[ViewVC] Diff of: cvs/AnyEvent/lib/AnyEvent.pm

Comparing AnyEvent/lib/AnyEvent.pm (file contents):
Revision 1.41 by root, Mon Apr 7 19:23:59 2008 UTC vs.
Revision 1.112 by root, Sat May 10 01:04:42 2008 UTC

…		…
1	=head1 NAME	1	=head1 NAME
2		2
3	AnyEvent - provide framework for multiple event loops	3	AnyEvent - provide framework for multiple event loops
4		4
5	Event, Coro, Glib, Tk, Perl - various supported event loops	5	EV, Event, Glib, Tk, Perl, Event::Lib, Qt, POE - various supported event loops
6		6
7	=head1 SYNOPSIS	7	=head1 SYNOPSIS
8		8
9	use AnyEvent;	9	use AnyEvent;
10		10
14		14
15	my $w = AnyEvent->timer (after => $seconds, cb => sub {	15	my $w = AnyEvent->timer (after => $seconds, cb => sub {
16	...	16	...
17	});	17	});
18		18
19	my $w = AnyEvent->condvar; # stores wether a condition was flagged	19	my $w = AnyEvent->condvar; # stores whether a condition was flagged
20	$w->wait; # enters "main loop" till $condvar gets ->broadcast	20	$w->wait; # enters "main loop" till $condvar gets ->send
21	$w->broadcast; # wake up current and all future wait's	21	$w->send; # wake up current and all future wait's
22		22
23	=head1 WHY YOU SHOULD USE THIS MODULE	23	=head1 WHY YOU SHOULD USE THIS MODULE (OR NOT)
24		24
25	Glib, POE, IO::Async, Event... CPAN offers event models by the dozen	25	Glib, POE, IO::Async, Event... CPAN offers event models by the dozen
26	nowadays. So what is different about AnyEvent?	26	nowadays. So what is different about AnyEvent?
27		27
28	Executive Summary: AnyEvent is I<compatible>, AnyEvent is I<free of	28	Executive Summary: AnyEvent is I<compatible>, AnyEvent is I<free of
29	policy> and AnyEvent is I<small and efficient>.	29	policy> and AnyEvent is I<small and efficient>.
30		30
31	First and foremost, I<AnyEvent is not an event model> itself, it only	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	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,	33	pragmatic way. For event models and certain classes of immortals alike,
34	the statement "there can only be one" is a bitter reality, and AnyEvent	34	the statement "there can only be one" is a bitter reality: In general,
35	helps hiding the differences.	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.
36		37
37	The goal of AnyEvent is to offer module authors the ability to do event	38	The goal of AnyEvent is to offer module authors the ability to do event
38	programming (waiting for I/O or timer events) without subscribing to a	39	programming (waiting for I/O or timer events) without subscribing to a
39	religion, a way of living, and most importantly: without forcing your	40	religion, a way of living, and most importantly: without forcing your
40	module users into the same thing by forcing them to use the same event	41	module users into the same thing by forcing them to use the same event
41	model you use.	42	model you use.
42		43
43	For modules like POE or IO::Async (the latter of which is actually	44	For modules like POE or IO::Async (which is a total misnomer as it is
44	named confusingly, as it does neither do I/O nor does it do anything	45	actually doing all I/O I<synchronously>...), using them in your module is
45	asynchronously...), using them in your module is like joining a
46	cult: After you joined, you are dependent on them and you cannot use	46	like joining a cult: After you joined, you are dependent on them and you
47	anything else, as it is simply incompatible to everything that isn't	47	cannot use anything else, as it is simply incompatible to everything that
48	itself.	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.
49		50
50	AnyEvent + POE works fine. AnyEvent + Glib works fine. AnyEvent + Tk	51	AnyEvent is different: AnyEvent + POE works fine. AnyEvent + Glib works
51	works fine etc. etc. but none of these work together with the rest: POE	52	fine. AnyEvent + Tk works fine etc. etc. but none of these work together
52	+ IO::Async? no go. Tk + Event? no go. If your module uses one of	53	with the rest: POE + IO::Async? no go. Tk + Event? no go. Again: if
53	those, every user of your module has to use it, too. If your module	54	your module uses one of those, every user of your module has to use it,
54	uses AnyEvent, it works transparently with all event models it supports	55	too. But if your module uses AnyEvent, it works transparently with all
55	(including stuff like POE and IO::Async).	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).
56		59
57	In addition of being free of having to use I<the one and only true event	60	In addition to being free of having to use I<the one and only true event
58	model>, AnyEvent also is free of bloat and policy: with POE or similar	61	model>, AnyEvent also is free of bloat and policy: with POE or similar
59	modules, you get an enourmous amount of code and strict rules you have	62	modules, you get an enourmous amount of code and strict rules you have to
60	to follow. AnyEvent, on the other hand, is lean and to the point by only	63	follow. AnyEvent, on the other hand, is lean and up to the point, by only
61	offering the functionality that is useful, in as thin as a wrapper as	64	offering the functionality that is necessary, in as thin as a wrapper as
62	technically possible.	65	technically possible.
63		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.
64		70
65	=head1 DESCRIPTION	71	=head1 DESCRIPTION
66		72
67	L<AnyEvent> provides an identical interface to multiple event loops. This	73	L<AnyEvent> provides an identical interface to multiple event loops. This
68	allows module authors to utilise an event loop without forcing module	74	allows module authors to utilise an event loop without forcing module
69	users to use the same event loop (as only a single event loop can coexist	75	users to use the same event loop (as only a single event loop can coexist
70	peacefully at any one time).	76	peacefully at any one time).
71		77
72	The interface itself is vaguely similar but not identical to the Event	78	The interface itself is vaguely similar, but not identical to the L<Event>
73	module.	79	module.
74		80
75	On the first call of any method, the module tries to detect the currently	81	During the first call of any watcher-creation method, the module tries
76	loaded event loop by probing wether any of the following modules is	82	to detect the currently loaded event loop by probing whether one of the
77	loaded: L<Coro::Event>, L<Event>, L<Glib>, L<Tk>. The first one found is	83	following modules is already loaded: L<EV>,
78	used. If none is found, the module tries to load these modules in the	84	L<Event>, L<Glib>, L<AnyEvent::Impl::Perl>, L<Tk>, L<Event::Lib>, L<Qt>,
79	order given. The first one that could be successfully loaded will be	85	L<POE>. The first one found is used. If none are found, the module tries
80	used. If still none could be found, AnyEvent will fall back to a pure-perl	86	to load these modules (excluding Tk, Event::Lib, Qt and POE as the pure perl
81	event loop, which is also not very efficient.	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.
82		91
83	Because AnyEvent first checks for modules that are already loaded, loading	92	Because AnyEvent first checks for modules that are already loaded, loading
84	an Event model explicitly before first using AnyEvent will likely make	93	an event model explicitly before first using AnyEvent will likely make
85	that model the default. For example:	94	that model the default. For example:
86		95
87	use Tk;	96	use Tk;
88	use AnyEvent;	97	use AnyEvent;
89		98
90	# .. AnyEvent will likely default to Tk	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...
91		104
92	The pure-perl implementation of AnyEvent is called	105	The pure-perl implementation of AnyEvent is called
93	C<AnyEvent::Impl::Perl>. Like other event modules you can load it	106	C<AnyEvent::Impl::Perl>. Like other event modules you can load it
94	explicitly.	107	explicitly.
95		108
…		…
98	AnyEvent has the central concept of a I<watcher>, which is an object that	111	AnyEvent has the central concept of a I<watcher>, which is an object that
99	stores relevant data for each kind of event you are waiting for, such as	112	stores relevant data for each kind of event you are waiting for, such as
100	the callback to call, the filehandle to watch, etc.	113	the callback to call, the filehandle to watch, etc.
101		114
102	These watchers are normal Perl objects with normal Perl lifetime. After	115	These watchers are normal Perl objects with normal Perl lifetime. After
103	creating a watcher it will immediately "watch" for events and invoke	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
104	the callback. To disable the watcher you have to destroy it (e.g. by	120	To disable the watcher you have to destroy it (e.g. by setting the
105	setting the variable that stores it to C<undef> or otherwise deleting all	121	variable you store it in to C<undef> or otherwise deleting all references
106	references to it).	122	to it).
107		123
108	All watchers are created by calling a method on the C<AnyEvent> class.	124	All watchers are created by calling a method on the C<AnyEvent> class.
109		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
110	=head2 IO WATCHERS	140	=head2 I/O WATCHERS
111		141
112	You can create I/O watcher by calling the C<< AnyEvent->io >> method with	142	You can create an I/O watcher by calling the C<< AnyEvent->io >> method
113	the following mandatory arguments:	143	with the following mandatory key-value pairs as arguments:
114		144
115	C<fh> the Perl I<filehandle> (not filedescriptor) to watch for	145	C<fh> the Perl I<file handle> (I<not> file descriptor) to watch
116	events. C<poll> must be a string that is either C<r> or C<w>, that creates	146	for events. C<poll> must be a string that is either C<r> or C<w>,
117	a watcher waiting for "r"eadable or "w"ritable events. C<cb> the callback	147	which creates a watcher waiting for "r"eadable or "w"ritable events,
118	to invoke everytime the filehandle becomes ready.	148	respectively. C<cb> is the callback to invoke each time the file handle
		149	becomes ready.
119		150
120	Only one io watcher per C<fh> and C<poll> combination is allowed (i.e. on	151	Although the callback might get passed parameters, their value and
121	a socket you can have one r + one w, not any more (limitation comes from	152	presence is undefined and you cannot rely on them. Portable AnyEvent
122	Tk - if you are sure you are not using Tk this limitation is gone).	153	callbacks cannot use arguments passed to I/O watcher callbacks.
123		154
124	Filehandles will be kept alive, so as long as the watcher exists, the	155	The I/O watcher might use the underlying file descriptor or a copy of it.
125	filehandle exists, too.	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.
126		162
127	Example:	163	Example:
128		164
129	# wait for readability of STDIN, then read a line and disable the watcher	165	# wait for readability of STDIN, then read a line and disable the watcher
130	my $w; $w = AnyEvent->io (fh => \*STDIN, poll => 'r', cb => sub {	166	my $w; $w = AnyEvent->io (fh => \*STDIN, poll => 'r', cb => sub {
…		…
136	=head2 TIME WATCHERS	172	=head2 TIME WATCHERS
137		173
138	You can create a time watcher by calling the C<< AnyEvent->timer >>	174	You can create a time watcher by calling the C<< AnyEvent->timer >>
139	method with the following mandatory arguments:	175	method with the following mandatory arguments:
140		176
141	C<after> after how many seconds (fractions are supported) should the timer	177	C<after> specifies after how many seconds (fractional values are
142	activate. C<cb> the callback to invoke.	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.
143		184
144	The timer callback will be invoked at most once: if you want a repeating	185	The timer callback will be invoked at most once: if you want a repeating
145	timer you have to create a new watcher (this is a limitation by both Tk	186	timer you have to create a new watcher (this is a limitation by both Tk
146	and Glib).	187	and Glib).
147		188
…		…
153	});	194	});
154		195
155	# to cancel the timer:	196	# to cancel the timer:
156	undef $w;	197	undef $w;
157		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 occurances 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 guarenteed 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	AnyEvent::detect; # force event module to be initialised
		285
		286	my $pid = fork or exit 5;
		287
		288	my $w = AnyEvent->child (
		289	pid => $pid,
		290	cb => sub {
		291	my ($pid, $status) = @_;
		292	warn "pid $pid exited with status $status";
		293	$done->send;
		294	},
		295	);
		296
		297	# do something else, then wait for process exit
		298	$done->wait;
		299
158	=head2 CONDITION WATCHERS	300	=head2 CONDITION VARIABLES
159		301
		302	If you are familiar with some event loops you will know that all of them
		303	require you to run some blocking "loop", "run" or similar function that
		304	will actively watch for new events and call your callbacks.
		305
		306	AnyEvent is different, it expects somebody else to run the event loop and
		307	will only block when necessary (usually when told by the user).
		308
		309	The instrument to do that is called a "condition variable", so called
		310	because they represent a condition that must become true.
		311
160	Condition watchers can be created by calling the C<< AnyEvent->condvar >>	312	Condition variables can be created by calling the C<< AnyEvent->condvar
161	method without any arguments.	313	>> method, usually without arguments. The only argument pair allowed is
		314	C<cb>, which specifies a callback to be called when the condition variable
		315	becomes true.
162		316
163	A condition watcher watches for a condition - precisely that the C<<	317	After creation, the conditon variable is "false" until it becomes "true"
164	->broadcast >> method has been called.	318	by calling the C<send> method.
165		319
		320	Condition variables are similar to callbacks, except that you can
		321	optionally wait for them. They can also be called merge points - points
		322	in time where multiple outstandign events have been processed. And yet
		323	another way to call them is transations - each condition variable can be
		324	used to represent a transaction, which finishes at some point and delivers
		325	a result.
		326
		327	Condition variables are very useful to signal that something has finished,
		328	for example, if you write a module that does asynchronous http requests,
		329	then a condition variable would be the ideal candidate to signal the
		330	availability of results. The user can either act when the callback is
		331	called or can synchronously C<< ->wait >> for the results.
		332
		333	You can also use them to simulate traditional event loops - for example,
		334	you can block your main program until an event occurs - for example, you
		335	could C<< ->wait >> in your main program until the user clicks the Quit
		336	button of your app, which would C<< ->send >> the "quit" event.
		337
166	Note that condition watchers recurse into the event loop - if you have	338	Note that condition variables recurse into the event loop - if you have
167	two watchers that call C<< ->wait >> in a round-robbin fashion, you	339	two pieces of code that call C<< ->wait >> in a round-robbin fashion, you
168	lose. Therefore, condition watchers are good to export to your caller, but	340	lose. Therefore, condition variables are good to export to your caller, but
169	you should avoid making a blocking wait, at least in callbacks, as this	341	you should avoid making a blocking wait yourself, at least in callbacks,
170	usually asks for trouble.	342	as this asks for trouble.
171		343
172	The watcher has only two methods:	344	Condition variables are represented by hash refs in perl, and the keys
		345	used by AnyEvent itself are all named C<_ae_XXX> to make subclassing
		346	easy (it is often useful to build your own transaction class on top of
		347	AnyEvent). To subclass, use C<AnyEvent::CondVar> as base class and call
		348	it's C<new> method in your own C<new> method.
173		349
174	=over 4	350	There are two "sides" to a condition variable - the "producer side" which
175		351	eventually calls C<< -> send >>, and the "consumer side", which waits
176	=item $cv->wait	352	for the send to occur.
177
178	Wait (blocking if necessary) until the C<< ->broadcast >> method has been
179	called on c<$cv>, while servicing other watchers normally.
180
181	Not all event models support a blocking wait - some die in that case, so
182	if you are using this from a module, never require a blocking wait, but
183	let the caller decide wether the call will block or not (for example,
184	by coupling condition variables with some kind of request results and
185	supporting callbacks so the caller knows that getting the result will not
186	block, while still suppporting blockign waits if the caller so desires).
187
188	You can only wait once on a condition - additional calls will return
189	immediately.
190
191	=item $cv->broadcast
192
193	Flag the condition as ready - a running C<< ->wait >> and all further
194	calls to C<wait> will return after this method has been called. If nobody
195	is waiting the broadcast will be remembered..
196		353
197	Example:	354	Example:
198		355
199	# wait till the result is ready	356	# wait till the result is ready
200	my $result_ready = AnyEvent->condvar;	357	my $result_ready = AnyEvent->condvar;
201		358
202	# do something such as adding a timer	359	# do something such as adding a timer
203	# or socket watcher the calls $result_ready->broadcast	360	# or socket watcher the calls $result_ready->send
204	# when the "result" is ready.	361	# when the "result" is ready.
		362	# in this case, we simply use a timer:
		363	my $w = AnyEvent->timer (
		364	after => 1,
		365	cb => sub { $result_ready->send },
		366	);
205		367
		368	# this "blocks" (while handling events) till the callback
		369	# calls send
206	$result_ready->wait;	370	$result_ready->wait;
207		371
		372	=head3 METHODS FOR PRODUCERS
		373
		374	These methods should only be used by the producing side, i.e. the
		375	code/module that eventually sends the signal. Note that it is also
		376	the producer side which creates the condvar in most cases, but it isn't
		377	uncommon for the consumer to create it as well.
		378
		379	=over 4
		380
		381	=item $cv->send (...)
		382
		383	Flag the condition as ready - a running C<< ->wait >> and all further
		384	calls to C<wait> will (eventually) return after this method has been
		385	called. If nobody is waiting the send will be remembered.
		386
		387	If a callback has been set on the condition variable, it is called
		388	immediately from within send.
		389
		390	Any arguments passed to the C<send> call will be returned by all
		391	future C<< ->wait >> calls.
		392
		393	=item $cv->croak ($error)
		394
		395	Similar to send, but causes all call's wait C<< ->wait >> to invoke
		396	C<Carp::croak> with the given error message/object/scalar.
		397
		398	This can be used to signal any errors to the condition variable
		399	user/consumer.
		400
		401	=item $cv->begin ([group callback])
		402
		403	=item $cv->end
		404
		405	These two methods can be used to combine many transactions/events into
		406	one. For example, a function that pings many hosts in parallel might want
		407	to use a condition variable for the whole process.
		408
		409	Every call to C<< ->begin >> will increment a counter, and every call to
		410	C<< ->end >> will decrement it. If the counter reaches C<0> in C<< ->end
		411	>>, the (last) callback passed to C<begin> will be executed. That callback
		412	is I<supposed> to call C<< ->send >>, but that is not required. If no
		413	callback was set, C<send> will be called without any arguments.
		414
		415	Let's clarify this with the ping example:
		416
		417	my $cv = AnyEvent->condvar;
		418
		419	my %result;
		420	$cv->begin (sub { $cv->send (\%result) });
		421
		422	for my $host (@list_of_hosts) {
		423	$cv->begin;
		424	ping_host_then_call_callback $host, sub {
		425	$result{$host} = ...;
		426	$cv->end;
		427	};
		428	}
		429
		430	$cv->end;
		431
		432	This code fragment supposedly pings a number of hosts and calls
		433	C<send> after results for all then have have been gathered - in any
		434	order. To achieve this, the code issues a call to C<begin> when it starts
		435	each ping request and calls C<end> when it has received some result for
		436	it. Since C<begin> and C<end> only maintain a counter, the order in which
		437	results arrive is not relevant.
		438
		439	There is an additional bracketing call to C<begin> and C<end> outside the
		440	loop, which serves two important purposes: first, it sets the callback
		441	to be called once the counter reaches C<0>, and second, it ensures that
		442	C<send> is called even when C<no> hosts are being pinged (the loop
		443	doesn't execute once).
		444
		445	This is the general pattern when you "fan out" into multiple subrequests:
		446	use an outer C<begin>/C<end> pair to set the callback and ensure C<end>
		447	is called at least once, and then, for each subrequest you start, call
		448	C<begin> and for eahc subrequest you finish, call C<end>.
		449
208	=back	450	=back
209		451
210	=head2 SIGNAL WATCHERS	452	=head3 METHODS FOR CONSUMERS
211		453
212	You can listen for signals using a signal watcher, C<signal> is the signal	454	These methods should only be used by the consuming side, i.e. the
213	I<name> without any C<SIG> prefix. Multiple signals events can be clumped	455	code awaits the condition.
214	together into one callback invocation, and callback invocation might or
215	might not be asynchronous.
216		456
217	These watchers might use C<%SIG>, so programs overwriting those signals	457	=over 4
218	directly will likely not work correctly.
219		458
220	Example: exit on SIGINT	459	=item $cv->wait
221		460
222	my $w = AnyEvent->signal (signal => "INT", cb => sub { exit 1 });	461	Wait (blocking if necessary) until the C<< ->send >> or C<< ->croak
		462	>> methods have been called on c<$cv>, while servicing other watchers
		463	normally.
223		464
224	=head2 CHILD PROCESS WATCHERS	465	You can only wait once on a condition - additional calls are valid but
		466	will return immediately.
225		467
226	You can also listen for the status of a child process specified by the	468	If an error condition has been set by calling C<< ->croak >>, then this
227	C<pid> argument (or any child if the pid argument is 0). The watcher will	469	function will call C<croak>.
228	trigger as often as status change for the child are received. This works
229	by installing a signal handler for C<SIGCHLD>. The callback will be called with
230	the pid and exit status (as returned by waitpid).
231		470
232	Example: wait for pid 1333	471	In list context, all parameters passed to C<send> will be returned,
		472	in scalar context only the first one will be returned.
233		473
234	my $w = AnyEvent->child (pid => 1333, cb => sub { warn "exit status $?" });	474	Not all event models support a blocking wait - some die in that case
		475	(programs might want to do that to stay interactive), so I<if you are
		476	using this from a module, never require a blocking wait>, but let the
		477	caller decide whether the call will block or not (for example, by coupling
		478	condition variables with some kind of request results and supporting
		479	callbacks so the caller knows that getting the result will not block,
		480	while still suppporting blocking waits if the caller so desires).
235		481
236	=head1 GLOBALS	482	Another reason I<never> to C<< ->wait >> in a module is that you cannot
		483	sensibly have two C<< ->wait >>'s in parallel, as that would require
		484	multiple interpreters or coroutines/threads, none of which C<AnyEvent>
		485	can supply.
		486
		487	The L<Coro> module, however, I<can> and I<does> supply coroutines and, in
		488	fact, L<Coro::AnyEvent> replaces AnyEvent's condvars by coroutine-safe
		489	versions and also integrates coroutines into AnyEvent, making blocking
		490	C<< ->wait >> calls perfectly safe as long as they are done from another
		491	coroutine (one that doesn't run the event loop).
		492
		493	You can ensure that C<< -wait >> never blocks by setting a callback and
		494	only calling C<< ->wait >> from within that callback (or at a later
		495	time). This will work even when the event loop does not support blocking
		496	waits otherwise.
		497
		498	=item $bool = $cv->ready
		499
		500	Returns true when the condition is "true", i.e. whether C<send> or
		501	C<croak> have been called.
		502
		503	=item $cb = $cv->cb ([new callback])
		504
		505	This is a mutator function that returns the callback set and optionally
		506	replaces it before doing so.
		507
		508	The callback will be called when the condition becomes "true", i.e. when
		509	C<send> or C<croak> are called. Calling C<wait> inside the callback
		510	or at any later time is guaranteed not to block.
		511
		512	=back
		513
		514	=head1 GLOBAL VARIABLES AND FUNCTIONS
237		515
238	=over 4	516	=over 4
239		517
240	=item $AnyEvent::MODEL	518	=item $AnyEvent::MODEL
241		519
…		…
245	C<AnyEvent::Impl:xxx> modules, but can be any other class in the case	523	C<AnyEvent::Impl:xxx> modules, but can be any other class in the case
246	AnyEvent has been extended at runtime (e.g. in I<rxvt-unicode>).	524	AnyEvent has been extended at runtime (e.g. in I<rxvt-unicode>).
247		525
248	The known classes so far are:	526	The known classes so far are:
249		527
250	AnyEvent::Impl::CoroEV based on Coro::EV, best choice.
251	AnyEvent::Impl::EV based on EV (an interface to libev, also best choice).	528	AnyEvent::Impl::EV based on EV (an interface to libev, best choice).
252	AnyEvent::Impl::CoroEvent based on Coro::Event, second best choice.
253	AnyEvent::Impl::Event based on Event, also second best choice :)	529	AnyEvent::Impl::Event based on Event, second best choice.
		530	AnyEvent::Impl::Perl pure-perl implementation, fast and portable.
254	AnyEvent::Impl::Glib based on Glib, second-best choice.	531	AnyEvent::Impl::Glib based on Glib, third-best choice.
255	AnyEvent::Impl::Tk based on Tk, very bad choice.	532	AnyEvent::Impl::Tk based on Tk, very bad choice.
256	AnyEvent::Impl::Perl pure-perl implementation, inefficient.	533	AnyEvent::Impl::Qt based on Qt, cannot be autoprobed (see its docs).
		534	AnyEvent::Impl::EventLib based on Event::Lib, leaks memory and worse.
		535	AnyEvent::Impl::POE based on POE, not generic enough for full support.
		536
		537	There is no support for WxWidgets, as WxWidgets has no support for
		538	watching file handles. However, you can use WxWidgets through the
		539	POE Adaptor, as POE has a Wx backend that simply polls 20 times per
		540	second, which was considered to be too horrible to even consider for
		541	AnyEvent. Likewise, other POE backends can be used by AnyEvent by using
		542	it's adaptor.
		543
		544	AnyEvent knows about L<Prima> and L<Wx> and will try to use L<POE> when
		545	autodetecting them.
257		546
258	=item AnyEvent::detect	547	=item AnyEvent::detect
259		548
260	Returns C<$AnyEvent::MODEL>, forcing autodetection of the event model if	549	Returns C<$AnyEvent::MODEL>, forcing autodetection of the event model
261	necessary. You should only call this function right before you would have	550	if necessary. You should only call this function right before you would
262	created an AnyEvent watcher anyway, that is, very late at runtime.	551	have created an AnyEvent watcher anyway, that is, as late as possible at
		552	runtime.
		553
		554	=item $guard = AnyEvent::post_detect { BLOCK }
		555
		556	Arranges for the code block to be executed as soon as the event model is
		557	autodetected (or immediately if this has already happened).
		558
		559	If called in scalar or list context, then it creates and returns an object
		560	that automatically removes the callback again when it is destroyed. See
		561	L<Coro::BDB> for a case where this is useful.
		562
		563	=item @AnyEvent::post_detect
		564
		565	If there are any code references in this array (you can C<push> to it
		566	before or after loading AnyEvent), then they will called directly after
		567	the event loop has been chosen.
		568
		569	You should check C<$AnyEvent::MODEL> before adding to this array, though:
		570	if it contains a true value then the event loop has already been detected,
		571	and the array will be ignored.
		572
		573	Best use C<AnyEvent::post_detect { BLOCK }> instead.
263		574
264	=back	575	=back
265		576
266	=head1 WHAT TO DO IN A MODULE	577	=head1 WHAT TO DO IN A MODULE
267		578
268	As a module author, you should "use AnyEvent" and call AnyEvent methods	579	As a module author, you should C<use AnyEvent> and call AnyEvent methods
269	freely, but you should not load a specific event module or rely on it.	580	freely, but you should not load a specific event module or rely on it.
270		581
271	Be careful when you create watchers in the module body - Anyevent will	582	Be careful when you create watchers in the module body - AnyEvent will
272	decide which event module to use as soon as the first method is called, so	583	decide which event module to use as soon as the first method is called, so
273	by calling AnyEvent in your module body you force the user of your module	584	by calling AnyEvent in your module body you force the user of your module
274	to load the event module first.	585	to load the event module first.
275		586
		587	Never call C<< ->wait >> on a condition variable unless you I<know> that
		588	the C<< ->send >> method has been called on it already. This is
		589	because it will stall the whole program, and the whole point of using
		590	events is to stay interactive.
		591
		592	It is fine, however, to call C<< ->wait >> when the user of your module
		593	requests it (i.e. if you create a http request object ad have a method
		594	called C<results> that returns the results, it should call C<< ->wait >>
		595	freely, as the user of your module knows what she is doing. always).
		596
276	=head1 WHAT TO DO IN THE MAIN PROGRAM	597	=head1 WHAT TO DO IN THE MAIN PROGRAM
277		598
278	There will always be a single main program - the only place that should	599	There will always be a single main program - the only place that should
279	dictate which event model to use.	600	dictate which event model to use.
280		601
281	If it doesn't care, it can just "use AnyEvent" and use it itself, or not	602	If it doesn't care, it can just "use AnyEvent" and use it itself, or not
282	do anything special and let AnyEvent decide which implementation to chose.	603	do anything special (it does not need to be event-based) and let AnyEvent
		604	decide which implementation to chose if some module relies on it.
283		605
284	If the main program relies on a specific event model (for example, in Gtk2	606	If the main program relies on a specific event model. For example, in
285	programs you have to rely on either Glib or Glib::Event), you should load	607	Gtk2 programs you have to rely on the Glib module. You should load the
286	it before loading AnyEvent or any module that uses it, generally, as early	608	event module before loading AnyEvent or any module that uses it: generally
287	as possible. The reason is that modules might create watchers when they	609	speaking, you should load it as early as possible. The reason is that
288	are loaded, and AnyEvent will decide on the event model to use as soon as	610	modules might create watchers when they are loaded, and AnyEvent will
289	it creates watchers, and it might chose the wrong one unless you load the	611	decide on the event model to use as soon as it creates watchers, and it
290	correct one yourself.	612	might chose the wrong one unless you load the correct one yourself.
291		613
292	You can chose to use a rather inefficient pure-perl implementation by	614	You can chose to use a rather inefficient pure-perl implementation by
293	loading the C<AnyEvent::Impl::Perl> module, but letting AnyEvent chose is	615	loading the C<AnyEvent::Impl::Perl> module, which gives you similar
294	generally better.	616	behaviour everywhere, but letting AnyEvent chose is generally better.
		617
		618	=head1 OTHER MODULES
		619
		620	The following is a non-exhaustive list of additional modules that use
		621	AnyEvent and can therefore be mixed easily with other AnyEvent modules
		622	in the same program. Some of the modules come with AnyEvent, some are
		623	available via CPAN.
		624
		625	=over 4
		626
		627	=item L<AnyEvent::Util>
		628
		629	Contains various utility functions that replace often-used but blocking
		630	functions such as C<inet_aton> by event-/callback-based versions.
		631
		632	=item L<AnyEvent::Handle>
		633
		634	Provide read and write buffers and manages watchers for reads and writes.
		635
		636	=item L<AnyEvent::Socket>
		637
		638	Provides a means to do non-blocking connects, accepts etc.
		639
		640	=item L<AnyEvent::HTTPD>
		641
		642	Provides a simple web application server framework.
		643
		644	=item L<AnyEvent::DNS>
		645
		646	Provides asynchronous DNS resolver capabilities, beyond what
		647	L<AnyEvent::Util> offers.
		648
		649	=item L<AnyEvent::FastPing>
		650
		651	The fastest ping in the west.
		652
		653	=item L<Net::IRC3>
		654
		655	AnyEvent based IRC client module family.
		656
		657	=item L<Net::XMPP2>
		658
		659	AnyEvent based XMPP (Jabber protocol) module family.
		660
		661	=item L<Net::FCP>
		662
		663	AnyEvent-based implementation of the Freenet Client Protocol, birthplace
		664	of AnyEvent.
		665
		666	=item L<Event::ExecFlow>
		667
		668	High level API for event-based execution flow control.
		669
		670	=item L<Coro>
		671
		672	Has special support for AnyEvent via L<Coro::AnyEvent>.
		673
		674	=item L<IO::Lambda>
		675
		676	The lambda approach to I/O - don't ask, look there. Can use AnyEvent.
		677
		678	=item L<IO::AIO>
		679
		680	Truly asynchronous I/O, should be in the toolbox of every event
		681	programmer. Can be trivially made to use AnyEvent.
		682
		683	=item L<BDB>
		684
		685	Truly asynchronous Berkeley DB access. Can be trivially made to use
		686	AnyEvent.
		687
		688	=back
295		689
296	=cut	690	=cut
297		691
298	package AnyEvent;	692	package AnyEvent;
299		693
300	no warnings;	694	no warnings;
301	use strict;	695	use strict;
302		696
303	use Carp;	697	use Carp;
304		698
305	our $VERSION = '3.0';	699	our $VERSION = '3.4';
306	our $MODEL;	700	our $MODEL;
307		701
308	our $AUTOLOAD;	702	our $AUTOLOAD;
309	our @ISA;	703	our @ISA;
310		704
311	our $verbose = $ENV{PERL_ANYEVENT_VERBOSE}*1;	705	our $verbose = $ENV{PERL_ANYEVENT_VERBOSE}*1;
312		706
313	our @REGISTRY;	707	our @REGISTRY;
314		708
315	my @models = (	709	my @models = (
316	[Coro::EV:: => AnyEvent::Impl::CoroEV::],
317	[EV:: => AnyEvent::Impl::EV::],	710	[EV:: => AnyEvent::Impl::EV::],
318	[Coro::Event:: => AnyEvent::Impl::CoroEvent::],
319	[Event:: => AnyEvent::Impl::Event::],	711	[Event:: => AnyEvent::Impl::Event::],
		712	[Tk:: => AnyEvent::Impl::Tk::],
		713	[Wx:: => AnyEvent::Impl::POE::],
		714	[Prima:: => AnyEvent::Impl::POE::],
		715	[AnyEvent::Impl::Perl:: => AnyEvent::Impl::Perl::],
		716	# everything below here will not be autoprobed as the pureperl backend should work everywhere
320	[Glib:: => AnyEvent::Impl::Glib::],	717	[Glib:: => AnyEvent::Impl::Glib::],
321	[Tk:: => AnyEvent::Impl::Tk::],	718	[Event::Lib:: => AnyEvent::Impl::EventLib::], # too buggy
322	[AnyEvent::Impl::Perl:: => AnyEvent::Impl::Perl::],	719	[Qt:: => AnyEvent::Impl::Qt::], # requires special main program
		720	[POE::Kernel:: => AnyEvent::Impl::POE::], # lasciate ogni speranza
323	);	721	);
324		722
325	our %method = map +($_ => 1), qw(io timer condvar broadcast wait signal one_event DESTROY);	723	our %method = map +($_ => 1), qw(io timer signal child condvar one_event DESTROY);
		724
		725	our @post_detect;
		726
		727	sub post_detect(&) {
		728	my ($cb) = @_;
		729
		730	if ($MODEL) {
		731	$cb->();
		732
		733	1
		734	} else {
		735	push @post_detect, $cb;
		736
		737	defined wantarray
		738	? bless \$cb, "AnyEvent::Util::Guard"
		739	: ()
		740	}
		741	}
		742
		743	sub AnyEvent::Util::Guard::DESTROY {
		744	@post_detect = grep $_ != ${$_[0]}, @post_detect;
		745	}
326		746
327	sub detect() {	747	sub detect() {
328	unless ($MODEL) {	748	unless ($MODEL) {
329	no strict 'refs';	749	no strict 'refs';
330		750
		751	if ($ENV{PERL_ANYEVENT_MODEL} =~ /^([a-zA-Z]+)$/) {
		752	my $model = "AnyEvent::Impl::$1";
		753	if (eval "require $model") {
		754	$MODEL = $model;
		755	warn "AnyEvent: loaded model '$model' (forced by \$PERL_ANYEVENT_MODEL), using it.\n" if $verbose > 1;
		756	} else {
		757	warn "AnyEvent: unable to load model '$model' (from \$PERL_ANYEVENT_MODEL):\n$@" if $verbose;
		758	}
		759	}
		760
331	# check for already loaded models	761	# check for already loaded models
		762	unless ($MODEL) {
332	for (@REGISTRY, @models) {	763	for (@REGISTRY, @models) {
333	my ($package, $model) = @$_;	764	my ($package, $model) = @$_;
334	if (${"$package\::VERSION"} > 0) {	765	if (${"$package\::VERSION"} > 0) {
335	if (eval "require $model") {	766	if (eval "require $model") {
336	$MODEL = $model;	767	$MODEL = $model;
337	warn "AnyEvent: found model '$model', using it.\n" if $verbose > 1;	768	warn "AnyEvent: autodetected model '$model', using it.\n" if $verbose > 1;
338	last;	769	last;
		770	}
339	}	771	}
340	}	772	}
341	}
342		773
343	unless ($MODEL) {	774	unless ($MODEL) {
344	# try to load a model	775	# try to load a model
345		776
346	for (@REGISTRY, @models) {	777	for (@REGISTRY, @models) {
347	my ($package, $model) = @$_;	778	my ($package, $model) = @$_;
348	if (eval "require $package"	779	if (eval "require $package"
349	and ${"$package\::VERSION"} > 0	780	and ${"$package\::VERSION"} > 0
350	and eval "require $model") {	781	and eval "require $model") {
351	$MODEL = $model;	782	$MODEL = $model;
352	warn "AnyEvent: autoprobed and loaded model '$model', using it.\n" if $verbose > 1;	783	warn "AnyEvent: autoprobed model '$model', using it.\n" if $verbose > 1;
353	last;	784	last;
		785	}
354	}	786	}
		787
		788	$MODEL
		789	or die "No event module selected for AnyEvent and autodetect failed. Install any one of these modules: EV, Event or Glib.";
355	}	790	}
356
357	$MODEL
358	or die "No event module selected for AnyEvent and autodetect failed. Install any one of these modules: EV (or Coro+EV), Event (or Coro+Event), Glib or Tk.";
359	}	791	}
360		792
361	unshift @ISA, $MODEL;	793	unshift @ISA, $MODEL;
362	push @{"$MODEL\::ISA"}, "AnyEvent::Base";	794	push @{"$MODEL\::ISA"}, "AnyEvent::Base";
		795
		796	(shift @post_detect)->() while @post_detect;
363	}	797	}
364		798
365	$MODEL	799	$MODEL
366	}	800	}
367		801
…		…
474	undef $CHLD_W unless keys %PID_CB;	908	undef $CHLD_W unless keys %PID_CB;
475	}	909	}
476		910
477	=head1 SUPPLYING YOUR OWN EVENT MODEL INTERFACE	911	=head1 SUPPLYING YOUR OWN EVENT MODEL INTERFACE
478		912
		913	This is an advanced topic that you do not normally need to use AnyEvent in
		914	a module. This section is only of use to event loop authors who want to
		915	provide AnyEvent compatibility.
		916
479	If you need to support another event library which isn't directly	917	If you need to support another event library which isn't directly
480	supported by AnyEvent, you can supply your own interface to it by	918	supported by AnyEvent, you can supply your own interface to it by
481	pushing, before the first watcher gets created, the package name of	919	pushing, before the first watcher gets created, the package name of
482	the event module and the package name of the interface to use onto	920	the event module and the package name of the interface to use onto
483	C<@AnyEvent::REGISTRY>. You can do that before and even without loading	921	C<@AnyEvent::REGISTRY>. You can do that before and even without loading
484	AnyEvent.	922	AnyEvent, so it is reasonably cheap.
485		923
486	Example:	924	Example:
487		925
488	push @AnyEvent::REGISTRY, [urxvt => urxvt::anyevent::];	926	push @AnyEvent::REGISTRY, [urxvt => urxvt::anyevent::];
489		927
490	This tells AnyEvent to (literally) use the C<urxvt::anyevent::>	928	This tells AnyEvent to (literally) use the C<urxvt::anyevent::>
491	package/class when it finds the C<urxvt> package/module is loaded. When	929	package/class when it finds the C<urxvt> package/module is already loaded.
		930
492	AnyEvent is loaded and asked to find a suitable event model, it will	931	When AnyEvent is loaded and asked to find a suitable event model, it
493	first check for the presence of urxvt.	932	will first check for the presence of urxvt by trying to C<use> the
		933	C<urxvt::anyevent> module.
494		934
495	The class should provide implementations for all watcher types (see	935	The class should provide implementations for all watcher types. See
496	L<AnyEvent::Impl::Event> (source code), L<AnyEvent::Impl::Glib>	936	L<AnyEvent::Impl::EV> (source code), L<AnyEvent::Impl::Glib> (Source code)
497	(Source code) and so on for actual examples, use C<perldoc -m	937	and so on for actual examples. Use C<perldoc -m AnyEvent::Impl::Glib> to
498	AnyEvent::Impl::Glib> to see the sources).	938	see the sources.
499		939
		940	If you don't provide C<signal> and C<child> watchers than AnyEvent will
		941	provide suitable (hopefully) replacements.
		942
500	The above isn't fictitious, the I<rxvt-unicode> (a.k.a. urxvt)	943	The above example isn't fictitious, the I<rxvt-unicode> (a.k.a. urxvt)
501	uses the above line as-is. An interface isn't included in AnyEvent	944	terminal emulator uses the above line as-is. An interface isn't included
502	because it doesn't make sense outside the embedded interpreter inside	945	in AnyEvent because it doesn't make sense outside the embedded interpreter
503	I<rxvt-unicode>, and it is updated and maintained as part of the	946	inside I<rxvt-unicode>, and it is updated and maintained as part of the
504	I<rxvt-unicode> distribution.	947	I<rxvt-unicode> distribution.
505		948
506	I<rxvt-unicode> also cheats a bit by not providing blocking access to	949	I<rxvt-unicode> also cheats a bit by not providing blocking access to
507	condition variables: code blocking while waiting for a condition will	950	condition variables: code blocking while waiting for a condition will
508	C<die>. This still works with most modules/usages, and blocking calls must	951	C<die>. This still works with most modules/usages, and blocking calls must
509	not be in an interactive application, so it makes sense.	952	not be done in an interactive application, so it makes sense.
510		953
511	=head1 ENVIRONMENT VARIABLES	954	=head1 ENVIRONMENT VARIABLES
512		955
513	The following environment variables are used by this module:	956	The following environment variables are used by this module:
514		957
515	C<PERL_ANYEVENT_VERBOSE> when set to C<2> or higher, reports which event	958	=over 4
516	model gets used.
517		959
		960	=item C<PERL_ANYEVENT_VERBOSE>
		961
		962	By default, AnyEvent will be completely silent except in fatal
		963	conditions. You can set this environment variable to make AnyEvent more
		964	talkative.
		965
		966	When set to C<1> or higher, causes AnyEvent to warn about unexpected
		967	conditions, such as not being able to load the event model specified by
		968	C<PERL_ANYEVENT_MODEL>.
		969
		970	When set to C<2> or higher, cause AnyEvent to report to STDERR which event
		971	model it chooses.
		972
		973	=item C<PERL_ANYEVENT_MODEL>
		974
		975	This can be used to specify the event model to be used by AnyEvent, before
		976	autodetection and -probing kicks in. It must be a string consisting
		977	entirely of ASCII letters. The string C<AnyEvent::Impl::> gets prepended
		978	and the resulting module name is loaded and if the load was successful,
		979	used as event model. If it fails to load AnyEvent will proceed with
		980	autodetection and -probing.
		981
		982	This functionality might change in future versions.
		983
		984	For example, to force the pure perl model (L<AnyEvent::Impl::Perl>) you
		985	could start your program like this:
		986
		987	PERL_ANYEVENT_MODEL=Perl perl ...
		988
		989	=back
		990
518	=head1 EXAMPLE	991	=head1 EXAMPLE PROGRAM
519		992
520	The following program uses an io watcher to read data from stdin, a timer	993	The following program uses an I/O watcher to read data from STDIN, a timer
521	to display a message once per second, and a condvar to exit the program	994	to display a message once per second, and a condition variable to quit the
522	when the user enters quit:	995	program when the user enters quit:
523		996
524	use AnyEvent;	997	use AnyEvent;
525		998
526	my $cv = AnyEvent->condvar;	999	my $cv = AnyEvent->condvar;
527		1000
528	my $io_watcher = AnyEvent->io (fh => \*STDIN, poll => 'r', cb => sub {	1001	my $io_watcher = AnyEvent->io (
		1002	fh => \*STDIN,
		1003	poll => 'r',
		1004	cb => sub {
529	warn "io event <$_[0]>\n"; # will always output <r>	1005	warn "io event <$_[0]>\n"; # will always output <r>
530	chomp (my $input = <STDIN>); # read a line	1006	chomp (my $input = <STDIN>); # read a line
531	warn "read: $input\n"; # output what has been read	1007	warn "read: $input\n"; # output what has been read
532	$cv->broadcast if $input =~ /^q/i; # quit program if /^q/i	1008	$cv->broadcast if $input =~ /^q/i; # quit program if /^q/i
		1009	},
533	});	1010	);
534		1011
535	my $time_watcher; # can only be used once	1012	my $time_watcher; # can only be used once
536		1013
537	sub new_timer {	1014	sub new_timer {
538	$timer = AnyEvent->timer (after => 1, cb => sub {	1015	$timer = AnyEvent->timer (after => 1, cb => sub {
…		…
620	$txn->{finished}->wait;	1097	$txn->{finished}->wait;
621	return $txn->{result};	1098	return $txn->{result};
622		1099
623	The actual code goes further and collects all errors (C<die>s, exceptions)	1100	The actual code goes further and collects all errors (C<die>s, exceptions)
624	that occured during request processing. The C<result> method detects	1101	that occured during request processing. The C<result> method detects
625	wether an exception as thrown (it is stored inside the $txn object)	1102	whether an exception as thrown (it is stored inside the $txn object)
626	and just throws the exception, which means connection errors and other	1103	and just throws the exception, which means connection errors and other
627	problems get reported tot he code that tries to use the result, not in a	1104	problems get reported tot he code that tries to use the result, not in a
628	random callback.	1105	random callback.
629		1106
630	All of this enables the following usage styles:	1107	All of this enables the following usage styles:
631		1108
632	1. Blocking:	1109	1. Blocking:
633		1110
634	my $data = $fcp->client_get ($url);	1111	my $data = $fcp->client_get ($url);
635		1112
636	2. Blocking, but parallelizing:	1113	2. Blocking, but running in parallel:
637		1114
638	my @datas = map $_->result,	1115	my @datas = map $_->result,
639	map $fcp->txn_client_get ($_),	1116	map $fcp->txn_client_get ($_),
640	@urls;	1117	@urls;
641		1118
642	Both blocking examples work without the module user having to know	1119	Both blocking examples work without the module user having to know
643	anything about events.	1120	anything about events.
644		1121
645	3a. Event-based in a main program, using any support Event module:	1122	3a. Event-based in a main program, using any supported event module:
646		1123
647	use Event;	1124	use EV;
648		1125
649	$fcp->txn_client_get ($url)->cb (sub {	1126	$fcp->txn_client_get ($url)->cb (sub {
650	my $txn = shift;	1127	my $txn = shift;
651	my $data = $txn->result;	1128	my $data = $txn->result;
652	...	1129	...
653	});	1130	});
654		1131
655	Event::loop;	1132	EV::loop;
656		1133
657	3b. The module user could use AnyEvent, too:	1134	3b. The module user could use AnyEvent, too:
658		1135
659	use AnyEvent;	1136	use AnyEvent;
660		1137
…		…
665	$quit->broadcast;	1142	$quit->broadcast;
666	});	1143	});
667		1144
668	$quit->wait;	1145	$quit->wait;
669		1146
		1147
		1148	=head1 BENCHMARKS
		1149
		1150	To give you an idea of the performance and overheads that AnyEvent adds
		1151	over the event loops themselves and to give you an impression of the speed
		1152	of various event loops I prepared some benchmarks.
		1153
		1154	=head2 BENCHMARKING ANYEVENT OVERHEAD
		1155
		1156	Here is a benchmark of various supported event models used natively and
		1157	through anyevent. The benchmark creates a lot of timers (with a zero
		1158	timeout) and I/O watchers (watching STDOUT, a pty, to become writable,
		1159	which it is), lets them fire exactly once and destroys them again.
		1160
		1161	Source code for this benchmark is found as F<eg/bench> in the AnyEvent
		1162	distribution.
		1163
		1164	=head3 Explanation of the columns
		1165
		1166	I<watcher> is the number of event watchers created/destroyed. Since
		1167	different event models feature vastly different performances, each event
		1168	loop was given a number of watchers so that overall runtime is acceptable
		1169	and similar between tested event loop (and keep them from crashing): Glib
		1170	would probably take thousands of years if asked to process the same number
		1171	of watchers as EV in this benchmark.
		1172
		1173	I<bytes> is the number of bytes (as measured by the resident set size,
		1174	RSS) consumed by each watcher. This method of measuring captures both C
		1175	and Perl-based overheads.
		1176
		1177	I<create> is the time, in microseconds (millionths of seconds), that it
		1178	takes to create a single watcher. The callback is a closure shared between
		1179	all watchers, to avoid adding memory overhead. That means closure creation
		1180	and memory usage is not included in the figures.
		1181
		1182	I<invoke> is the time, in microseconds, used to invoke a simple
		1183	callback. The callback simply counts down a Perl variable and after it was
		1184	invoked "watcher" times, it would C<< ->broadcast >> a condvar once to
		1185	signal the end of this phase.
		1186
		1187	I<destroy> is the time, in microseconds, that it takes to destroy a single
		1188	watcher.
		1189
		1190	=head3 Results
		1191
		1192	name watchers bytes create invoke destroy comment
		1193	EV/EV 400000 244 0.56 0.46 0.31 EV native interface
		1194	EV/Any 100000 244 2.50 0.46 0.29 EV + AnyEvent watchers
		1195	CoroEV/Any 100000 244 2.49 0.44 0.29 coroutines + Coro::Signal
		1196	Perl/Any 100000 513 4.92 0.87 1.12 pure perl implementation
		1197	Event/Event 16000 516 31.88 31.30 0.85 Event native interface
		1198	Event/Any 16000 590 35.75 31.42 1.08 Event + AnyEvent watchers
		1199	Glib/Any 16000 1357 98.22 12.41 54.00 quadratic behaviour
		1200	Tk/Any 2000 1860 26.97 67.98 14.00 SEGV with >> 2000 watchers
		1201	POE/Event 2000 6644 108.64 736.02 14.73 via POE::Loop::Event
		1202	POE/Select 2000 6343 94.13 809.12 565.96 via POE::Loop::Select
		1203
		1204	=head3 Discussion
		1205
		1206	The benchmark does I<not> measure scalability of the event loop very
		1207	well. For example, a select-based event loop (such as the pure perl one)
		1208	can never compete with an event loop that uses epoll when the number of
		1209	file descriptors grows high. In this benchmark, all events become ready at
		1210	the same time, so select/poll-based implementations get an unnatural speed
		1211	boost.
		1212
		1213	Also, note that the number of watchers usually has a nonlinear effect on
		1214	overall speed, that is, creating twice as many watchers doesn't take twice
		1215	the time - usually it takes longer. This puts event loops tested with a
		1216	higher number of watchers at a disadvantage.
		1217
		1218	To put the range of results into perspective, consider that on the
		1219	benchmark machine, handling an event takes roughly 1600 CPU cycles with
		1220	EV, 3100 CPU cycles with AnyEvent's pure perl loop and almost 3000000 CPU
		1221	cycles with POE.
		1222
		1223	C<EV> is the sole leader regarding speed and memory use, which are both
		1224	maximal/minimal, respectively. Even when going through AnyEvent, it uses
		1225	far less memory than any other event loop and is still faster than Event
		1226	natively.
		1227
		1228	The pure perl implementation is hit in a few sweet spots (both the
		1229	constant timeout and the use of a single fd hit optimisations in the perl
		1230	interpreter and the backend itself). Nevertheless this shows that it
		1231	adds very little overhead in itself. Like any select-based backend its
		1232	performance becomes really bad with lots of file descriptors (and few of
		1233	them active), of course, but this was not subject of this benchmark.
		1234
		1235	The C<Event> module has a relatively high setup and callback invocation
		1236	cost, but overall scores in on the third place.
		1237
		1238	C<Glib>'s memory usage is quite a bit higher, but it features a
		1239	faster callback invocation and overall ends up in the same class as
		1240	C<Event>. However, Glib scales extremely badly, doubling the number of
		1241	watchers increases the processing time by more than a factor of four,
		1242	making it completely unusable when using larger numbers of watchers
		1243	(note that only a single file descriptor was used in the benchmark, so
		1244	inefficiencies of C<poll> do not account for this).
		1245
		1246	The C<Tk> adaptor works relatively well. The fact that it crashes with
		1247	more than 2000 watchers is a big setback, however, as correctness takes
		1248	precedence over speed. Nevertheless, its performance is surprising, as the
		1249	file descriptor is dup()ed for each watcher. This shows that the dup()
		1250	employed by some adaptors is not a big performance issue (it does incur a
		1251	hidden memory cost inside the kernel which is not reflected in the figures
		1252	above).
		1253
		1254	C<POE>, regardless of underlying event loop (whether using its pure perl
		1255	select-based backend or the Event module, the POE-EV backend couldn't
		1256	be tested because it wasn't working) shows abysmal performance and
		1257	memory usage with AnyEvent: Watchers use almost 30 times as much memory
		1258	as EV watchers, and 10 times as much memory as Event (the high memory
		1259	requirements are caused by requiring a session for each watcher). Watcher
		1260	invocation speed is almost 900 times slower than with AnyEvent's pure perl
		1261	implementation.
		1262
		1263	The design of the POE adaptor class in AnyEvent can not really account
		1264	for the performance issues, though, as session creation overhead is
		1265	small compared to execution of the state machine, which is coded pretty
		1266	optimally within L<AnyEvent::Impl::POE> (and while everybody agrees that
		1267	using multiple sessions is not a good approach, especially regarding
		1268	memory usage, even the author of POE could not come up with a faster
		1269	design).
		1270
		1271	=head3 Summary
		1272
		1273	=over 4
		1274
		1275	=item * Using EV through AnyEvent is faster than any other event loop
		1276	(even when used without AnyEvent), but most event loops have acceptable
		1277	performance with or without AnyEvent.
		1278
		1279	=item * The overhead AnyEvent adds is usually much smaller than the overhead of
		1280	the actual event loop, only with extremely fast event loops such as EV
		1281	adds AnyEvent significant overhead.
		1282
		1283	=item * You should avoid POE like the plague if you want performance or
		1284	reasonable memory usage.
		1285
		1286	=back
		1287
		1288	=head2 BENCHMARKING THE LARGE SERVER CASE
		1289
		1290	This benchmark atcually benchmarks the event loop itself. It works by
		1291	creating a number of "servers": each server consists of a socketpair, a
		1292	timeout watcher that gets reset on activity (but never fires), and an I/O
		1293	watcher waiting for input on one side of the socket. Each time the socket
		1294	watcher reads a byte it will write that byte to a random other "server".
		1295
		1296	The effect is that there will be a lot of I/O watchers, only part of which
		1297	are active at any one point (so there is a constant number of active
		1298	fds for each loop iterstaion, but which fds these are is random). The
		1299	timeout is reset each time something is read because that reflects how
		1300	most timeouts work (and puts extra pressure on the event loops).
		1301
		1302	In this benchmark, we use 10000 socketpairs (20000 sockets), of which 100
		1303	(1%) are active. This mirrors the activity of large servers with many
		1304	connections, most of which are idle at any one point in time.
		1305
		1306	Source code for this benchmark is found as F<eg/bench2> in the AnyEvent
		1307	distribution.
		1308
		1309	=head3 Explanation of the columns
		1310
		1311	I<sockets> is the number of sockets, and twice the number of "servers" (as
		1312	each server has a read and write socket end).
		1313
		1314	I<create> is the time it takes to create a socketpair (which is
		1315	nontrivial) and two watchers: an I/O watcher and a timeout watcher.
		1316
		1317	I<request>, the most important value, is the time it takes to handle a
		1318	single "request", that is, reading the token from the pipe and forwarding
		1319	it to another server. This includes deleting the old timeout and creating
		1320	a new one that moves the timeout into the future.
		1321
		1322	=head3 Results
		1323
		1324	name sockets create request
		1325	EV 20000 69.01 11.16
		1326	Perl 20000 73.32 35.87
		1327	Event 20000 212.62 257.32
		1328	Glib 20000 651.16 1896.30
		1329	POE 20000 349.67 12317.24 uses POE::Loop::Event
		1330
		1331	=head3 Discussion
		1332
		1333	This benchmark I<does> measure scalability and overall performance of the
		1334	particular event loop.
		1335
		1336	EV is again fastest. Since it is using epoll on my system, the setup time
		1337	is relatively high, though.
		1338
		1339	Perl surprisingly comes second. It is much faster than the C-based event
		1340	loops Event and Glib.
		1341
		1342	Event suffers from high setup time as well (look at its code and you will
		1343	understand why). Callback invocation also has a high overhead compared to
		1344	the C<< $_->() for .. >>-style loop that the Perl event loop uses. Event
		1345	uses select or poll in basically all documented configurations.
		1346
		1347	Glib is hit hard by its quadratic behaviour w.r.t. many watchers. It
		1348	clearly fails to perform with many filehandles or in busy servers.
		1349
		1350	POE is still completely out of the picture, taking over 1000 times as long
		1351	as EV, and over 100 times as long as the Perl implementation, even though
		1352	it uses a C-based event loop in this case.
		1353
		1354	=head3 Summary
		1355
		1356	=over 4
		1357
		1358	=item * The pure perl implementation performs extremely well.
		1359
		1360	=item * Avoid Glib or POE in large projects where performance matters.
		1361
		1362	=back
		1363
		1364	=head2 BENCHMARKING SMALL SERVERS
		1365
		1366	While event loops should scale (and select-based ones do not...) even to
		1367	large servers, most programs we (or I :) actually write have only a few
		1368	I/O watchers.
		1369
		1370	In this benchmark, I use the same benchmark program as in the large server
		1371	case, but it uses only eight "servers", of which three are active at any
		1372	one time. This should reflect performance for a small server relatively
		1373	well.
		1374
		1375	The columns are identical to the previous table.
		1376
		1377	=head3 Results
		1378
		1379	name sockets create request
		1380	EV 16 20.00 6.54
		1381	Perl 16 25.75 12.62
		1382	Event 16 81.27 35.86
		1383	Glib 16 32.63 15.48
		1384	POE 16 261.87 276.28 uses POE::Loop::Event
		1385
		1386	=head3 Discussion
		1387
		1388	The benchmark tries to test the performance of a typical small
		1389	server. While knowing how various event loops perform is interesting, keep
		1390	in mind that their overhead in this case is usually not as important, due
		1391	to the small absolute number of watchers (that is, you need efficiency and
		1392	speed most when you have lots of watchers, not when you only have a few of
		1393	them).
		1394
		1395	EV is again fastest.
		1396
		1397	Perl again comes second. It is noticably faster than the C-based event
		1398	loops Event and Glib, although the difference is too small to really
		1399	matter.
		1400
		1401	POE also performs much better in this case, but is is still far behind the
		1402	others.
		1403
		1404	=head3 Summary
		1405
		1406	=over 4
		1407
		1408	=item * C-based event loops perform very well with small number of
		1409	watchers, as the management overhead dominates.
		1410
		1411	=back
		1412
		1413
		1414	=head1 FORK
		1415
		1416	Most event libraries are not fork-safe. The ones who are usually are
		1417	because they rely on inefficient but fork-safe C<select> or C<poll>
		1418	calls. Only L<EV> is fully fork-aware.
		1419
		1420	If you have to fork, you must either do so I<before> creating your first
		1421	watcher OR you must not use AnyEvent at all in the child.
		1422
		1423
		1424	=head1 SECURITY CONSIDERATIONS
		1425
		1426	AnyEvent can be forced to load any event model via
		1427	$ENV{PERL_ANYEVENT_MODEL}. While this cannot (to my knowledge) be used to
		1428	execute arbitrary code or directly gain access, it can easily be used to
		1429	make the program hang or malfunction in subtle ways, as AnyEvent watchers
		1430	will not be active when the program uses a different event model than
		1431	specified in the variable.
		1432
		1433	You can make AnyEvent completely ignore this variable by deleting it
		1434	before the first watcher gets created, e.g. with a C<BEGIN> block:
		1435
		1436	BEGIN { delete $ENV{PERL_ANYEVENT_MODEL} }
		1437
		1438	use AnyEvent;
		1439
		1440	Similar considerations apply to $ENV{PERL_ANYEVENT_VERBOSE}, as that can
		1441	be used to probe what backend is used and gain other information (which is
		1442	probably even less useful to an attacker than PERL_ANYEVENT_MODEL).
		1443
		1444
670	=head1 SEE ALSO	1445	=head1 SEE ALSO
671		1446
672	Event modules: L<Coro::Event>, L<Coro>, L<Event>, L<Glib::Event>, L<Glib>.	1447	Event modules: L<EV>, L<EV::Glib>, L<Glib::EV>, L<Event>, L<Glib::Event>,
		1448	L<Glib>, L<Tk>, L<Event::Lib>, L<Qt>, L<POE>.
673		1449
674	Implementations: L<AnyEvent::Impl::Coro>, L<AnyEvent::Impl::Event>, L<AnyEvent::Impl::Glib>, L<AnyEvent::Impl::Tk>.	1450	Implementations: L<AnyEvent::Impl::EV>, L<AnyEvent::Impl::Event>,
		1451	L<AnyEvent::Impl::Glib>, L<AnyEvent::Impl::Tk>, L<AnyEvent::Impl::Perl>,
		1452	L<AnyEvent::Impl::EventLib>, L<AnyEvent::Impl::Qt>,
		1453	L<AnyEvent::Impl::POE>.
675		1454
		1455	Coroutine support: L<Coro>, L<Coro::AnyEvent>, L<Coro::EV>, L<Coro::Event>,
		1456
676	Nontrivial usage example: L<Net::FCP>.	1457	Nontrivial usage examples: L<Net::FCP>, L<Net::XMPP2>.
677		1458
678	=head1	1459
		1460	=head1 AUTHOR
		1461
		1462	Marc Lehmann <schmorp@schmorp.de>
		1463	http://home.schmorp.de/
679		1464
680	=cut	1465	=cut
681		1466
682	1	1467	1
683		1468

Diff Legend

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

Comparing AnyEvent/lib/AnyEvent.pm (file contents): Revision 1.41 by root, Mon Apr 7 19:23:59 2008 UTC vs. Revision 1.112 by root, Sat May 10 01:04:42 2008 UTC

Diff Legend

Comparing AnyEvent/lib/AnyEvent.pm (file contents):
Revision 1.41 by root, Mon Apr 7 19:23:59 2008 UTC vs.
Revision 1.112 by root, Sat May 10 01:04:42 2008 UTC