status (as returned by waitpid), so unlike other watcher types,
		266	you I<can> rely on child watcher callback arguments.
		267
		268	There is a slight catch to child watchers, however: you usually start them
		269	I<after> the child process was created, and this means the process could
		270	have exited already (and no SIGCHLD will be sent anymore).
		271
		272	Not all event models handle this correctly (POE doesn't), but even for
		273	event models that I<do> handle this correctly, they usually need to be
		274	loaded before the process exits (i.e. before you fork in the first place).
		275
		276	This means you cannot create a child watcher as the very first thing in an
		277	AnyEvent program, you I<have> to create at least one watcher before you
		278	C<fork> the child (alternatively, you can call C<AnyEvent::detect>).
		279
		280	Example: fork a process and wait for it
		281
		282	my $done = AnyEvent->condvar;
		283
		284	my $pid = fork or exit 5;
		285
		286	my $w = AnyEvent->child (
		287	pid => $pid,
		288	cb => sub {
		289	my ($pid, $status) = @_;
		290	warn "pid $pid exited with status $status";
		291	$done->send;
		292	},
		293	);
		294
		295	# do something else, then wait for process exit
		296	$done->recv;
		297
		298	=head2 CONDITION VARIABLES
		299
		300	If you are familiar with some event loops you will know that all of them
		301	require you to run some blocking "loop", "run" or similar function that
		302	will actively watch for new events and call your callbacks.
		303
		304	AnyEvent is different, it expects somebody else to run the event loop and
		305	will only block when necessary (usually when told by the user).
		306
		307	The instrument to do that is called a "condition variable", so called
		308	because they represent a condition that must become true.
		309
		310	Condition variables can be created by calling the C<< AnyEvent->condvar
		311	>> method, usually without arguments. The only argument pair allowed is
		312	C<cb>, which specifies a callback to be called when the condition variable
		313	becomes true.
		314
		315	After creation, the condition variable is "false" until it becomes "true"
		316	by calling the C<send> method (or calling the condition variable as if it
		317	were a callback).
		318
		319	Condition variables are similar to callbacks, except that you can
		320	optionally wait for them. They can also be called merge points - points
		321	in time where multiple outstanding events have been processed. And yet
		322	another way to call them is transactions - each condition variable can be
		323	used to represent a transaction, which finishes at some point and delivers
		324	a result.
		325
		326	Condition variables are very useful to signal that something has finished,
		327	for example, if you write a module that does asynchronous http requests,
		328	then a condition variable would be the ideal candidate to signal the
		329	availability of results. The user can either act when the callback is
		330	called or can synchronously C<< ->recv >> for the results.
		331
		332	You can also use them to simulate traditional event loops - for example,
		333	you can block your main program until an event occurs - for example, you
		334	could C<< ->recv >> in your main program until the user clicks the Quit
		335	button of your app, which would C<< ->send >> the "quit" event.
		336
		337	Note that condition variables recurse into the event loop - if you have
		338	two pieces of code that call C<< ->recv >> in a round-robin fashion, you
		339	lose. Therefore, condition variables are good to export to your caller, but
		340	you should avoid making a blocking wait yourself, at least in callbacks,
		341	as this asks for trouble.
		342
		343	Condition variables are represented by hash refs in perl, and the keys
		344	used by AnyEvent itself are all named C<_ae_XXX> to make subclassing
		345	easy (it is often useful to build your own transaction class on top of
		346	AnyEvent). To subclass, use C<AnyEvent::CondVar> as base class and call
		347	it's C<new> method in your own C<new> method.
		348
		349	There are two "sides" to a condition variable - the "producer side" which
		350	eventually calls C<< -> send >>, and the "consumer side", which waits
		351	for the send to occur.
		352
		353	Example: wait for a timer.
		354
		355	# wait till the result is ready
		356	my $result_ready = AnyEvent->condvar;
		357
		358	# do something such as adding a timer
		359	# or socket watcher the calls $result_ready->send
		360	# when the "result" is ready.
		361	# in this case, we simply use a timer:
		362	my $w = AnyEvent->timer (
		363	after => 1,
		364	cb => sub { $result_ready->send },
		365	);
		366
		367	# this "blocks" (while handling events) till the callback
		368	# calls send
		369	$result_ready->recv;
		370
		371	Example: wait for a timer, but take advantage of the fact that
		372	condition variables are also code references.
		373
		374	my $done = AnyEvent->condvar;
		375	my $delay = AnyEvent->timer (after => 5, cb => $done);
		376	$done->recv;
		377
		378	=head3 METHODS FOR PRODUCERS
		379
		380	These methods should only be used by the producing side, i.e. the
		381	code/module that eventually sends the signal. Note that it is also
		382	the producer side which creates the condvar in most cases, but it isn't
		383	uncommon for the consumer to create it as well.
		384
9	=over 4	385	=over 4
10		386
		387	=item $cv->send (...)
		388
		389	Flag the condition as ready - a running C<< ->recv >> and all further
		390	calls to C<recv> will (eventually) return after this method has been
		391	called. If nobody is waiting the send will be remembered.
		392
		393	If a callback has been set on the condition variable, it is called
		394	immediately from within send.
		395
		396	Any arguments passed to the C<send> call will be returned by all
		397	future C<< ->recv >> calls.
		398
		399	Condition variables are overloaded so one can call them directly (as a
		400	code reference). Calling them directly is the same as calling C<send>.
		401
		402	=item $cv->croak ($error)
		403
		404	Similar to send, but causes all call's to C<< ->recv >> to invoke
		405	C<Carp::croak> with the given error message/object/scalar.
		406
		407	This can be used to signal any errors to the condition variable
		408	user/consumer.
		409
		410	=item $cv->begin ([group callback])
		411
		412	=item $cv->end
		413
		414	These two methods are EXPERIMENTAL and MIGHT CHANGE.
		415
		416	These two methods can be used to combine many transactions/events into
		417	one. For example, a function that pings many hosts in parallel might want
		418	to use a condition variable for the whole process.
		419
		420	Every call to C<< ->begin >> will increment a counter, and every call to
		421	C<< ->end >> will decrement it. If the counter reaches C<0> in C<< ->end
		422	>>, the (last) callback passed to C<begin> will be executed. That callback
		423	is I<supposed> to call C<< ->send >>, but that is not required. If no
		424	callback was set, C<send> will be called without any arguments.
		425
		426	Let's clarify this with the ping example:
		427
		428	my $cv = AnyEvent->condvar;
		429
		430	my %result;
		431	$cv->begin (sub { $cv->send (\%result) });
		432
		433	for my $host (@list_of_hosts) {
		434	$cv->begin;
		435	ping_host_then_call_callback $host, sub {
		436	$result{$host} = ...;
		437	$cv->end;
		438	};
		439	}
		440
		441	$cv->end;
		442
		443	This code fragment supposedly pings a number of hosts and calls
		444	C<send> after results for all then have have been gathered - in any
		445	order. To achieve this, the code issues a call to C<begin> when it starts
		446	each ping request and calls C<end> when it has received some result for
		447	it. Since C<begin> and C<end> only maintain a counter, the order in which
		448	results arrive is not relevant.
		449
		450	There is an additional bracketing call to C<begin> and C<end> outside the
		451	loop, which serves two important purposes: first, it sets the callback
		452	to be called once the counter reaches C<0>, and second, it ensures that
		453	C<send> is called even when C<no> hosts are being pinged (the loop
		454	doesn't execute once).
		455
		456	This is the general pattern when you "fan out" into multiple subrequests:
		457	use an outer C<begin>/C<end> pair to set the callback and ensure C<end>
		458	is called at least once, and then, for each subrequest you start, call
		459	C<begin> and for each subrequest you finish, call C<end>.
		460
		461	=back
		462
		463	=head3 METHODS FOR CONSUMERS
		464
		465	These methods should only be used by the consuming side, i.e. the
		466	code awaits the condition.
		467
		468	=over 4
		469
		470	=item $cv->recv
		471
		472	Wait (blocking if necessary) until the C<< ->send >> or C<< ->croak
		473	>> methods have been called on c<$cv>, while servicing other watchers
		474	normally.
		475
		476	You can only wait once on a condition - additional calls are valid but
		477	will return immediately.
		478
		479	If an error condition has been set by calling C<< ->croak >>, then this
		480	function will call C<croak>.
		481
		482	In list context, all parameters passed to C<send> will be returned,
		483	in scalar context only the first one will be returned.
		484
		485	Not all event models support a blocking wait - some die in that case
		486	(programs might want to do that to stay interactive), so I<if you are
		487	using this from a module, never require a blocking wait>, but let the
		488	caller decide whether the call will block or not (for example, by coupling
		489	condition variables with some kind of request results and supporting
		490	callbacks so the caller knows that getting the result will not block,
		491	while still supporting blocking waits if the caller so desires).
		492
		493	Another reason I<never> to C<< ->recv >> in a module is that you cannot
		494	sensibly have two C<< ->recv >>'s in parallel, as that would require
		495	multiple interpreters or coroutines/threads, none of which C<AnyEvent>
		496	can supply.
		497
		498	The L<Coro> module, however, I<can> and I<does> supply coroutines and, in
		499	fact, L<Coro::AnyEvent> replaces AnyEvent's condvars by coroutine-safe
		500	versions and also integrates coroutines into AnyEvent, making blocking
		501	C<< ->recv >> calls perfectly safe as long as they are done from another
		502	coroutine (one that doesn't run the event loop).
		503
		504	You can ensure that C<< -recv >> never blocks by setting a callback and
		505	only calling C<< ->recv >> from within that callback (or at a later
		506	time). This will work even when the event loop does not support blocking
		507	waits otherwise.
		508
		509	=item $bool = $cv->ready
		510
		511	Returns true when the condition is "true", i.e. whether C<send> or
		512	C<croak> have been called.
		513
		514	=item $cb = $cv->cb ([new callback])
		515
		516	This is a mutator function that returns the callback set and optionally
		517	replaces it before doing so.
		518
		519	The callback will be called when the condition becomes "true", i.e. when
		520	C<send> or C<croak> are called. Calling C<recv> inside the callback
		521	or at any later time is guaranteed not to block.
		522
		523	=back
		524
		525	=head1 GLOBAL VARIABLES AND FUNCTIONS
		526
		527	=over 4
		528
		529	=item $AnyEvent::MODEL
		530
		531	Contains C<undef> until the first watcher is being created. Then it
		532	contains the event model that is being used, which is the name of the
		533	Perl class implementing the model. This class is usually one of the
		534	C<AnyEvent::Impl:xxx> modules, but can be any other class in the case
		535	AnyEvent has been extended at runtime (e.g. in I<rxvt-unicode>).
		536
		537	The known classes so far are:
		538
		539	AnyEvent::Impl::EV based on EV (an interface to libev, best choice).
		540	AnyEvent::Impl::Event based on Event, second best choice.
		541	AnyEvent::Impl::Perl pure-perl implementation, fast and portable.
		542	AnyEvent::Impl::Glib based on Glib, third-best choice.
		543	AnyEvent::Impl::Tk based on Tk, very bad choice.
		544	AnyEvent::Impl::Qt based on Qt, cannot be autoprobed (see its docs).
		545	AnyEvent::Impl::EventLib based on Event::Lib, leaks memory and worse.
		546	AnyEvent::Impl::POE based on POE, not generic enough for full support.
		547
		548	There is no support for WxWidgets, as WxWidgets has no support for
		549	watching file handles. However, you can use WxWidgets through the
		550	POE Adaptor, as POE has a Wx backend that simply polls 20 times per
		551	second, which was considered to be too horrible to even consider for
		552	AnyEvent. Likewise, other POE backends can be used by AnyEvent by using
		553	it's adaptor.
		554
		555	AnyEvent knows about L<Prima> and L<Wx> and will try to use L<POE> when
		556	autodetecting them.
		557
		558	=item AnyEvent::detect
		559
		560	Returns C<$AnyEvent::MODEL>, forcing autodetection of the event model
		561	if necessary. You should only call this function right before you would
		562	have created an AnyEvent watcher anyway, that is, as late as possible at
		563	runtime.
		564
		565	=item $guard = AnyEvent::post_detect { BLOCK }
		566
		567	Arranges for the code block to be executed as soon as the event model is
		568	autodetected (or immediately if this has already happened).
		569
		570	If called in scalar or list context, then it creates and returns an object
		571	that automatically removes the callback again when it is destroyed. See
		572	L<Coro::BDB> for a case where this is useful.
		573
		574	=item @AnyEvent::post_detect
		575
		576	If there are any code references in this array (you can C<push> to it
		577	before or after loading AnyEvent), then they will called directly after
		578	the event loop has been chosen.
		579
		580	You should check C<$AnyEvent::MODEL> before adding to this array, though:
		581	if it contains a true value then the event loop has already been detected,
		582	and the array will be ignored.
		583
		584	Best use C<AnyEvent::post_detect { BLOCK }> instead.
		585
		586	=back
		587
		588	=head1 WHAT TO DO IN A MODULE
		589
		590	As a module author, you should C<use AnyEvent> and call AnyEvent methods
		591	freely, but you should not load a specific event module or rely on it.
		592
		593	Be careful when you create watchers in the module body - AnyEvent will
		594	decide which event module to use as soon as the first method is called, so
		595	by calling AnyEvent in your module body you force the user of your module
		596	to load the event module first.
		597
		598	Never call C<< ->recv >> on a condition variable unless you I<know> that
		599	the C<< ->send >> method has been called on it already. This is
		600	because it will stall the whole program, and the whole point of using
		601	events is to stay interactive.
		602
		603	It is fine, however, to call C<< ->recv >> when the user of your module
		604	requests it (i.e. if you create a http request object ad have a method
		605	called C<results> that returns the results, it should call C<< ->recv >>
		606	freely, as the user of your module knows what she is doing. always).
		607
		608	=head1 WHAT TO DO IN THE MAIN PROGRAM
		609
		610	There will always be a single main program - the only place that should
		611	dictate which event model to use.
		612
		613	If it doesn't care, it can just "use AnyEvent" and use it itself, or not
		614	do anything special (it does not need to be event-based) and let AnyEvent
		615	decide which implementation to chose if some module relies on it.
		616
		617	If the main program relies on a specific event model - for example, in
		618	Gtk2 programs you have to rely on the Glib module - you should load the
		619	event module before loading AnyEvent or any module that uses it: generally
		620	speaking, you should load it as early as possible. The reason is that
		621	modules might create watchers when they are loaded, and AnyEvent will
		622	decide on the event model to use as soon as it creates watchers, and it
		623	might chose the wrong one unless you load the correct one yourself.
		624
		625	You can chose to use a pure-perl implementation by loading the
		626	C<AnyEvent::Impl::Perl> module, which gives you similar behaviour
		627	everywhere, but letting AnyEvent chose the model is generally better.
		628
		629	=head2 MAINLOOP EMULATION
		630
		631	Sometimes (often for short test scripts, or even standalone programs who
		632	only want to use AnyEvent), you do not want to run a specific event loop.
		633
		634	In that case, you can use a condition variable like this:
		635
		636	AnyEvent->condvar->recv;
		637
		638	This has the effect of entering the event loop and looping forever.
		639
		640	Note that usually your program has some exit condition, in which case
		641	it is better to use the "traditional" approach of storing a condition
		642	variable somewhere, waiting for it, and sending it when the program should
		643	exit cleanly.
		644
		645
		646	=head1 OTHER MODULES
		647
		648	The following is a non-exhaustive list of additional modules that use
		649	AnyEvent and can therefore be mixed easily with other AnyEvent modules
		650	in the same program. Some of the modules come with AnyEvent, some are
		651	available via CPAN.
		652
		653	=over 4
		654
		655	=item L<AnyEvent::Util>
		656
		657	Contains various utility functions that replace often-used but blocking
		658	functions such as C<inet_aton> by event-/callback-based versions.
		659
		660	=item L<AnyEvent::Handle>
		661
		662	Provide read and write buffers and manages watchers for reads and writes.
		663
		664	=item L<AnyEvent::Socket>
		665
		666	Provides various utility functions for (internet protocol) sockets,
		667	addresses and name resolution. Also functions to create non-blocking tcp
		668	connections or tcp servers, with IPv6 and SRV record support and more.
		669
		670	=item L<AnyEvent::DNS>
		671
		672	Provides rich asynchronous DNS resolver capabilities.
		673
		674	=item L<AnyEvent::HTTPD>
		675
		676	Provides a simple web application server framework.
		677
		678	=item L<AnyEvent::FastPing>
		679
		680	The fastest ping in the west.
		681
		682	=item L<Net::IRC3>
		683
		684	AnyEvent based IRC client module family.
		685
		686	=item L<Net::XMPP2>
		687
		688	AnyEvent based XMPP (Jabber protocol) module family.
		689
		690	=item L<Net::FCP>
		691
		692	AnyEvent-based implementation of the Freenet Client Protocol, birthplace
		693	of AnyEvent.
		694
		695	=item L<Event::ExecFlow>
		696
		697	High level API for event-based execution flow control.
		698
		699	=item L<Coro>
		700
		701	Has special support for AnyEvent via L<Coro::AnyEvent>.
		702
		703	=item L<AnyEvent::AIO>, L<IO::AIO>
		704
		705	Truly asynchronous I/O, should be in the toolbox of every event
		706	programmer. AnyEvent::AIO transparently fuses IO::AIO and AnyEvent
		707	together.
		708
		709	=item L<AnyEvent::BDB>, L<BDB>
		710
		711	Truly asynchronous Berkeley DB access. AnyEvent::AIO transparently fuses
		712	IO::AIO and AnyEvent together.
		713
		714	=item L<IO::Lambda>
		715
		716	The lambda approach to I/O - don't ask, look there. Can use AnyEvent.
		717
		718	=back
		719
11	=cut	720	=cut
12		721
13	package AnyEvent;	722	package AnyEvent;
14		723
		724	no warnings;
		725	use strict;
		726
15	use Carp;	727	use Carp;
16		728
17	$VERSION = 0.1;	729	our $VERSION = '4.03';
		730	our $MODEL;
18		731
19	no warnings;	732	our $AUTOLOAD;
		733	our @ISA;
		734
		735	our $verbose = $ENV{PERL_ANYEVENT_VERBOSE}*1;
		736
		737	our @REGISTRY;
		738
		739	our %PROTOCOL; # (ipv4|ipv6) => (1|2)
		740
		741	{
		742	my $idx;
		743	$PROTOCOL{$_} = ++$idx
		744	for split /\s*,\s*/, $ENV{PERL_ANYEVENT_PROTOCOLS} || "ipv4,ipv6";
		745	}
20		746
21	my @models = (	747	my @models = (
22	[Coro => Coro::Event::],	748	[EV:: => AnyEvent::Impl::EV::],
23	[Event => Event::],	749	[Event:: => AnyEvent::Impl::Event::],
24	[Glib => Glib::],	750	[Tk:: => AnyEvent::Impl::Tk::],
25	[Tk => Tk::],	751	[Wx:: => AnyEvent::Impl::POE::],
		752	[Prima:: => AnyEvent::Impl::POE::],
		753	[AnyEvent::Impl::Perl:: => AnyEvent::Impl::Perl::],
		754	# everything below here will not be autoprobed as the pureperl backend should work everywhere
		755	[Glib:: => AnyEvent::Impl::Glib::],
		756	[Event::Lib:: => AnyEvent::Impl::EventLib::], # too buggy
		757	[Qt:: => AnyEvent::Impl::Qt::], # requires special main program
		758	[POE::Kernel:: => AnyEvent::Impl::POE::], # lasciate ogni speranza
26	);	759	);
27		760
		761	our %method = map +($_ => 1), qw(io timer signal child condvar one_event DESTROY);
		762
		763	our @post_detect;
		764
		765	sub post_detect(&) {
		766	my ($cb) = @_;
		767
		768	if ($MODEL) {
		769	$cb->();
		770
		771	1
		772	} else {
		773	push @post_detect, $cb;
		774
		775	defined wantarray
		776	? bless \$cb, "AnyEvent::Util::PostDetect"
		777	: ()
		778	}
		779	}
		780
		781	sub AnyEvent::Util::PostDetect::DESTROY {
		782	@post_detect = grep $_ != ${$_[0]}, @post_detect;
		783	}
		784
		785	sub detect() {
		786	unless ($MODEL) {
		787	no strict 'refs';
		788
		789	if ($ENV{PERL_ANYEVENT_MODEL} =~ /^([a-zA-Z]+)$/) {
		790	my $model = "AnyEvent::Impl::$1";
		791	if (eval "require $model") {
		792	$MODEL = $model;
		793	warn "AnyEvent: loaded model '$model' (forced by \$PERL_ANYEVENT_MODEL), using it.\n" if $verbose > 1;
		794	} else {
		795	warn "AnyEvent: unable to load model '$model' (from \$PERL_ANYEVENT_MODEL):\n$@" if $verbose;
		796	}
		797	}
		798
		799	# check for already loaded models
		800	unless ($MODEL) {
		801	for (@REGISTRY, @models) {
		802	my ($package, $model) = @$_;
		803	if (${"$package\::VERSION"} > 0) {
		804	if (eval "require $model") {
		805	$MODEL = $model;
		806	warn "AnyEvent: autodetected model '$model', using it.\n" if $verbose > 1;
		807	last;
		808	}
		809	}
		810	}
		811
		812	unless ($MODEL) {
		813	# try to load a model
		814
		815	for (@REGISTRY, @models) {
		816	my ($package, $model) = @$_;
		817	if (eval "require $package"
		818	and ${"$package\::VERSION"} > 0
		819	and eval "require $model") {
		820	$MODEL = $model;
		821	warn "AnyEvent: autoprobed model '$model', using it.\n" if $verbose > 1;
		822	last;
		823	}
		824	}
		825
		826	$MODEL
		827	or die "No event module selected for AnyEvent and autodetect failed. Install any one of these modules: EV, Event or Glib.";
		828	}
		829	}
		830
		831	unshift @ISA, $MODEL;
		832	push @{"$MODEL\::ISA"}, "AnyEvent::Base";
		833
		834	(shift @post_detect)->() while @post_detect;
		835	}
		836
		837	$MODEL
		838	}
		839
28	sub AUTOLOAD {	840	sub AUTOLOAD {
29	$AUTOLOAD =~ s/.*://;	841	(my $func = $AUTOLOAD) =~ s/.*://;
30		842
31	for (@models) {	843	$method{$func}
32	my ($model, $package) = @$_;	844	or croak "$func: not a valid method for AnyEvent objects";
33	if (defined ${"$package\::VERSION"}) {	845
34	$EVENT = "AnyEvent::Impl::$model";	846	detect unless $MODEL;
35	eval "require $EVENT"; die if $@;	847
36	goto &{"$EVENT\::$AUTOLOAD"};	848	my $class = shift;
37	}	849	$class->$func (@_);
		850	}
		851
		852	package AnyEvent::Base;
		853
		854	# default implementation for ->condvar
		855
		856	sub condvar {
		857	bless { @_ == 3 ? (_ae_cb => $_[2]) : () }, AnyEvent::CondVar::
		858	}
		859
		860	# default implementation for ->signal
		861
		862	our %SIG_CB;
		863
		864	sub signal {
		865	my (undef, %arg) = @_;
		866
		867	my $signal = uc $arg{signal}
		868	or Carp::croak "required option 'signal' is missing";
		869
		870	$SIG_CB{$signal}{$arg{cb}} = $arg{cb};
		871	$SIG{$signal} ||= sub {
		872	$_->() for values %{ $SIG_CB{$signal} || {} };
		873	};
		874
		875	bless [$signal, $arg{cb}], "AnyEvent::Base::Signal"
		876	}
		877
		878	sub AnyEvent::Base::Signal::DESTROY {
		879	my ($signal, $cb) = @{$_[0]};
		880
		881	delete $SIG_CB{$signal}{$cb};
		882
		883	$SIG{$signal} = 'DEFAULT' unless keys %{ $SIG_CB{$signal} };
		884	}
		885
		886	# default implementation for ->child
		887
		888	our %PID_CB;
		889	our $CHLD_W;
		890	our $CHLD_DELAY_W;
		891	our $PID_IDLE;
		892	our $WNOHANG;
		893
		894	sub _child_wait {
		895	while (0 < (my $pid = waitpid -1, $WNOHANG)) {
		896	$_->($pid, $?) for (values %{ $PID_CB{$pid} || {} }),
		897	(values %{ $PID_CB{0} || {} });
38	}	898	}
39		899
40	for (@models) {	900	undef $PID_IDLE;
41	my ($model, $package) = @$_;	901	}
42	$EVENT = "AnyEvent::Impl::$model";	902
43	if (eval "require $EVENT") {	903	sub _sigchld {
44	goto &{"$EVENT\::$AUTOLOAD"};	904	# make sure we deliver these changes "synchronous" with the event loop.
45	}	905	$CHLD_DELAY_W ||= AnyEvent->timer (after => 0, cb => sub {
		906	undef $CHLD_DELAY_W;
		907	&_child_wait;
		908	});
		909	}
		910
		911	sub child {
		912	my (undef, %arg) = @_;
		913
		914	defined (my $pid = $arg{pid} + 0)
		915	or Carp::croak "required option 'pid' is missing";
		916
		917	$PID_CB{$pid}{$arg{cb}} = $arg{cb};
		918
		919	unless ($WNOHANG) {
		920	$WNOHANG = eval { require POSIX; &POSIX::WNOHANG } || 1;
46	}	921	}
47		922
48	die "No event module selected for AnyEvent and autodetect failed. Install any of these: Coro, Event, Glib or Tk.";	923	unless ($CHLD_W) {
49	}	924	$CHLD_W = AnyEvent->signal (signal => 'CHLD', cb => \&_sigchld);
		925	# child could be a zombie already, so make at least one round
		926	&_sigchld;
		927	}
50		928
51	1;	929	bless [$pid, $arg{cb}], "AnyEvent::Base::Child"
		930	}
52		931
		932	sub AnyEvent::Base::Child::DESTROY {
		933	my ($pid, $cb) = @{$_[0]};
		934
		935	delete $PID_CB{$pid}{$cb};
		936	delete $PID_CB{$pid} unless keys %{ $PID_CB{$pid} };
		937
		938	undef $CHLD_W unless keys %PID_CB;
		939	}
		940
		941	package AnyEvent::CondVar;
		942
		943	our @ISA = AnyEvent::CondVar::Base::;
		944
		945	package AnyEvent::CondVar::Base;
		946
		947	use overload
		948	'&{}' => sub { my $self = shift; sub { $self->send (@_) } },
		949	fallback => 1;
		950
		951	sub _send {
		952	# nop
		953	}
		954
		955	sub send {
		956	my $cv = shift;
		957	$cv->{_ae_sent} = [@_];
		958	(delete $cv->{_ae_cb})->($cv) if $cv->{_ae_cb};
		959	$cv->_send;
		960	}
		961
		962	sub croak {
		963	$_[0]{_ae_croak} = $_[1];
		964	$_[0]->send;
		965	}
		966
		967	sub ready {
		968	$_[0]{_ae_sent}
		969	}
		970
		971	sub _wait {
		972	AnyEvent->one_event while !$_[0]{_ae_sent};
		973	}
		974
		975	sub recv {
		976	$_[0]->_wait;
		977
		978	Carp::croak $_[0]{_ae_croak} if $_[0]{_ae_croak};
		979	wantarray ? @{ $_[0]{_ae_sent} } : $_[0]{_ae_sent}[0]
		980	}
		981
		982	sub cb {
		983	$_[0]{_ae_cb} = $_[1] if @_ > 1;
		984	$_[0]{_ae_cb}
		985	}
		986
		987	sub begin {
		988	++$_[0]{_ae_counter};
		989	$_[0]{_ae_end_cb} = $_[1] if @_ > 1;
		990	}
		991
		992	sub end {
		993	return if --$_[0]{_ae_counter};
		994	&{ $_[0]{_ae_end_cb} || sub { $_[0]->send } };
		995	}
		996
		997	# undocumented/compatibility with pre-3.4
		998	*broadcast = \&send;
		999	*wait = \&_wait;
		1000
		1001	=head1 SUPPLYING YOUR OWN EVENT MODEL INTERFACE
		1002
		1003	This is an advanced topic that you do not normally need to use AnyEvent in
		1004	a module. This section is only of use to event loop authors who want to
		1005	provide AnyEvent compatibility.
		1006
		1007	If you need to support another event library which isn't directly
		1008	supported by AnyEvent, you can supply your own interface to it by
		1009	pushing, before the first watcher gets created, the package name of
		1010	the event module and the package name of the interface to use onto
		1011	C<@AnyEvent::REGISTRY>. You can do that before and even without loading
		1012	AnyEvent, so it is reasonably cheap.
		1013
		1014	Example:
		1015
		1016	push @AnyEvent::REGISTRY, [urxvt => urxvt::anyevent::];
		1017
		1018	This tells AnyEvent to (literally) use the C<urxvt::anyevent::>
		1019	package/class when it finds the C<urxvt> package/module is already loaded.
		1020
		1021	When AnyEvent is loaded and asked to find a suitable event model, it
		1022	will first check for the presence of urxvt by trying to C<use> the
		1023	C<urxvt::anyevent> module.
		1024
		1025	The class should provide implementations for all watcher types. See
		1026	L<AnyEvent::Impl::EV> (source code), L<AnyEvent::Impl::Glib> (Source code)
		1027	and so on for actual examples. Use C<perldoc -m AnyEvent::Impl::Glib> to
		1028	see the sources.
		1029
		1030	If you don't provide C<signal> and C<child> watchers than AnyEvent will
		1031	provide suitable (hopefully) replacements.
		1032
		1033	The above example isn't fictitious, the I<rxvt-unicode> (a.k.a. urxvt)
		1034	terminal emulator uses the above line as-is. An interface isn't included
		1035	in AnyEvent because it doesn't make sense outside the embedded interpreter
		1036	inside I<rxvt-unicode>, and it is updated and maintained as part of the
		1037	I<rxvt-unicode> distribution.
		1038
		1039	I<rxvt-unicode> also cheats a bit by not providing blocking access to
		1040	condition variables: code blocking while waiting for a condition will
		1041	C<die>. This still works with most modules/usages, and blocking calls must
		1042	not be done in an interactive application, so it makes sense.
		1043
		1044	=head1 ENVIRONMENT VARIABLES
		1045
		1046	The following environment variables are used by this module:
		1047
		1048	=over 4
		1049
		1050	=item C<PERL_ANYEVENT_VERBOSE>
		1051
		1052	By default, AnyEvent will be completely silent except in fatal
		1053	conditions. You can set this environment variable to make AnyEvent more
		1054	talkative.
		1055
		1056	When set to C<1> or higher, causes AnyEvent to warn about unexpected
		1057	conditions, such as not being able to load the event model specified by
		1058	C<PERL_ANYEVENT_MODEL>.
		1059
		1060	When set to C<2> or higher, cause AnyEvent to report to STDERR which event
		1061	model it chooses.
		1062
		1063	=item C<PERL_ANYEVENT_MODEL>
		1064
		1065	This can be used to specify the event model to be used by AnyEvent, before
		1066	auto detection and -probing kicks in. It must be a string consisting
		1067	entirely of ASCII letters. The string C<AnyEvent::Impl::> gets prepended
		1068	and the resulting module name is loaded and if the load was successful,
		1069	used as event model. If it fails to load AnyEvent will proceed with
		1070	auto detection and -probing.
		1071
		1072	This functionality might change in future versions.
		1073
		1074	For example, to force the pure perl model (L<AnyEvent::Impl::Perl>) you
		1075	could start your program like this:
		1076
		1077	PERL_ANYEVENT_MODEL=Perl perl ...
		1078
		1079	=item C<PERL_ANYEVENT_PROTOCOLS>
		1080
		1081	Used by both L<AnyEvent::DNS> and L<AnyEvent::Socket> to determine preferences
		1082	for IPv4 or IPv6. The default is unspecified (and might change, or be the result
		1083	of auto probing).
		1084
		1085	Must be set to a comma-separated list of protocols or address families,
		1086	current supported: C<ipv4> and C<ipv6>. Only protocols mentioned will be
		1087	used, and preference will be given to protocols mentioned earlier in the
		1088	list.
		1089
		1090	This variable can effectively be used for denial-of-service attacks
		1091	against local programs (e.g. when setuid), although the impact is likely
		1092	small, as the program has to handle connection errors already-
		1093
		1094	Examples: C<PERL_ANYEVENT_PROTOCOLS=ipv4,ipv6> - prefer IPv4 over IPv6,
		1095	but support both and try to use both. C<PERL_ANYEVENT_PROTOCOLS=ipv4>
		1096	- only support IPv4, never try to resolve or contact IPv6
		1097	addresses. C<PERL_ANYEVENT_PROTOCOLS=ipv6,ipv4> support either IPv4 or
		1098	IPv6, but prefer IPv6 over IPv4.
		1099
		1100	=item C<PERL_ANYEVENT_EDNS0>
		1101
		1102	Used by L<AnyEvent::DNS> to decide whether to use the EDNS0 extension
		1103	for DNS. This extension is generally useful to reduce DNS traffic, but
		1104	some (broken) firewalls drop such DNS packets, which is why it is off by
		1105	default.
		1106
		1107	Setting this variable to C<1> will cause L<AnyEvent::DNS> to announce
		1108	EDNS0 in its DNS requests.
		1109
		1110	=back
		1111
		1112	=head1 EXAMPLE PROGRAM
		1113
		1114	The following program uses an I/O watcher to read data from STDIN, a timer
		1115	to display a message once per second, and a condition variable to quit the
		1116	program when the user enters quit:
		1117
		1118	use AnyEvent;
		1119
		1120	my $cv = AnyEvent->condvar;
		1121
		1122	my $io_watcher = AnyEvent->io (
		1123	fh => \*STDIN,
		1124	poll => 'r',
		1125	cb => sub {
		1126	warn "io event <$_[0]>\n"; # will always output <r>
		1127	chomp (my $input = <STDIN>); # read a line
		1128	warn "read: $input\n"; # output what has been read
		1129	$cv->send if $input =~ /^q/i; # quit program if /^q/i
		1130	},
		1131	);
		1132
		1133	my $time_watcher; # can only be used once
		1134
		1135	sub new_timer {
		1136	$timer = AnyEvent->timer (after => 1, cb => sub {
		1137	warn "timeout\n"; # print 'timeout' about every second
		1138	&new_timer; # and restart the time
		1139	});
		1140	}
		1141
		1142	new_timer; # create first timer
		1143
		1144	$cv->recv; # wait until user enters /^q/i
		1145
		1146	=head1 REAL-WORLD EXAMPLE
		1147
		1148	Consider the L<Net::FCP> module. It features (among others) the following
		1149	API calls, which are to freenet what HTTP GET requests are to http:
		1150
		1151	my $data = $fcp->client_get ($url); # blocks
		1152
		1153	my $transaction = $fcp->txn_client_get ($url); # does not block
		1154	$transaction->cb ( sub { ... } ); # set optional result callback
		1155	my $data = $transaction->result; # possibly blocks
		1156
		1157	The C<client_get> method works like C<LWP::Simple::get>: it requests the
		1158	given URL and waits till the data has arrived. It is defined to be:
		1159
		1160	sub client_get { $_[0]->txn_client_get ($_[1])->result }
		1161
		1162	And in fact is automatically generated. This is the blocking API of
		1163	L<Net::FCP>, and it works as simple as in any other, similar, module.
		1164
		1165	More complicated is C<txn_client_get>: It only creates a transaction
		1166	(completion, result, ...) object and initiates the transaction.
		1167
		1168	my $txn = bless { }, Net::FCP::Txn::;
		1169
		1170	It also creates a condition variable that is used to signal the completion
		1171	of the request:
		1172
		1173	$txn->{finished} = AnyAvent->condvar;
		1174
		1175	It then creates a socket in non-blocking mode.
		1176
		1177	socket $txn->{fh}, ...;
		1178	fcntl $txn->{fh}, F_SETFL, O_NONBLOCK;
		1179	connect $txn->{fh}, ...
		1180	and !$!{EWOULDBLOCK}
		1181	and !$!{EINPROGRESS}
		1182	and Carp::croak "unable to connect: $!\n";
		1183
		1184	Then it creates a write-watcher which gets called whenever an error occurs
		1185	or the connection succeeds:
		1186
		1187	$txn->{w} = AnyEvent->io (fh => $txn->{fh}, poll => 'w', cb => sub { $txn->fh_ready_w });
		1188
		1189	And returns this transaction object. The C<fh_ready_w> callback gets
		1190	called as soon as the event loop detects that the socket is ready for
		1191	writing.
		1192
		1193	The C<fh_ready_w> method makes the socket blocking again, writes the
		1194	request data and replaces the watcher by a read watcher (waiting for reply
		1195	data). The actual code is more complicated, but that doesn't matter for
		1196	this example:
		1197
		1198	fcntl $txn->{fh}, F_SETFL, 0;
		1199	syswrite $txn->{fh}, $txn->{request}
		1200	or die "connection or write error";
		1201	$txn->{w} = AnyEvent->io (fh => $txn->{fh}, poll => 'r', cb => sub { $txn->fh_ready_r });
		1202
		1203	Again, C<fh_ready_r> waits till all data has arrived, and then stores the
		1204	result and signals any possible waiters that the request has finished:
		1205
		1206	sysread $txn->{fh}, $txn->{buf}, length $txn->{$buf};
		1207
		1208	if (end-of-file or data complete) {
		1209	$txn->{result} = $txn->{buf};
		1210	$txn->{finished}->send;
		1211	$txb->{cb}->($txn) of $txn->{cb}; # also call callback
		1212	}
		1213
		1214	The C<result> method, finally, just waits for the finished signal (if the
		1215	request was already finished, it doesn't wait, of course, and returns the
		1216	data:
		1217
		1218	$txn->{finished}->recv;
		1219	return $txn->{result};
		1220
		1221	The actual code goes further and collects all errors (C<die>s, exceptions)
		1222	that occurred during request processing. The C<result> method detects
		1223	whether an exception as thrown (it is stored inside the $txn object)
		1224	and just throws the exception, which means connection errors and other
		1225	problems get reported tot he code that tries to use the result, not in a
		1226	random callback.
		1227
		1228	All of this enables the following usage styles:
		1229
		1230	1. Blocking:
		1231
		1232	my $data = $fcp->client_get ($url);
		1233
		1234	2. Blocking, but running in parallel:
		1235
		1236	my @datas = map $_->result,
		1237	map $fcp->txn_client_get ($_),
		1238	@urls;
		1239
		1240	Both blocking examples work without the module user having to know
		1241	anything about events.
		1242
		1243	3a. Event-based in a main program, using any supported event module:
		1244
		1245	use EV;
		1246
		1247	$fcp->txn_client_get ($url)->cb (sub {
		1248	my $txn = shift;
		1249	my $data = $txn->result;
		1250	...
		1251	});
		1252
		1253	EV::loop;
		1254
		1255	3b. The module user could use AnyEvent, too:
		1256
		1257	use AnyEvent;
		1258
		1259	my $quit = AnyEvent->condvar;
		1260
		1261	$fcp->txn_client_get ($url)->cb (sub {
		1262	...
		1263	$quit->send;
		1264	});
		1265
		1266	$quit->recv;
		1267
		1268
		1269	=head1 BENCHMARKS
		1270
		1271	To give you an idea of the performance and overheads that AnyEvent adds
		1272	over the event loops themselves and to give you an impression of the speed
		1273	of various event loops I prepared some benchmarks.
		1274
		1275	=head2 BENCHMARKING ANYEVENT OVERHEAD
		1276
		1277	Here is a benchmark of various supported event models used natively and
		1278	through AnyEvent. The benchmark creates a lot of timers (with a zero
		1279	timeout) and I/O watchers (watching STDOUT, a pty, to become writable,
		1280	which it is), lets them fire exactly once and destroys them again.
		1281
		1282	Source code for this benchmark is found as F<eg/bench> in the AnyEvent
		1283	distribution.
		1284
		1285	=head3 Explanation of the columns
		1286
		1287	I<watcher> is the number of event watchers created/destroyed. Since
		1288	different event models feature vastly different performances, each event
		1289	loop was given a number of watchers so that overall runtime is acceptable
		1290	and similar between tested event loop (and keep them from crashing): Glib
		1291	would probably take thousands of years if asked to process the same number
		1292	of watchers as EV in this benchmark.
		1293
		1294	I<bytes> is the number of bytes (as measured by the resident set size,
		1295	RSS) consumed by each watcher. This method of measuring captures both C
		1296	and Perl-based overheads.
		1297
		1298	I<create> is the time, in microseconds (millionths of seconds), that it
		1299	takes to create a single watcher. The callback is a closure shared between
		1300	all watchers, to avoid adding memory overhead. That means closure creation
		1301	and memory usage is not included in the figures.
		1302
		1303	I<invoke> is the time, in microseconds, used to invoke a simple
		1304	callback. The callback simply counts down a Perl variable and after it was
		1305	invoked "watcher" times, it would C<< ->send >> a condvar once to
		1306	signal the end of this phase.
		1307
		1308	I<destroy> is the time, in microseconds, that it takes to destroy a single
		1309	watcher.
		1310
		1311	=head3 Results
		1312
		1313	name watchers bytes create invoke destroy comment
		1314	EV/EV 400000 244 0.56 0.46 0.31 EV native interface
		1315	EV/Any 100000 244 2.50 0.46 0.29 EV + AnyEvent watchers
		1316	CoroEV/Any 100000 244 2.49 0.44 0.29 coroutines + Coro::Signal
		1317	Perl/Any 100000 513 4.92 0.87 1.12 pure perl implementation
		1318	Event/Event 16000 516 31.88 31.30 0.85 Event native interface
		1319	Event/Any 16000 590 35.75 31.42 1.08 Event + AnyEvent watchers
		1320	Glib/Any 16000 1357 98.22 12.41 54.00 quadratic behaviour
		1321	Tk/Any 2000 1860 26.97 67.98 14.00 SEGV with >> 2000 watchers
		1322	POE/Event 2000 6644 108.64 736.02 14.73 via POE::Loop::Event
		1323	POE/Select 2000 6343 94.13 809.12 565.96 via POE::Loop::Select
		1324
		1325	=head3 Discussion
		1326
		1327	The benchmark does I<not> measure scalability of the event loop very
		1328	well. For example, a select-based event loop (such as the pure perl one)
		1329	can never compete with an event loop that uses epoll when the number of
		1330	file descriptors grows high. In this benchmark, all events become ready at
		1331	the same time, so select/poll-based implementations get an unnatural speed
		1332	boost.
		1333
		1334	Also, note that the number of watchers usually has a nonlinear effect on
		1335	overall speed, that is, creating twice as many watchers doesn't take twice
		1336	the time - usually it takes longer. This puts event loops tested with a
		1337	higher number of watchers at a disadvantage.
		1338
		1339	To put the range of results into perspective, consider that on the
		1340	benchmark machine, handling an event takes roughly 1600 CPU cycles with
		1341	EV, 3100 CPU cycles with AnyEvent's pure perl loop and almost 3000000 CPU
		1342	cycles with POE.
		1343
		1344	C<EV> is the sole leader regarding speed and memory use, which are both
		1345	maximal/minimal, respectively. Even when going through AnyEvent, it uses
		1346	far less memory than any other event loop and is still faster than Event
		1347	natively.
		1348
		1349	The pure perl implementation is hit in a few sweet spots (both the
		1350	constant timeout and the use of a single fd hit optimisations in the perl
		1351	interpreter and the backend itself). Nevertheless this shows that it
		1352	adds very little overhead in itself. Like any select-based backend its
		1353	performance becomes really bad with lots of file descriptors (and few of
		1354	them active), of course, but this was not subject of this benchmark.
		1355
		1356	The C<Event> module has a relatively high setup and callback invocation
		1357	cost, but overall scores in on the third place.
		1358
		1359	C<Glib>'s memory usage is quite a bit higher, but it features a
		1360	faster callback invocation and overall ends up in the same class as
		1361	C<Event>. However, Glib scales extremely badly, doubling the number of
		1362	watchers increases the processing time by more than a factor of four,
		1363	making it completely unusable when using larger numbers of watchers
		1364	(note that only a single file descriptor was used in the benchmark, so
		1365	inefficiencies of C<poll> do not account for this).
		1366
		1367	The C<Tk> adaptor works relatively well. The fact that it crashes with
		1368	more than 2000 watchers is a big setback, however, as correctness takes
		1369	precedence over speed. Nevertheless, its performance is surprising, as the
		1370	file descriptor is dup()ed for each watcher. This shows that the dup()
		1371	employed by some adaptors is not a big performance issue (it does incur a
		1372	hidden memory cost inside the kernel which is not reflected in the figures
		1373	above).
		1374
		1375	C<POE>, regardless of underlying event loop (whether using its pure perl
		1376	select-based backend or the Event module, the POE-EV backend couldn't
		1377	be tested because it wasn't working) shows abysmal performance and
		1378	memory usage with AnyEvent: Watchers use almost 30 times as much memory
		1379	as EV watchers, and 10 times as much memory as Event (the high memory
		1380	requirements are caused by requiring a session for each watcher). Watcher
		1381	invocation speed is almost 900 times slower than with AnyEvent's pure perl
		1382	implementation.
		1383
		1384	The design of the POE adaptor class in AnyEvent can not really account
		1385	for the performance issues, though, as session creation overhead is
		1386	small compared to execution of the state machine, which is coded pretty
		1387	optimally within L<AnyEvent::Impl::POE> (and while everybody agrees that
		1388	using multiple sessions is not a good approach, especially regarding
		1389	memory usage, even the author of POE could not come up with a faster
		1390	design).
		1391
		1392	=head3 Summary
		1393
		1394	=over 4
		1395
		1396	=item * Using EV through AnyEvent is faster than any other event loop
		1397	(even when used without AnyEvent), but most event loops have acceptable
		1398	performance with or without AnyEvent.
		1399
		1400	=item * The overhead AnyEvent adds is usually much smaller than the overhead of
		1401	the actual event loop, only with extremely fast event loops such as EV
		1402	adds AnyEvent significant overhead.
		1403
		1404	=item * You should avoid POE like the plague if you want performance or
		1405	reasonable memory usage.
		1406
		1407	=back
		1408
		1409	=head2 BENCHMARKING THE LARGE SERVER CASE
		1410
		1411	This benchmark actually benchmarks the event loop itself. It works by
		1412	creating a number of "servers": each server consists of a socket pair, a
		1413	timeout watcher that gets reset on activity (but never fires), and an I/O
		1414	watcher waiting for input on one side of the socket. Each time the socket
		1415	watcher reads a byte it will write that byte to a random other "server".
		1416
		1417	The effect is that there will be a lot of I/O watchers, only part of which
		1418	are active at any one point (so there is a constant number of active
		1419	fds for each loop iteration, but which fds these are is random). The
		1420	timeout is reset each time something is read because that reflects how
		1421	most timeouts work (and puts extra pressure on the event loops).
		1422
		1423	In this benchmark, we use 10000 socket pairs (20000 sockets), of which 100
		1424	(1%) are active. This mirrors the activity of large servers with many
		1425	connections, most of which are idle at any one point in time.
		1426
		1427	Source code for this benchmark is found as F<eg/bench2> in the AnyEvent
		1428	distribution.
		1429
		1430	=head3 Explanation of the columns
		1431
		1432	I<sockets> is the number of sockets, and twice the number of "servers" (as
		1433	each server has a read and write socket end).
		1434
		1435	I<create> is the time it takes to create a socket pair (which is
		1436	nontrivial) and two watchers: an I/O watcher and a timeout watcher.
		1437
		1438	I<request>, the most important value, is the time it takes to handle a
		1439	single "request", that is, reading the token from the pipe and forwarding
		1440	it to another server. This includes deleting the old timeout and creating
		1441	a new one that moves the timeout into the future.
		1442
		1443	=head3 Results
		1444
		1445	name sockets create request
		1446	EV 20000 69.01 11.16
		1447	Perl 20000 73.32 35.87
		1448	Event 20000 212.62 257.32
		1449	Glib 20000 651.16 1896.30
		1450	POE 20000 349.67 12317.24 uses POE::Loop::Event
		1451
		1452	=head3 Discussion
		1453
		1454	This benchmark I<does> measure scalability and overall performance of the
		1455	particular event loop.
		1456
		1457	EV is again fastest. Since it is using epoll on my system, the setup time
		1458	is relatively high, though.
		1459
		1460	Perl surprisingly comes second. It is much faster than the C-based event
		1461	loops Event and Glib.
		1462
		1463	Event suffers from high setup time as well (look at its code and you will
		1464	understand why). Callback invocation also has a high overhead compared to
		1465	the C<< $_->() for .. >>-style loop that the Perl event loop uses. Event
		1466	uses select or poll in basically all documented configurations.
		1467
		1468	Glib is hit hard by its quadratic behaviour w.r.t. many watchers. It
		1469	clearly fails to perform with many filehandles or in busy servers.
		1470
		1471	POE is still completely out of the picture, taking over 1000 times as long
		1472	as EV, and over 100 times as long as the Perl implementation, even though
		1473	it uses a C-based event loop in this case.
		1474
		1475	=head3 Summary
		1476
		1477	=over 4
		1478
		1479	=item * The pure perl implementation performs extremely well.
		1480
		1481	=item * Avoid Glib or POE in large projects where performance matters.
		1482
		1483	=back
		1484
		1485	=head2 BENCHMARKING SMALL SERVERS
		1486
		1487	While event loops should scale (and select-based ones do not...) even to
		1488	large servers, most programs we (or I :) actually write have only a few
		1489	I/O watchers.
		1490
		1491	In this benchmark, I use the same benchmark program as in the large server
		1492	case, but it uses only eight "servers", of which three are active at any
		1493	one time. This should reflect performance for a small server relatively
		1494	well.
		1495
		1496	The columns are identical to the previous table.
		1497
		1498	=head3 Results
		1499
		1500	name sockets create request
		1501	EV 16 20.00 6.54
		1502	Perl 16 25.75 12.62
		1503	Event 16 81.27 35.86
		1504	Glib 16 32.63 15.48
		1505	POE 16 261.87 276.28 uses POE::Loop::Event
		1506
		1507	=head3 Discussion
		1508
		1509	The benchmark tries to test the performance of a typical small
		1510	server. While knowing how various event loops perform is interesting, keep
		1511	in mind that their overhead in this case is usually not as important, due
		1512	to the small absolute number of watchers (that is, you need efficiency and
		1513	speed most when you have lots of watchers, not when you only have a few of
		1514	them).
		1515
		1516	EV is again fastest.
		1517
		1518	Perl again comes second. It is noticeably faster than the C-based event
		1519	loops Event and Glib, although the difference is too small to really
		1520	matter.
		1521
		1522	POE also performs much better in this case, but is is still far behind the
		1523	others.
		1524
		1525	=head3 Summary
		1526
		1527	=over 4
		1528
		1529	=item * C-based event loops perform very well with small number of
		1530	watchers, as the management overhead dominates.
		1531
		1532	=back
		1533
		1534
		1535	=head1 FORK
		1536
		1537	Most event libraries are not fork-safe. The ones who are usually are
		1538	because they rely on inefficient but fork-safe C<select> or C<poll>
		1539	calls. Only L<EV> is fully fork-aware.
		1540
		1541	If you have to fork, you must either do so I<before> creating your first
		1542	watcher OR you must not use AnyEvent at all in the child.
		1543
		1544
		1545	=head1 SECURITY CONSIDERATIONS
		1546
		1547	AnyEvent can be forced to load any event model via
		1548	$ENV{PERL_ANYEVENT_MODEL}. While this cannot (to my knowledge) be used to
		1549	execute arbitrary code or directly gain access, it can easily be used to
		1550	make the program hang or malfunction in subtle ways, as AnyEvent watchers
		1551	will not be active when the program uses a different event model than
		1552	specified in the variable.
		1553
		1554	You can make AnyEvent completely ignore this variable by deleting it
		1555	before the first watcher gets created, e.g. with a C<BEGIN> block:
		1556
		1557	BEGIN { delete $ENV{PERL_ANYEVENT_MODEL} }
		1558
		1559	use AnyEvent;
		1560
		1561	Similar considerations apply to $ENV{PERL_ANYEVENT_VERBOSE}, as that can
		1562	be used to probe what backend is used and gain other information (which is
		1563	probably even less useful to an attacker than PERL_ANYEVENT_MODEL).
		1564
		1565
		1566	=head1 SEE ALSO
		1567
		1568	Utility functions: L<AnyEvent::Util>.
		1569
		1570	Event modules: L<EV>, L<EV::Glib>, L<Glib::EV>, L<Event>, L<Glib::Event>,
		1571	L<Glib>, L<Tk>, L<Event::Lib>, L<Qt>, L<POE>.
		1572
		1573	Implementations: L<AnyEvent::Impl::EV>, L<AnyEvent::Impl::Event>,
		1574	L<AnyEvent::Impl::Glib>, L<AnyEvent::Impl::Tk>, L<AnyEvent::Impl::Perl>,
		1575	L<AnyEvent::Impl::EventLib>, L<AnyEvent::Impl::Qt>,
		1576	L<AnyEvent::Impl::POE>.
		1577
		1578	Non-blocking file handles, sockets, TCP clients and
		1579	servers: L<AnyEvent::Handle>, L<AnyEvent::Socket>.
		1580
		1581	Asynchronous DNS: L<AnyEvent::DNS>.
		1582
		1583	Coroutine support: L<Coro>, L<Coro::AnyEvent>, L<Coro::EV>, L<Coro::Event>,
		1584
		1585	Nontrivial usage examples: L<Net::FCP>, L<Net::XMPP2>, L<AnyEvent::DNS>.
		1586
		1587
		1588	=head1 AUTHOR
		1589
		1590	Marc Lehmann <schmorp@schmorp.de>
		1591	http://home.schmorp.de/
		1592
		1593	=cut
		1594
		1595	1
		1596

Diff Legend

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