ViewVC Help

View File | Revision Log | Show Annotations | Download File

/cvs/AnyEvent/lib/AnyEvent.pm

Comparing AnyEvent/lib/AnyEvent.pm (file contents):
Revision 1.89 by root, Fri Apr 25 14:19:23 2008 UTC vs.
Revision 1.181 by root, Sat Sep 6 10:54:32 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	EV, Event, Coro::EV, Coro::Event, Glib, Tk, Perl, Event::Lib, Qt, POE - 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
11	my $w = AnyEvent->io (fh => $fh, poll => "r|w", cb => sub {	11	my $w = AnyEvent->io (fh => $fh, poll => "r|w", cb => sub { ... });
		12
		13	my $w = AnyEvent->timer (after => $seconds, cb => sub { ... });
		14	my $w = AnyEvent->timer (after => $seconds, interval => $seconds, cb => ...
		15
		16	print AnyEvent->now; # prints current event loop time
		17	print AnyEvent->time; # think Time::HiRes::time or simply CORE::time.
		18
		19	my $w = AnyEvent->signal (signal => "TERM", cb => sub { ... });
		20
		21	my $w = AnyEvent->child (pid => $pid, cb => sub {
		22	my ($pid, $status) = @_;
12	...	23	...
13	});	24	});
14		25
15	my $w = AnyEvent->timer (after => $seconds, cb => sub {
16	...
17	});
18
19	my $w = AnyEvent->condvar; # stores whether a condition was flagged	26	my $w = AnyEvent->condvar; # stores whether a condition was flagged
		27	$w->send; # wake up current and all future recv's
20	$w->wait; # enters "main loop" till $condvar gets ->broadcast	28	$w->recv; # enters "main loop" till $condvar gets ->send
21	$w->broadcast; # wake up current and all future wait's	29	# use a condvar in callback mode:
		30	$w->cb (sub { $_[0]->recv });
		31
		32	=head1 INTRODUCTION/TUTORIAL
		33
		34	This manpage is mainly a reference manual. If you are interested
		35	in a tutorial or some gentle introduction, have a look at the
		36	L<AnyEvent::Intro> manpage.
22		37
23	=head1 WHY YOU SHOULD USE THIS MODULE (OR NOT)	38	=head1 WHY YOU SHOULD USE THIS MODULE (OR NOT)
24		39
25	Glib, POE, IO::Async, Event... CPAN offers event models by the dozen	40	Glib, POE, IO::Async, Event... CPAN offers event models by the dozen
26	nowadays. So what is different about AnyEvent?	41	nowadays. So what is different about AnyEvent?
27		42
28	Executive Summary: AnyEvent is I<compatible>, AnyEvent is I<free of	43	Executive Summary: AnyEvent is I<compatible>, AnyEvent is I<free of
29	policy> and AnyEvent is I<small and efficient>.	44	policy> and AnyEvent is I<small and efficient>.
30		45
31	First and foremost, I<AnyEvent is not an event model> itself, it only	46	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	47	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,	48	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,	49	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	50	only one event loop can be active at the same time in a process. AnyEvent
36	helps hiding the differences between those event loops.	51	cannot change this, but it can hide the differences between those event
		52	loops.
37		53
38	The goal of AnyEvent is to offer module authors the ability to do event	54	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	55	programming (waiting for I/O or timer events) without subscribing to a
40	religion, a way of living, and most importantly: without forcing your	56	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	57	module users into the same thing by forcing them to use the same event
42	model you use.	58	model you use.
43		59
44	For modules like POE or IO::Async (which is a total misnomer as it is	60	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	61	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	62	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	63	cannot use anything else, as they are simply incompatible to everything
48	isn't itself. What's worse, all the potential users of your module are	64	that isn't them. What's worse, all the potential users of your
49	I<also> forced to use the same event loop you use.	65	module are I<also> forced to use the same event loop you use.
50		66
51	AnyEvent is different: AnyEvent + POE works fine. AnyEvent + Glib works	67	AnyEvent is different: AnyEvent + POE works fine. AnyEvent + Glib works
52	fine. AnyEvent + Tk works fine etc. etc. but none of these work together	68	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	69	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,	70	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	71	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	72	event models it supports (including stuff like IO::Async, as long as those
57	as those use one of the supported event loops. It is trivial to add new	73	use one of the supported event loops. It is trivial to add new event loops
58	event loops to AnyEvent, too, so it is future-proof).	74	to AnyEvent, too, so it is future-proof).
59		75
60	In addition to being free of having to use I<the one and only true event	76	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	77	model>, AnyEvent also is free of bloat and policy: with POE or similar
62	modules, you get an enourmous amount of code and strict rules you have to	78	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	79	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	80	offering the functionality that is necessary, in as thin as a wrapper as
65	technically possible.	81	technically possible.
66		82
		83	Of course, AnyEvent comes with a big (and fully optional!) toolbox
		84	of useful functionality, such as an asynchronous DNS resolver, 100%
		85	non-blocking connects (even with TLS/SSL, IPv6 and on broken platforms
		86	such as Windows) and lots of real-world knowledge and workarounds for
		87	platform bugs and differences.
		88
67	Of course, if you want lots of policy (this can arguably be somewhat	89	Now, if you I<do 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	90	useful) and you want to force your users to use the one and only event
69	model, you should I<not> use this module.	91	model, you should I<not> use this module.
70
71		92
72	=head1 DESCRIPTION	93	=head1 DESCRIPTION
73		94
74	L<AnyEvent> provides an identical interface to multiple event loops. This	95	L<AnyEvent> provides an identical interface to multiple event loops. This
75	allows module authors to utilise an event loop without forcing module	96	allows module authors to utilise an event loop without forcing module
…		…
79	The interface itself is vaguely similar, but not identical to the L<Event>	100	The interface itself is vaguely similar, but not identical to the L<Event>
80	module.	101	module.
81		102
82	During the first call of any watcher-creation method, the module tries	103	During the first call of any watcher-creation method, the module tries
83	to detect the currently loaded event loop by probing whether one of the	104	to detect the currently loaded event loop by probing whether one of the
84	following modules is already loaded: L<Coro::EV>, L<Coro::Event>, L<EV>,	105	following modules is already loaded: L<EV>,
85	L<Event>, L<Glib>, L<AnyEvent::Impl::Perl>, L<Tk>, L<Event::Lib>, L<Qt>,	106	L<Event>, L<Glib>, L<AnyEvent::Impl::Perl>, L<Tk>, L<Event::Lib>, L<Qt>,
86	L<POE>. The first one found is used. If none are found, the module tries	107	L<POE>. The first one found is used. If none are found, the module tries
87	to load these modules (excluding Tk, Event::Lib, Qt and POE as the pure perl	108	to load these modules (excluding Tk, Event::Lib, Qt and POE as the pure perl
88	adaptor should always succeed) in the order given. The first one that can	109	adaptor should always succeed) in the order given. The first one that can
89	be successfully loaded will be used. If, after this, still none could be	110	be successfully loaded will be used. If, after this, still none could be
…		…
103	starts using it, all bets are off. Maybe you should tell their authors to	124	starts using it, all bets are off. Maybe you should tell their authors to
104	use AnyEvent so their modules work together with others seamlessly...	125	use AnyEvent so their modules work together with others seamlessly...
105		126
106	The pure-perl implementation of AnyEvent is called	127	The pure-perl implementation of AnyEvent is called
107	C<AnyEvent::Impl::Perl>. Like other event modules you can load it	128	C<AnyEvent::Impl::Perl>. Like other event modules you can load it
108	explicitly.	129	explicitly and enjoy the high availability of that event loop :)
109		130
110	=head1 WATCHERS	131	=head1 WATCHERS
111		132
112	AnyEvent has the central concept of a I<watcher>, which is an object that	133	AnyEvent has the central concept of a I<watcher>, which is an object that
113	stores relevant data for each kind of event you are waiting for, such as	134	stores relevant data for each kind of event you are waiting for, such as
114	the callback to call, the filehandle to watch, etc.	135	the callback to call, the file handle to watch, etc.
115		136
116	These watchers are normal Perl objects with normal Perl lifetime. After	137	These watchers are normal Perl objects with normal Perl lifetime. After
117	creating a watcher it will immediately "watch" for events and invoke the	138	creating a watcher it will immediately "watch" for events and invoke the
118	callback when the event occurs (of course, only when the event model	139	callback when the event occurs (of course, only when the event model
119	is in control).	140	is in control).
…		…
127	Many watchers either are used with "recursion" (repeating timers for	148	Many watchers either are used with "recursion" (repeating timers for
128	example), or need to refer to their watcher object in other ways.	149	example), or need to refer to their watcher object in other ways.
129		150
130	An any way to achieve that is this pattern:	151	An any way to achieve that is this pattern:
131		152
132	my $w; $w = AnyEvent->type (arg => value ..., cb => sub {	153	my $w; $w = AnyEvent->type (arg => value ..., cb => sub {
133	# you can use $w here, for example to undef it	154	# you can use $w here, for example to undef it
134	undef $w;	155	undef $w;
135	});	156	});
136		157
137	Note that C<my $w; $w => combination. This is necessary because in Perl,	158	Note that C<my $w; $w => combination. This is necessary because in Perl,
138	my variables are only visible after the statement in which they are	159	my variables are only visible after the statement in which they are
139	declared.	160	declared.
140		161
141	=head2 I/O WATCHERS	162	=head2 I/O WATCHERS
142		163
143	You can create an I/O watcher by calling the C<< AnyEvent->io >> method	164	You can create an I/O watcher by calling the C<< AnyEvent->io >> method
144	with the following mandatory key-value pairs as arguments:	165	with the following mandatory key-value pairs as arguments:
145		166
146	C<fh> the Perl I<file handle> (I<not> file descriptor) to watch	167	C<fh> the Perl I<file handle> (I<not> file descriptor) to watch for events
147	for events. C<poll> must be a string that is either C<r> or C<w>,	168	(AnyEvent might or might not keep a reference to this file handle). C<poll>
148	which creates a watcher waiting for "r"eadable or "w"ritable events,	169	must be a string that is either C<r> or C<w>, which creates a watcher
149	respectively. C<cb> is the callback to invoke each time the file handle	170	waiting for "r"eadable or "w"ritable events, respectively. C<cb> is the
150	becomes ready.	171	callback to invoke each time the file handle becomes ready.
151		172
152	Although the callback might get passed parameters, their value and	173	Although the callback might get passed parameters, their value and
153	presence is undefined and you cannot rely on them. Portable AnyEvent	174	presence is undefined and you cannot rely on them. Portable AnyEvent
154	callbacks cannot use arguments passed to I/O watcher callbacks.	175	callbacks cannot use arguments passed to I/O watcher callbacks.
155		176
…		…
159		180
160	Some event loops issue spurious readyness notifications, so you should	181	Some event loops issue spurious readyness notifications, so you should
161	always use non-blocking calls when reading/writing from/to your file	182	always use non-blocking calls when reading/writing from/to your file
162	handles.	183	handles.
163		184
164	Example:
165
166	# wait for readability of STDIN, then read a line and disable the watcher	185	Example: wait for readability of STDIN, then read a line and disable the
		186	watcher.
		187
167	my $w; $w = AnyEvent->io (fh => \*STDIN, poll => 'r', cb => sub {	188	my $w; $w = AnyEvent->io (fh => \*STDIN, poll => 'r', cb => sub {
168	chomp (my $input = <STDIN>);	189	chomp (my $input = <STDIN>);
169	warn "read: $input\n";	190	warn "read: $input\n";
170	undef $w;	191	undef $w;
171	});	192	});
…		…
181		202
182	Although the callback might get passed parameters, their value and	203	Although the callback might get passed parameters, their value and
183	presence is undefined and you cannot rely on them. Portable AnyEvent	204	presence is undefined and you cannot rely on them. Portable AnyEvent
184	callbacks cannot use arguments passed to time watcher callbacks.	205	callbacks cannot use arguments passed to time watcher callbacks.
185		206
186	The timer callback will be invoked at most once: if you want a repeating	207	The callback will normally be invoked once only. If you specify another
187	timer you have to create a new watcher (this is a limitation by both Tk	208	parameter, C<interval>, as a strictly positive number (> 0), then the
188	and Glib).	209	callback will be invoked regularly at that interval (in fractional
		210	seconds) after the first invocation. If C<interval> is specified with a
		211	false value, then it is treated as if it were missing.
189		212
190	Example:	213	The callback will be rescheduled before invoking the callback, but no
		214	attempt is done to avoid timer drift in most backends, so the interval is
		215	only approximate.
191		216
192	# fire an event after 7.7 seconds	217	Example: fire an event after 7.7 seconds.
		218
193	my $w = AnyEvent->timer (after => 7.7, cb => sub {	219	my $w = AnyEvent->timer (after => 7.7, cb => sub {
194	warn "timeout\n";	220	warn "timeout\n";
195	});	221	});
196		222
197	# to cancel the timer:	223	# to cancel the timer:
198	undef $w;	224	undef $w;
199		225
200	Example 2:
201
202	# fire an event after 0.5 seconds, then roughly every second	226	Example 2: fire an event after 0.5 seconds, then roughly every second.
203	my $w;
204		227
205	my $cb = sub {
206	# cancel the old timer while creating a new one
207	$w = AnyEvent->timer (after => 1, cb => $cb);	228	my $w = AnyEvent->timer (after => 0.5, interval => 1, cb => sub {
		229	warn "timeout\n";
208	};	230	};
209
210	# start the "loop" by creating the first watcher
211	$w = AnyEvent->timer (after => 0.5, cb => $cb);
212		231
213	=head3 TIMING ISSUES	232	=head3 TIMING ISSUES
214		233
215	There are two ways to handle timers: based on real time (relative, "fire	234	There are two ways to handle timers: based on real time (relative, "fire
216	in 10 seconds") and based on wallclock time (absolute, "fire at 12	235	in 10 seconds") and based on wallclock time (absolute, "fire at 12
…		…
228	timers.	247	timers.
229		248
230	AnyEvent always prefers relative timers, if available, matching the	249	AnyEvent always prefers relative timers, if available, matching the
231	AnyEvent API.	250	AnyEvent API.
232		251
		252	AnyEvent has two additional methods that return the "current time":
		253
		254	=over 4
		255
		256	=item AnyEvent->time
		257
		258	This returns the "current wallclock time" as a fractional number of
		259	seconds since the Epoch (the same thing as C<time> or C<Time::HiRes::time>
		260	return, and the result is guaranteed to be compatible with those).
		261
		262	It progresses independently of any event loop processing, i.e. each call
		263	will check the system clock, which usually gets updated frequently.
		264
		265	=item AnyEvent->now
		266
		267	This also returns the "current wallclock time", but unlike C<time>, above,
		268	this value might change only once per event loop iteration, depending on
		269	the event loop (most return the same time as C<time>, above). This is the
		270	time that AnyEvent's timers get scheduled against.
		271
		272	I<In almost all cases (in all cases if you don't care), this is the
		273	function to call when you want to know the current time.>
		274
		275	This function is also often faster then C<< AnyEvent->time >>, and
		276	thus the preferred method if you want some timestamp (for example,
		277	L<AnyEvent::Handle> uses this to update it's activity timeouts).
		278
		279	The rest of this section is only of relevance if you try to be very exact
		280	with your timing, you can skip it without bad conscience.
		281
		282	For a practical example of when these times differ, consider L<Event::Lib>
		283	and L<EV> and the following set-up:
		284
		285	The event loop is running and has just invoked one of your callback at
		286	time=500 (assume no other callbacks delay processing). In your callback,
		287	you wait a second by executing C<sleep 1> (blocking the process for a
		288	second) and then (at time=501) you create a relative timer that fires
		289	after three seconds.
		290
		291	With L<Event::Lib>, C<< AnyEvent->time >> and C<< AnyEvent->now >> will
		292	both return C<501>, because that is the current time, and the timer will
		293	be scheduled to fire at time=504 (C<501> + C<3>).
		294
		295	With L<EV>, C<< AnyEvent->time >> returns C<501> (as that is the current
		296	time), but C<< AnyEvent->now >> returns C<500>, as that is the time the
		297	last event processing phase started. With L<EV>, your timer gets scheduled
		298	to run at time=503 (C<500> + C<3>).
		299
		300	In one sense, L<Event::Lib> is more exact, as it uses the current time
		301	regardless of any delays introduced by event processing. However, most
		302	callbacks do not expect large delays in processing, so this causes a
		303	higher drift (and a lot more system calls to get the current time).
		304
		305	In another sense, L<EV> is more exact, as your timer will be scheduled at
		306	the same time, regardless of how long event processing actually took.
		307
		308	In either case, if you care (and in most cases, you don't), then you
		309	can get whatever behaviour you want with any event loop, by taking the
		310	difference between C<< AnyEvent->time >> and C<< AnyEvent->now >> into
		311	account.
		312
		313	=back
		314
233	=head2 SIGNAL WATCHERS	315	=head2 SIGNAL WATCHERS
234		316
235	You can watch for signals using a signal watcher, C<signal> is the signal	317	You can watch for signals using a signal watcher, C<signal> is the signal
236	I<name> without any C<SIG> prefix, C<cb> is the Perl callback to	318	I<name> in uppercase and without any C<SIG> prefix, C<cb> is the Perl
237	be invoked whenever a signal occurs.	319	callback to be invoked whenever a signal occurs.
238		320
239	Although the callback might get passed parameters, their value and	321	Although the callback might get passed parameters, their value and
240	presence is undefined and you cannot rely on them. Portable AnyEvent	322	presence is undefined and you cannot rely on them. Portable AnyEvent
241	callbacks cannot use arguments passed to signal watcher callbacks.	323	callbacks cannot use arguments passed to signal watcher callbacks.
242		324
243	Multiple signal occurances can be clumped together into one callback	325	Multiple signal occurrences can be clumped together into one callback
244	invocation, and callback invocation will be synchronous. synchronous means	326	invocation, and callback invocation will be synchronous. Synchronous means
245	that it might take a while until the signal gets handled by the process,	327	that it might take a while until the signal gets handled by the process,
246	but it is guarenteed not to interrupt any other callbacks.	328	but it is guaranteed not to interrupt any other callbacks.
247		329
248	The main advantage of using these watchers is that you can share a signal	330	The main advantage of using these watchers is that you can share a signal
249	between multiple watchers.	331	between multiple watchers.
250		332
251	This watcher might use C<%SIG>, so programs overwriting those signals	333	This watcher might use C<%SIG>, so programs overwriting those signals
…		…
258	=head2 CHILD PROCESS WATCHERS	340	=head2 CHILD PROCESS WATCHERS
259		341
260	You can also watch on a child process exit and catch its exit status.	342	You can also watch on a child process exit and catch its exit status.
261		343
262	The child process is specified by the C<pid> argument (if set to C<0>, it	344	The child process is specified by the C<pid> argument (if set to C<0>, it
263	watches for any child process exit). The watcher will trigger as often	345	watches for any child process exit). The watcher will triggered only when
264	as status change for the child are received. This works by installing a	346	the child process has finished and an exit status is available, not on
265	signal handler for C<SIGCHLD>. The callback will be called with the pid	347	any trace events (stopped/continued).
266	and exit status (as returned by waitpid), so unlike other watcher types,	348
267	you I<can> rely on child watcher callback arguments.	349	The callback will be called with the pid and exit status (as returned by
		350	waitpid), so unlike other watcher types, you I<can> rely on child watcher
		351	callback arguments.
		352
		353	This watcher type works by installing a signal handler for C<SIGCHLD>,
		354	and since it cannot be shared, nothing else should use SIGCHLD or reap
		355	random child processes (waiting for specific child processes, e.g. inside
		356	C<system>, is just fine).
268		357
269	There is a slight catch to child watchers, however: you usually start them	358	There is a slight catch to child watchers, however: you usually start them
270	I<after> the child process was created, and this means the process could	359	I<after> the child process was created, and this means the process could
271	have exited already (and no SIGCHLD will be sent anymore).	360	have exited already (and no SIGCHLD will be sent anymore).
272		361
…		…
278	AnyEvent program, you I<have> to create at least one watcher before you	367	AnyEvent program, you I<have> to create at least one watcher before you
279	C<fork> the child (alternatively, you can call C<AnyEvent::detect>).	368	C<fork> the child (alternatively, you can call C<AnyEvent::detect>).
280		369
281	Example: fork a process and wait for it	370	Example: fork a process and wait for it
282		371
283	my $done = AnyEvent->condvar;	372	my $done = AnyEvent->condvar;
284		373
285	AnyEvent::detect; # force event module to be initialised
286
287	my $pid = fork or exit 5;	374	my $pid = fork or exit 5;
288		375
289	my $w = AnyEvent->child (	376	my $w = AnyEvent->child (
290	pid => $pid,	377	pid => $pid,
291	cb => sub {	378	cb => sub {
292	my ($pid, $status) = @_;	379	my ($pid, $status) = @_;
293	warn "pid $pid exited with status $status";	380	warn "pid $pid exited with status $status";
294	$done->broadcast;	381	$done->send;
295	},	382	},
296	);	383	);
297		384
298	# do something else, then wait for process exit	385	# do something else, then wait for process exit
299	$done->wait;	386	$done->recv;
300		387
301	=head2 CONDITION VARIABLES	388	=head2 CONDITION VARIABLES
302		389
		390	If you are familiar with some event loops you will know that all of them
		391	require you to run some blocking "loop", "run" or similar function that
		392	will actively watch for new events and call your callbacks.
		393
		394	AnyEvent is different, it expects somebody else to run the event loop and
		395	will only block when necessary (usually when told by the user).
		396
		397	The instrument to do that is called a "condition variable", so called
		398	because they represent a condition that must become true.
		399
303	Condition variables can be created by calling the C<< AnyEvent->condvar >>	400	Condition variables can be created by calling the C<< AnyEvent->condvar
304	method without any arguments.	401	>> method, usually without arguments. The only argument pair allowed is
305		402
306	A condition variable waits for a condition - precisely that the C<<	403	C<cb>, which specifies a callback to be called when the condition variable
307	->broadcast >> method has been called.	404	becomes true, with the condition variable as the first argument (but not
		405	the results).
308		406
309	They are very useful to signal that a condition has been fulfilled, for	407	After creation, the condition variable is "false" until it becomes "true"
		408	by calling the C<send> method (or calling the condition variable as if it
		409	were a callback, read about the caveats in the description for the C<<
		410	->send >> method).
		411
		412	Condition variables are similar to callbacks, except that you can
		413	optionally wait for them. They can also be called merge points - points
		414	in time where multiple outstanding events have been processed. And yet
		415	another way to call them is transactions - each condition variable can be
		416	used to represent a transaction, which finishes at some point and delivers
		417	a result.
		418
		419	Condition variables are very useful to signal that something has finished,
310	example, if you write a module that does asynchronous http requests,	420	for example, if you write a module that does asynchronous http requests,
311	then a condition variable would be the ideal candidate to signal the	421	then a condition variable would be the ideal candidate to signal the
312	availability of results.	422	availability of results. The user can either act when the callback is
		423	called or can synchronously C<< ->recv >> for the results.
313		424
314	You can also use condition variables to block your main program until	425	You can also use them to simulate traditional event loops - for example,
315	an event occurs - for example, you could C<< ->wait >> in your main	426	you can block your main program until an event occurs - for example, you
316	program until the user clicks the Quit button in your app, which would C<<	427	could C<< ->recv >> in your main program until the user clicks the Quit
317	->broadcast >> the "quit" event.	428	button of your app, which would C<< ->send >> the "quit" event.
318		429
319	Note that condition variables recurse into the event loop - if you have	430	Note that condition variables recurse into the event loop - if you have
320	two pirces of code that call C<< ->wait >> in a round-robbin fashion, you	431	two pieces of code that call C<< ->recv >> in a round-robin fashion, you
321	lose. Therefore, condition variables are good to export to your caller, but	432	lose. Therefore, condition variables are good to export to your caller, but
322	you should avoid making a blocking wait yourself, at least in callbacks,	433	you should avoid making a blocking wait yourself, at least in callbacks,
323	as this asks for trouble.	434	as this asks for trouble.
324		435
325	This object has two methods:	436	Condition variables are represented by hash refs in perl, and the keys
		437	used by AnyEvent itself are all named C<_ae_XXX> to make subclassing
		438	easy (it is often useful to build your own transaction class on top of
		439	AnyEvent). To subclass, use C<AnyEvent::CondVar> as base class and call
		440	it's C<new> method in your own C<new> method.
		441
		442	There are two "sides" to a condition variable - the "producer side" which
		443	eventually calls C<< -> send >>, and the "consumer side", which waits
		444	for the send to occur.
		445
		446	Example: wait for a timer.
		447
		448	# wait till the result is ready
		449	my $result_ready = AnyEvent->condvar;
		450
		451	# do something such as adding a timer
		452	# or socket watcher the calls $result_ready->send
		453	# when the "result" is ready.
		454	# in this case, we simply use a timer:
		455	my $w = AnyEvent->timer (
		456	after => 1,
		457	cb => sub { $result_ready->send },
		458	);
		459
		460	# this "blocks" (while handling events) till the callback
		461	# calls send
		462	$result_ready->recv;
		463
		464	Example: wait for a timer, but take advantage of the fact that
		465	condition variables are also code references.
		466
		467	my $done = AnyEvent->condvar;
		468	my $delay = AnyEvent->timer (after => 5, cb => $done);
		469	$done->recv;
		470
		471	Example: Imagine an API that returns a condvar and doesn't support
		472	callbacks. This is how you make a synchronous call, for example from
		473	the main program:
		474
		475	use AnyEvent::CouchDB;
		476
		477	...
		478
		479	my @info = $couchdb->info->recv;
		480
		481	And this is how you would just ste a callback to be called whenever the
		482	results are available:
		483
		484	$couchdb->info->cb (sub {
		485	my @info = $_[0]->recv;
		486	});
		487
		488	=head3 METHODS FOR PRODUCERS
		489
		490	These methods should only be used by the producing side, i.e. the
		491	code/module that eventually sends the signal. Note that it is also
		492	the producer side which creates the condvar in most cases, but it isn't
		493	uncommon for the consumer to create it as well.
326		494
327	=over 4	495	=over 4
328		496
		497	=item $cv->send (...)
		498
		499	Flag the condition as ready - a running C<< ->recv >> and all further
		500	calls to C<recv> will (eventually) return after this method has been
		501	called. If nobody is waiting the send will be remembered.
		502
		503	If a callback has been set on the condition variable, it is called
		504	immediately from within send.
		505
		506	Any arguments passed to the C<send> call will be returned by all
		507	future C<< ->recv >> calls.
		508
		509	Condition variables are overloaded so one can call them directly
		510	(as a code reference). Calling them directly is the same as calling
		511	C<send>. Note, however, that many C-based event loops do not handle
		512	overloading, so as tempting as it may be, passing a condition variable
		513	instead of a callback does not work. Both the pure perl and EV loops
		514	support overloading, however, as well as all functions that use perl to
		515	invoke a callback (as in L<AnyEvent::Socket> and L<AnyEvent::DNS> for
		516	example).
		517
		518	=item $cv->croak ($error)
		519
		520	Similar to send, but causes all call's to C<< ->recv >> to invoke
		521	C<Carp::croak> with the given error message/object/scalar.
		522
		523	This can be used to signal any errors to the condition variable
		524	user/consumer.
		525
		526	=item $cv->begin ([group callback])
		527
329	=item $cv->wait	528	=item $cv->end
330		529
331	Wait (blocking if necessary) until the C<< ->broadcast >> method has been	530	These two methods are EXPERIMENTAL and MIGHT CHANGE.
		531
		532	These two methods can be used to combine many transactions/events into
		533	one. For example, a function that pings many hosts in parallel might want
		534	to use a condition variable for the whole process.
		535
		536	Every call to C<< ->begin >> will increment a counter, and every call to
		537	C<< ->end >> will decrement it. If the counter reaches C<0> in C<< ->end
		538	>>, the (last) callback passed to C<begin> will be executed. That callback
		539	is I<supposed> to call C<< ->send >>, but that is not required. If no
		540	callback was set, C<send> will be called without any arguments.
		541
		542	Let's clarify this with the ping example:
		543
		544	my $cv = AnyEvent->condvar;
		545
		546	my %result;
		547	$cv->begin (sub { $cv->send (\%result) });
		548
		549	for my $host (@list_of_hosts) {
		550	$cv->begin;
		551	ping_host_then_call_callback $host, sub {
		552	$result{$host} = ...;
		553	$cv->end;
		554	};
		555	}
		556
		557	$cv->end;
		558
		559	This code fragment supposedly pings a number of hosts and calls
		560	C<send> after results for all then have have been gathered - in any
		561	order. To achieve this, the code issues a call to C<begin> when it starts
		562	each ping request and calls C<end> when it has received some result for
		563	it. Since C<begin> and C<end> only maintain a counter, the order in which
		564	results arrive is not relevant.
		565
		566	There is an additional bracketing call to C<begin> and C<end> outside the
		567	loop, which serves two important purposes: first, it sets the callback
		568	to be called once the counter reaches C<0>, and second, it ensures that
		569	C<send> is called even when C<no> hosts are being pinged (the loop
		570	doesn't execute once).
		571
		572	This is the general pattern when you "fan out" into multiple subrequests:
		573	use an outer C<begin>/C<end> pair to set the callback and ensure C<end>
		574	is called at least once, and then, for each subrequest you start, call
		575	C<begin> and for each subrequest you finish, call C<end>.
		576
		577	=back
		578
		579	=head3 METHODS FOR CONSUMERS
		580
		581	These methods should only be used by the consuming side, i.e. the
		582	code awaits the condition.
		583
		584	=over 4
		585
		586	=item $cv->recv
		587
		588	Wait (blocking if necessary) until the C<< ->send >> or C<< ->croak
332	called on c<$cv>, while servicing other watchers normally.	589	>> methods have been called on c<$cv>, while servicing other watchers
		590	normally.
333		591
334	You can only wait once on a condition - additional calls will return	592	You can only wait once on a condition - additional calls are valid but
335	immediately.	593	will return immediately.
		594
		595	If an error condition has been set by calling C<< ->croak >>, then this
		596	function will call C<croak>.
		597
		598	In list context, all parameters passed to C<send> will be returned,
		599	in scalar context only the first one will be returned.
336		600
337	Not all event models support a blocking wait - some die in that case	601	Not all event models support a blocking wait - some die in that case
338	(programs might want to do that to stay interactive), so I<if you are	602	(programs might want to do that to stay interactive), so I<if you are
339	using this from a module, never require a blocking wait>, but let the	603	using this from a module, never require a blocking wait>, but let the
340	caller decide whether the call will block or not (for example, by coupling	604	caller decide whether the call will block or not (for example, by coupling
341	condition variables with some kind of request results and supporting	605	condition variables with some kind of request results and supporting
342	callbacks so the caller knows that getting the result will not block,	606	callbacks so the caller knows that getting the result will not block,
343	while still suppporting blocking waits if the caller so desires).	607	while still supporting blocking waits if the caller so desires).
344		608
345	Another reason I<never> to C<< ->wait >> in a module is that you cannot	609	Another reason I<never> to C<< ->recv >> in a module is that you cannot
346	sensibly have two C<< ->wait >>'s in parallel, as that would require	610	sensibly have two C<< ->recv >>'s in parallel, as that would require
347	multiple interpreters or coroutines/threads, none of which C<AnyEvent>	611	multiple interpreters or coroutines/threads, none of which C<AnyEvent>
348	can supply (the coroutine-aware backends L<AnyEvent::Impl::CoroEV> and	612	can supply.
349	L<AnyEvent::Impl::CoroEvent> explicitly support concurrent C<< ->wait >>'s
350	from different coroutines, however).
351		613
352	=item $cv->broadcast	614	The L<Coro> module, however, I<can> and I<does> supply coroutines and, in
		615	fact, L<Coro::AnyEvent> replaces AnyEvent's condvars by coroutine-safe
		616	versions and also integrates coroutines into AnyEvent, making blocking
		617	C<< ->recv >> calls perfectly safe as long as they are done from another
		618	coroutine (one that doesn't run the event loop).
353		619
354	Flag the condition as ready - a running C<< ->wait >> and all further	620	You can ensure that C<< -recv >> never blocks by setting a callback and
355	calls to C<wait> will (eventually) return after this method has been	621	only calling C<< ->recv >> from within that callback (or at a later
356	called. If nobody is waiting the broadcast will be remembered..	622	time). This will work even when the event loop does not support blocking
		623	waits otherwise.
		624
		625	=item $bool = $cv->ready
		626
		627	Returns true when the condition is "true", i.e. whether C<send> or
		628	C<croak> have been called.
		629
		630	=item $cb = $cv->cb ($cb->($cv))
		631
		632	This is a mutator function that returns the callback set and optionally
		633	replaces it before doing so.
		634
		635	The callback will be called when the condition becomes "true", i.e. when
		636	C<send> or C<croak> are called, with the only argument being the condition
		637	variable itself. Calling C<recv> inside the callback or at any later time
		638	is guaranteed not to block.
357		639
358	=back	640	=back
359
360	Example:
361
362	# wait till the result is ready
363	my $result_ready = AnyEvent->condvar;
364
365	# do something such as adding a timer
366	# or socket watcher the calls $result_ready->broadcast
367	# when the "result" is ready.
368	# in this case, we simply use a timer:
369	my $w = AnyEvent->timer (
370	after => 1,
371	cb => sub { $result_ready->broadcast },
372	);
373
374	# this "blocks" (while handling events) till the watcher
375	# calls broadcast
376	$result_ready->wait;
377		641
378	=head1 GLOBAL VARIABLES AND FUNCTIONS	642	=head1 GLOBAL VARIABLES AND FUNCTIONS
379		643
380	=over 4	644	=over 4
381		645
…		…
387	C<AnyEvent::Impl:xxx> modules, but can be any other class in the case	651	C<AnyEvent::Impl:xxx> modules, but can be any other class in the case
388	AnyEvent has been extended at runtime (e.g. in I<rxvt-unicode>).	652	AnyEvent has been extended at runtime (e.g. in I<rxvt-unicode>).
389		653
390	The known classes so far are:	654	The known classes so far are:
391		655
392	AnyEvent::Impl::CoroEV based on Coro::EV, best choice.
393	AnyEvent::Impl::CoroEvent based on Coro::Event, second best choice.
394	AnyEvent::Impl::EV based on EV (an interface to libev, best choice).	656	AnyEvent::Impl::EV based on EV (an interface to libev, best choice).
395	AnyEvent::Impl::Event based on Event, second best choice.	657	AnyEvent::Impl::Event based on Event, second best choice.
		658	AnyEvent::Impl::Perl pure-perl implementation, fast and portable.
396	AnyEvent::Impl::Glib based on Glib, third-best choice.	659	AnyEvent::Impl::Glib based on Glib, third-best choice.
397	AnyEvent::Impl::Perl pure-perl implementation, inefficient but portable.
398	AnyEvent::Impl::Tk based on Tk, very bad choice.	660	AnyEvent::Impl::Tk based on Tk, very bad choice.
399	AnyEvent::Impl::Qt based on Qt, cannot be autoprobed (see its docs).	661	AnyEvent::Impl::Qt based on Qt, cannot be autoprobed (see its docs).
400	AnyEvent::Impl::EventLib based on Event::Lib, leaks memory and worse.	662	AnyEvent::Impl::EventLib based on Event::Lib, leaks memory and worse.
401	AnyEvent::Impl::POE based on POE, not generic enough for full support.	663	AnyEvent::Impl::POE based on POE, not generic enough for full support.
402		664
…		…
415	Returns C<$AnyEvent::MODEL>, forcing autodetection of the event model	677	Returns C<$AnyEvent::MODEL>, forcing autodetection of the event model
416	if necessary. You should only call this function right before you would	678	if necessary. You should only call this function right before you would
417	have created an AnyEvent watcher anyway, that is, as late as possible at	679	have created an AnyEvent watcher anyway, that is, as late as possible at
418	runtime.	680	runtime.
419		681
		682	=item $guard = AnyEvent::post_detect { BLOCK }
		683
		684	Arranges for the code block to be executed as soon as the event model is
		685	autodetected (or immediately if this has already happened).
		686
		687	If called in scalar or list context, then it creates and returns an object
		688	that automatically removes the callback again when it is destroyed. See
		689	L<Coro::BDB> for a case where this is useful.
		690
		691	=item @AnyEvent::post_detect
		692
		693	If there are any code references in this array (you can C<push> to it
		694	before or after loading AnyEvent), then they will called directly after
		695	the event loop has been chosen.
		696
		697	You should check C<$AnyEvent::MODEL> before adding to this array, though:
		698	if it contains a true value then the event loop has already been detected,
		699	and the array will be ignored.
		700
		701	Best use C<AnyEvent::post_detect { BLOCK }> instead.
		702
420	=back	703	=back
421		704
422	=head1 WHAT TO DO IN A MODULE	705	=head1 WHAT TO DO IN A MODULE
423		706
424	As a module author, you should C<use AnyEvent> and call AnyEvent methods	707	As a module author, you should C<use AnyEvent> and call AnyEvent methods
…		…
427	Be careful when you create watchers in the module body - AnyEvent will	710	Be careful when you create watchers in the module body - AnyEvent will
428	decide which event module to use as soon as the first method is called, so	711	decide which event module to use as soon as the first method is called, so
429	by calling AnyEvent in your module body you force the user of your module	712	by calling AnyEvent in your module body you force the user of your module
430	to load the event module first.	713	to load the event module first.
431		714
432	Never call C<< ->wait >> on a condition variable unless you I<know> that	715	Never call C<< ->recv >> on a condition variable unless you I<know> that
433	the C<< ->broadcast >> method has been called on it already. This is	716	the C<< ->send >> method has been called on it already. This is
434	because it will stall the whole program, and the whole point of using	717	because it will stall the whole program, and the whole point of using
435	events is to stay interactive.	718	events is to stay interactive.
436		719
437	It is fine, however, to call C<< ->wait >> when the user of your module	720	It is fine, however, to call C<< ->recv >> when the user of your module
438	requests it (i.e. if you create a http request object ad have a method	721	requests it (i.e. if you create a http request object ad have a method
439	called C<results> that returns the results, it should call C<< ->wait >>	722	called C<results> that returns the results, it should call C<< ->recv >>
440	freely, as the user of your module knows what she is doing. always).	723	freely, as the user of your module knows what she is doing. always).
441		724
442	=head1 WHAT TO DO IN THE MAIN PROGRAM	725	=head1 WHAT TO DO IN THE MAIN PROGRAM
443		726
444	There will always be a single main program - the only place that should	727	There will always be a single main program - the only place that should
…		…
446		729
447	If it doesn't care, it can just "use AnyEvent" and use it itself, or not	730	If it doesn't care, it can just "use AnyEvent" and use it itself, or not
448	do anything special (it does not need to be event-based) and let AnyEvent	731	do anything special (it does not need to be event-based) and let AnyEvent
449	decide which implementation to chose if some module relies on it.	732	decide which implementation to chose if some module relies on it.
450		733
451	If the main program relies on a specific event model. For example, in	734	If the main program relies on a specific event model - for example, in
452	Gtk2 programs you have to rely on the Glib module. You should load the	735	Gtk2 programs you have to rely on the Glib module - you should load the
453	event module before loading AnyEvent or any module that uses it: generally	736	event module before loading AnyEvent or any module that uses it: generally
454	speaking, you should load it as early as possible. The reason is that	737	speaking, you should load it as early as possible. The reason is that
455	modules might create watchers when they are loaded, and AnyEvent will	738	modules might create watchers when they are loaded, and AnyEvent will
456	decide on the event model to use as soon as it creates watchers, and it	739	decide on the event model to use as soon as it creates watchers, and it
457	might chose the wrong one unless you load the correct one yourself.	740	might chose the wrong one unless you load the correct one yourself.
458		741
459	You can chose to use a rather inefficient pure-perl implementation by	742	You can chose to use a pure-perl implementation by loading the
460	loading the C<AnyEvent::Impl::Perl> module, which gives you similar	743	C<AnyEvent::Impl::Perl> module, which gives you similar behaviour
461	behaviour everywhere, but letting AnyEvent chose is generally better.	744	everywhere, but letting AnyEvent chose the model is generally better.
		745
		746	=head2 MAINLOOP EMULATION
		747
		748	Sometimes (often for short test scripts, or even standalone programs who
		749	only want to use AnyEvent), you do not want to run a specific event loop.
		750
		751	In that case, you can use a condition variable like this:
		752
		753	AnyEvent->condvar->recv;
		754
		755	This has the effect of entering the event loop and looping forever.
		756
		757	Note that usually your program has some exit condition, in which case
		758	it is better to use the "traditional" approach of storing a condition
		759	variable somewhere, waiting for it, and sending it when the program should
		760	exit cleanly.
		761
		762
		763	=head1 OTHER MODULES
		764
		765	The following is a non-exhaustive list of additional modules that use
		766	AnyEvent and can therefore be mixed easily with other AnyEvent modules
		767	in the same program. Some of the modules come with AnyEvent, some are
		768	available via CPAN.
		769
		770	=over 4
		771
		772	=item L<AnyEvent::Util>
		773
		774	Contains various utility functions that replace often-used but blocking
		775	functions such as C<inet_aton> by event-/callback-based versions.
		776
		777	=item L<AnyEvent::Socket>
		778
		779	Provides various utility functions for (internet protocol) sockets,
		780	addresses and name resolution. Also functions to create non-blocking tcp
		781	connections or tcp servers, with IPv6 and SRV record support and more.
		782
		783	=item L<AnyEvent::Handle>
		784
		785	Provide read and write buffers, manages watchers for reads and writes,
		786	supports raw and formatted I/O, I/O queued and fully transparent and
		787	non-blocking SSL/TLS.
		788
		789	=item L<AnyEvent::DNS>
		790
		791	Provides rich asynchronous DNS resolver capabilities.
		792
		793	=item L<AnyEvent::HTTP>
		794
		795	A simple-to-use HTTP library that is capable of making a lot of concurrent
		796	HTTP requests.
		797
		798	=item L<AnyEvent::HTTPD>
		799
		800	Provides a simple web application server framework.
		801
		802	=item L<AnyEvent::FastPing>
		803
		804	The fastest ping in the west.
		805
		806	=item L<AnyEvent::DBI>
		807
		808	Executes L<DBI> requests asynchronously in a proxy process.
		809
		810	=item L<AnyEvent::AIO>
		811
		812	Truly asynchronous I/O, should be in the toolbox of every event
		813	programmer. AnyEvent::AIO transparently fuses L<IO::AIO> and AnyEvent
		814	together.
		815
		816	=item L<AnyEvent::BDB>
		817
		818	Truly asynchronous Berkeley DB access. AnyEvent::BDB transparently fuses
		819	L<BDB> and AnyEvent together.
		820
		821	=item L<AnyEvent::GPSD>
		822
		823	A non-blocking interface to gpsd, a daemon delivering GPS information.
		824
		825	=item L<AnyEvent::IGS>
		826
		827	A non-blocking interface to the Internet Go Server protocol (used by
		828	L<App::IGS>).
		829
		830	=item L<Net::IRC3>
		831
		832	AnyEvent based IRC client module family.
		833
		834	=item L<Net::XMPP2>
		835
		836	AnyEvent based XMPP (Jabber protocol) module family.
		837
		838	=item L<Net::FCP>
		839
		840	AnyEvent-based implementation of the Freenet Client Protocol, birthplace
		841	of AnyEvent.
		842
		843	=item L<Event::ExecFlow>
		844
		845	High level API for event-based execution flow control.
		846
		847	=item L<Coro>
		848
		849	Has special support for AnyEvent via L<Coro::AnyEvent>.
		850
		851	=item L<IO::Lambda>
		852
		853	The lambda approach to I/O - don't ask, look there. Can use AnyEvent.
		854
		855	=back
462		856
463	=cut	857	=cut
464		858
465	package AnyEvent;	859	package AnyEvent;
466		860
467	no warnings;	861	no warnings;
468	use strict;	862	use strict qw(vars subs);
469		863
470	use Carp;	864	use Carp;
471		865
472	our $VERSION = '3.3';	866	our $VERSION = 4.234;
473	our $MODEL;	867	our $MODEL;
474		868
475	our $AUTOLOAD;	869	our $AUTOLOAD;
476	our @ISA;	870	our @ISA;
477		871
		872	our @REGISTRY;
		873
		874	our $WIN32;
		875
		876	BEGIN {
		877	my $win32 = ! ! ($^O =~ /mswin32/i);
		878	eval "sub WIN32(){ $win32 }";
		879	}
		880
478	our $verbose = $ENV{PERL_ANYEVENT_VERBOSE}*1;	881	our $verbose = $ENV{PERL_ANYEVENT_VERBOSE}*1;
479		882
480	our @REGISTRY;	883	our %PROTOCOL; # (ipv4|ipv6) => (1|2), higher numbers are preferred
		884
		885	{
		886	my $idx;
		887	$PROTOCOL{$_} = ++$idx
		888	for reverse split /\s*,\s*/,
		889	$ENV{PERL_ANYEVENT_PROTOCOLS} || "ipv4,ipv6";
		890	}
481		891
482	my @models = (	892	my @models = (
483	[Coro::EV:: => AnyEvent::Impl::CoroEV::],
484	[Coro::Event:: => AnyEvent::Impl::CoroEvent::],
485	[EV:: => AnyEvent::Impl::EV::],	893	[EV:: => AnyEvent::Impl::EV::],
486	[Event:: => AnyEvent::Impl::Event::],	894	[Event:: => AnyEvent::Impl::Event::],
487	[Glib:: => AnyEvent::Impl::Glib::],
488	[Tk:: => AnyEvent::Impl::Tk::],
489	[Wx:: => AnyEvent::Impl::POE::],
490	[Prima:: => AnyEvent::Impl::POE::],
491	[AnyEvent::Impl::Perl:: => AnyEvent::Impl::Perl::],	895	[AnyEvent::Impl::Perl:: => AnyEvent::Impl::Perl::],
492	# everything below here will not be autoprobed as the pureperl backend should work everywhere	896	# everything below here will not be autoprobed
		897	# as the pureperl backend should work everywhere
		898	# and is usually faster
		899	[Tk:: => AnyEvent::Impl::Tk::], # crashes with many handles
		900	[Glib:: => AnyEvent::Impl::Glib::], # becomes extremely slow with many watchers
493	[Event::Lib:: => AnyEvent::Impl::EventLib::], # too buggy	901	[Event::Lib:: => AnyEvent::Impl::EventLib::], # too buggy
494	[Qt:: => AnyEvent::Impl::Qt::], # requires special main program	902	[Qt:: => AnyEvent::Impl::Qt::], # requires special main program
495	[POE::Kernel:: => AnyEvent::Impl::POE::], # lasciate ogni speranza	903	[POE::Kernel:: => AnyEvent::Impl::POE::], # lasciate ogni speranza
		904	[Wx:: => AnyEvent::Impl::POE::],
		905	[Prima:: => AnyEvent::Impl::POE::],
496	);	906	);
497		907
498	our %method = map +($_ => 1), qw(io timer signal child condvar broadcast wait one_event DESTROY);	908	our %method = map +($_ => 1), qw(io timer time now signal child condvar one_event DESTROY);
		909
		910	our @post_detect;
		911
		912	sub post_detect(&) {
		913	my ($cb) = @_;
		914
		915	if ($MODEL) {
		916	$cb->();
		917
		918	1
		919	} else {
		920	push @post_detect, $cb;
		921
		922	defined wantarray
		923	? bless \$cb, "AnyEvent::Util::PostDetect"
		924	: ()
		925	}
		926	}
		927
		928	sub AnyEvent::Util::PostDetect::DESTROY {
		929	@post_detect = grep $_ != ${$_[0]}, @post_detect;
		930	}
499		931
500	sub detect() {	932	sub detect() {
501	unless ($MODEL) {	933	unless ($MODEL) {
502	no strict 'refs';	934	no strict 'refs';
		935	local $SIG{__DIE__};
503		936
504	if ($ENV{PERL_ANYEVENT_MODEL} =~ /^([a-zA-Z]+)$/) {	937	if ($ENV{PERL_ANYEVENT_MODEL} =~ /^([a-zA-Z]+)$/) {
505	my $model = "AnyEvent::Impl::$1";	938	my $model = "AnyEvent::Impl::$1";
506	if (eval "require $model") {	939	if (eval "require $model") {
507	$MODEL = $model;	940	$MODEL = $model;
…		…
537	last;	970	last;
538	}	971	}
539	}	972	}
540		973
541	$MODEL	974	$MODEL
542	or die "No event module selected for AnyEvent and autodetect failed. Install any one of these modules: EV (or Coro+EV), Event (or Coro+Event) or Glib.";	975	or die "No event module selected for AnyEvent and autodetect failed. Install any one of these modules: EV, Event or Glib.";
543	}	976	}
544	}	977	}
545		978
		979	push @{"$MODEL\::ISA"}, "AnyEvent::Base";
		980
546	unshift @ISA, $MODEL;	981	unshift @ISA, $MODEL;
547	push @{"$MODEL\::ISA"}, "AnyEvent::Base";	982
		983	require AnyEvent::Strict if $ENV{PERL_ANYEVENT_STRICT};
		984
		985	(shift @post_detect)->() while @post_detect;
548	}	986	}
549		987
550	$MODEL	988	$MODEL
551	}	989	}
552		990
…		…
560		998
561	my $class = shift;	999	my $class = shift;
562	$class->$func (@_);	1000	$class->$func (@_);
563	}	1001	}
564		1002
		1003	# utility function to dup a filehandle. this is used by many backends
		1004	# to support binding more than one watcher per filehandle (they usually
		1005	# allow only one watcher per fd, so we dup it to get a different one).
		1006	sub _dupfh($$$$) {
		1007	my ($poll, $fh, $r, $w) = @_;
		1008
		1009	require Fcntl;
		1010
		1011	# cygwin requires the fh mode to be matching, unix doesn't
		1012	my ($rw, $mode) = $poll eq "r" ? ($r, "<")
		1013	: $poll eq "w" ? ($w, ">")
		1014	: Carp::croak "AnyEvent->io requires poll set to either 'r' or 'w'";
		1015
		1016	open my $fh2, "$mode&" . fileno $fh
		1017	or die "cannot dup() filehandle: $!";
		1018
		1019	# we assume CLOEXEC is already set by perl in all important cases
		1020
		1021	($fh2, $rw)
		1022	}
		1023
565	package AnyEvent::Base;	1024	package AnyEvent::Base;
566		1025
		1026	# default implementation for now and time
		1027
		1028	BEGIN {
		1029	if (eval "use Time::HiRes (); time (); 1") {
		1030	*_time = \&Time::HiRes::time;
		1031	# if (eval "use POSIX (); (POSIX::times())...
		1032	} else {
		1033	*_time = \&CORE::time; # epic fail
		1034	}
		1035	}
		1036
		1037	sub time { _time }
		1038	sub now { _time }
		1039
567	# default implementation for ->condvar, ->wait, ->broadcast	1040	# default implementation for ->condvar
568		1041
569	sub condvar {	1042	sub condvar {
570	bless \my $flag, "AnyEvent::Base::CondVar"	1043	bless { @_ == 3 ? (_ae_cb => $_[2]) : () }, AnyEvent::CondVar::
571	}
572
573	sub AnyEvent::Base::CondVar::broadcast {
574	${$_[0]}++;
575	}
576
577	sub AnyEvent::Base::CondVar::wait {
578	AnyEvent->one_event while !${$_[0]};
579	}	1044	}
580		1045
581	# default implementation for ->signal	1046	# default implementation for ->signal
582		1047
583	our %SIG_CB;	1048	our %SIG_CB;
…		…
599	sub AnyEvent::Base::Signal::DESTROY {	1064	sub AnyEvent::Base::Signal::DESTROY {
600	my ($signal, $cb) = @{$_[0]};	1065	my ($signal, $cb) = @{$_[0]};
601		1066
602	delete $SIG_CB{$signal}{$cb};	1067	delete $SIG_CB{$signal}{$cb};
603		1068
604	$SIG{$signal} = 'DEFAULT' unless keys %{ $SIG_CB{$signal} };	1069	delete $SIG{$signal} unless keys %{ $SIG_CB{$signal} };
605	}	1070	}
606		1071
607	# default implementation for ->child	1072	# default implementation for ->child
608		1073
609	our %PID_CB;	1074	our %PID_CB;
…		…
636	or Carp::croak "required option 'pid' is missing";	1101	or Carp::croak "required option 'pid' is missing";
637		1102
638	$PID_CB{$pid}{$arg{cb}} = $arg{cb};	1103	$PID_CB{$pid}{$arg{cb}} = $arg{cb};
639		1104
640	unless ($WNOHANG) {	1105	unless ($WNOHANG) {
641	$WNOHANG = eval { require POSIX; &POSIX::WNOHANG } || 1;	1106	$WNOHANG = eval { local $SIG{__DIE__}; require POSIX; &POSIX::WNOHANG } || 1;
642	}	1107	}
643		1108
644	unless ($CHLD_W) {	1109	unless ($CHLD_W) {
645	$CHLD_W = AnyEvent->signal (signal => 'CHLD', cb => \&_sigchld);	1110	$CHLD_W = AnyEvent->signal (signal => 'CHLD', cb => \&_sigchld);
646	# child could be a zombie already, so make at least one round	1111	# child could be a zombie already, so make at least one round
…		…
656	delete $PID_CB{$pid}{$cb};	1121	delete $PID_CB{$pid}{$cb};
657	delete $PID_CB{$pid} unless keys %{ $PID_CB{$pid} };	1122	delete $PID_CB{$pid} unless keys %{ $PID_CB{$pid} };
658		1123
659	undef $CHLD_W unless keys %PID_CB;	1124	undef $CHLD_W unless keys %PID_CB;
660	}	1125	}
		1126
		1127	package AnyEvent::CondVar;
		1128
		1129	our @ISA = AnyEvent::CondVar::Base::;
		1130
		1131	package AnyEvent::CondVar::Base;
		1132
		1133	use overload
		1134	'&{}' => sub { my $self = shift; sub { $self->send (@_) } },
		1135	fallback => 1;
		1136
		1137	sub _send {
		1138	# nop
		1139	}
		1140
		1141	sub send {
		1142	my $cv = shift;
		1143	$cv->{_ae_sent} = [@_];
		1144	(delete $cv->{_ae_cb})->($cv) if $cv->{_ae_cb};
		1145	$cv->_send;
		1146	}
		1147
		1148	sub croak {
		1149	$_[0]{_ae_croak} = $_[1];
		1150	$_[0]->send;
		1151	}
		1152
		1153	sub ready {
		1154	$_[0]{_ae_sent}
		1155	}
		1156
		1157	sub _wait {
		1158	AnyEvent->one_event while !$_[0]{_ae_sent};
		1159	}
		1160
		1161	sub recv {
		1162	$_[0]->_wait;
		1163
		1164	Carp::croak $_[0]{_ae_croak} if $_[0]{_ae_croak};
		1165	wantarray ? @{ $_[0]{_ae_sent} } : $_[0]{_ae_sent}[0]
		1166	}
		1167
		1168	sub cb {
		1169	$_[0]{_ae_cb} = $_[1] if @_ > 1;
		1170	$_[0]{_ae_cb}
		1171	}
		1172
		1173	sub begin {
		1174	++$_[0]{_ae_counter};
		1175	$_[0]{_ae_end_cb} = $_[1] if @_ > 1;
		1176	}
		1177
		1178	sub end {
		1179	return if --$_[0]{_ae_counter};
		1180	&{ $_[0]{_ae_end_cb} || sub { $_[0]->send } };
		1181	}
		1182
		1183	# undocumented/compatibility with pre-3.4
		1184	*broadcast = \&send;
		1185	*wait = \&_wait;
		1186
		1187	=head1 ERROR AND EXCEPTION HANDLING
		1188
		1189	In general, AnyEvent does not do any error handling - it relies on the
		1190	caller to do that if required. The L<AnyEvent::Strict> module (see also
		1191	the C<PERL_ANYEVENT_STRICT> environment variable, below) provides strict
		1192	checking of all AnyEvent methods, however, which is highly useful during
		1193	development.
		1194
		1195	As for exception handling (i.e. runtime errors and exceptions thrown while
		1196	executing a callback), this is not only highly event-loop specific, but
		1197	also not in any way wrapped by this module, as this is the job of the main
		1198	program.
		1199
		1200	The pure perl event loop simply re-throws the exception (usually
		1201	within C<< condvar->recv >>), the L<Event> and L<EV> modules call C<<
		1202	$Event/EV::DIED->() >>, L<Glib> uses C<< install_exception_handler >> and
		1203	so on.
		1204
		1205	=head1 ENVIRONMENT VARIABLES
		1206
		1207	The following environment variables are used by this module or its
		1208	submodules:
		1209
		1210	=over 4
		1211
		1212	=item C<PERL_ANYEVENT_VERBOSE>
		1213
		1214	By default, AnyEvent will be completely silent except in fatal
		1215	conditions. You can set this environment variable to make AnyEvent more
		1216	talkative.
		1217
		1218	When set to C<1> or higher, causes AnyEvent to warn about unexpected
		1219	conditions, such as not being able to load the event model specified by
		1220	C<PERL_ANYEVENT_MODEL>.
		1221
		1222	When set to C<2> or higher, cause AnyEvent to report to STDERR which event
		1223	model it chooses.
		1224
		1225	=item C<PERL_ANYEVENT_STRICT>
		1226
		1227	AnyEvent does not do much argument checking by default, as thorough
		1228	argument checking is very costly. Setting this variable to a true value
		1229	will cause AnyEvent to load C<AnyEvent::Strict> and then to thoroughly
		1230	check the arguments passed to most method calls. If it finds any problems
		1231	it will croak.
		1232
		1233	In other words, enables "strict" mode.
		1234
		1235	Unlike C<use strict>, it is definitely recommended ot keep it off in
		1236	production. Keeping C<PERL_ANYEVENT_STRICT=1> in your environment while
		1237	developing programs can be very useful, however.
		1238
		1239	=item C<PERL_ANYEVENT_MODEL>
		1240
		1241	This can be used to specify the event model to be used by AnyEvent, before
		1242	auto detection and -probing kicks in. It must be a string consisting
		1243	entirely of ASCII letters. The string C<AnyEvent::Impl::> gets prepended
		1244	and the resulting module name is loaded and if the load was successful,
		1245	used as event model. If it fails to load AnyEvent will proceed with
		1246	auto detection and -probing.
		1247
		1248	This functionality might change in future versions.
		1249
		1250	For example, to force the pure perl model (L<AnyEvent::Impl::Perl>) you
		1251	could start your program like this:
		1252
		1253	PERL_ANYEVENT_MODEL=Perl perl ...
		1254
		1255	=item C<PERL_ANYEVENT_PROTOCOLS>
		1256
		1257	Used by both L<AnyEvent::DNS> and L<AnyEvent::Socket> to determine preferences
		1258	for IPv4 or IPv6. The default is unspecified (and might change, or be the result
		1259	of auto probing).
		1260
		1261	Must be set to a comma-separated list of protocols or address families,
		1262	current supported: C<ipv4> and C<ipv6>. Only protocols mentioned will be
		1263	used, and preference will be given to protocols mentioned earlier in the
		1264	list.
		1265
		1266	This variable can effectively be used for denial-of-service attacks
		1267	against local programs (e.g. when setuid), although the impact is likely
		1268	small, as the program has to handle connection errors already-
		1269
		1270	Examples: C<PERL_ANYEVENT_PROTOCOLS=ipv4,ipv6> - prefer IPv4 over IPv6,
		1271	but support both and try to use both. C<PERL_ANYEVENT_PROTOCOLS=ipv4>
		1272	- only support IPv4, never try to resolve or contact IPv6
		1273	addresses. C<PERL_ANYEVENT_PROTOCOLS=ipv6,ipv4> support either IPv4 or
		1274	IPv6, but prefer IPv6 over IPv4.
		1275
		1276	=item C<PERL_ANYEVENT_EDNS0>
		1277
		1278	Used by L<AnyEvent::DNS> to decide whether to use the EDNS0 extension
		1279	for DNS. This extension is generally useful to reduce DNS traffic, but
		1280	some (broken) firewalls drop such DNS packets, which is why it is off by
		1281	default.
		1282
		1283	Setting this variable to C<1> will cause L<AnyEvent::DNS> to announce
		1284	EDNS0 in its DNS requests.
		1285
		1286	=item C<PERL_ANYEVENT_MAX_FORKS>
		1287
		1288	The maximum number of child processes that C<AnyEvent::Util::fork_call>
		1289	will create in parallel.
		1290
		1291	=back
661		1292
662	=head1 SUPPLYING YOUR OWN EVENT MODEL INTERFACE	1293	=head1 SUPPLYING YOUR OWN EVENT MODEL INTERFACE
663		1294
664	This is an advanced topic that you do not normally need to use AnyEvent in	1295	This is an advanced topic that you do not normally need to use AnyEvent in
665	a module. This section is only of use to event loop authors who want to	1296	a module. This section is only of use to event loop authors who want to
…		…
699		1330
700	I<rxvt-unicode> also cheats a bit by not providing blocking access to	1331	I<rxvt-unicode> also cheats a bit by not providing blocking access to
701	condition variables: code blocking while waiting for a condition will	1332	condition variables: code blocking while waiting for a condition will
702	C<die>. This still works with most modules/usages, and blocking calls must	1333	C<die>. This still works with most modules/usages, and blocking calls must
703	not be done in an interactive application, so it makes sense.	1334	not be done in an interactive application, so it makes sense.
704
705	=head1 ENVIRONMENT VARIABLES
706
707	The following environment variables are used by this module:
708
709	=over 4
710
711	=item C<PERL_ANYEVENT_VERBOSE>
712
713	By default, AnyEvent will be completely silent except in fatal
714	conditions. You can set this environment variable to make AnyEvent more
715	talkative.
716
717	When set to C<1> or higher, causes AnyEvent to warn about unexpected
718	conditions, such as not being able to load the event model specified by
719	C<PERL_ANYEVENT_MODEL>.
720
721	When set to C<2> or higher, cause AnyEvent to report to STDERR which event
722	model it chooses.
723
724	=item C<PERL_ANYEVENT_MODEL>
725
726	This can be used to specify the event model to be used by AnyEvent, before
727	autodetection and -probing kicks in. It must be a string consisting
728	entirely of ASCII letters. The string C<AnyEvent::Impl::> gets prepended
729	and the resulting module name is loaded and if the load was successful,
730	used as event model. If it fails to load AnyEvent will proceed with
731	autodetection and -probing.
732
733	This functionality might change in future versions.
734
735	For example, to force the pure perl model (L<AnyEvent::Impl::Perl>) you
736	could start your program like this:
737
738	PERL_ANYEVENT_MODEL=Perl perl ...
739
740	=back
741		1335
742	=head1 EXAMPLE PROGRAM	1336	=head1 EXAMPLE PROGRAM
743		1337
744	The following program uses an I/O watcher to read data from STDIN, a timer	1338	The following program uses an I/O watcher to read data from STDIN, a timer
745	to display a message once per second, and a condition variable to quit the	1339	to display a message once per second, and a condition variable to quit the
…		…
754	poll => 'r',	1348	poll => 'r',
755	cb => sub {	1349	cb => sub {
756	warn "io event <$_[0]>\n"; # will always output <r>	1350	warn "io event <$_[0]>\n"; # will always output <r>
757	chomp (my $input = <STDIN>); # read a line	1351	chomp (my $input = <STDIN>); # read a line
758	warn "read: $input\n"; # output what has been read	1352	warn "read: $input\n"; # output what has been read
759	$cv->broadcast if $input =~ /^q/i; # quit program if /^q/i	1353	$cv->send if $input =~ /^q/i; # quit program if /^q/i
760	},	1354	},
761	);	1355	);
762		1356
763	my $time_watcher; # can only be used once	1357	my $time_watcher; # can only be used once
764		1358
…		…
769	});	1363	});
770	}	1364	}
771		1365
772	new_timer; # create first timer	1366	new_timer; # create first timer
773		1367
774	$cv->wait; # wait until user enters /^q/i	1368	$cv->recv; # wait until user enters /^q/i
775		1369
776	=head1 REAL-WORLD EXAMPLE	1370	=head1 REAL-WORLD EXAMPLE
777		1371
778	Consider the L<Net::FCP> module. It features (among others) the following	1372	Consider the L<Net::FCP> module. It features (among others) the following
779	API calls, which are to freenet what HTTP GET requests are to http:	1373	API calls, which are to freenet what HTTP GET requests are to http:
…		…
829	syswrite $txn->{fh}, $txn->{request}	1423	syswrite $txn->{fh}, $txn->{request}
830	or die "connection or write error";	1424	or die "connection or write error";
831	$txn->{w} = AnyEvent->io (fh => $txn->{fh}, poll => 'r', cb => sub { $txn->fh_ready_r });	1425	$txn->{w} = AnyEvent->io (fh => $txn->{fh}, poll => 'r', cb => sub { $txn->fh_ready_r });
832		1426
833	Again, C<fh_ready_r> waits till all data has arrived, and then stores the	1427	Again, C<fh_ready_r> waits till all data has arrived, and then stores the
834	result and signals any possible waiters that the request ahs finished:	1428	result and signals any possible waiters that the request has finished:
835		1429
836	sysread $txn->{fh}, $txn->{buf}, length $txn->{$buf};	1430	sysread $txn->{fh}, $txn->{buf}, length $txn->{$buf};
837		1431
838	if (end-of-file or data complete) {	1432	if (end-of-file or data complete) {
839	$txn->{result} = $txn->{buf};	1433	$txn->{result} = $txn->{buf};
840	$txn->{finished}->broadcast;	1434	$txn->{finished}->send;
841	$txb->{cb}->($txn) of $txn->{cb}; # also call callback	1435	$txb->{cb}->($txn) of $txn->{cb}; # also call callback
842	}	1436	}
843		1437
844	The C<result> method, finally, just waits for the finished signal (if the	1438	The C<result> method, finally, just waits for the finished signal (if the
845	request was already finished, it doesn't wait, of course, and returns the	1439	request was already finished, it doesn't wait, of course, and returns the
846	data:	1440	data:
847		1441
848	$txn->{finished}->wait;	1442	$txn->{finished}->recv;
849	return $txn->{result};	1443	return $txn->{result};
850		1444
851	The actual code goes further and collects all errors (C<die>s, exceptions)	1445	The actual code goes further and collects all errors (C<die>s, exceptions)
852	that occured during request processing. The C<result> method detects	1446	that occurred during request processing. The C<result> method detects
853	whether an exception as thrown (it is stored inside the $txn object)	1447	whether an exception as thrown (it is stored inside the $txn object)
854	and just throws the exception, which means connection errors and other	1448	and just throws the exception, which means connection errors and other
855	problems get reported tot he code that tries to use the result, not in a	1449	problems get reported tot he code that tries to use the result, not in a
856	random callback.	1450	random callback.
857		1451
…		…
888		1482
889	my $quit = AnyEvent->condvar;	1483	my $quit = AnyEvent->condvar;
890		1484
891	$fcp->txn_client_get ($url)->cb (sub {	1485	$fcp->txn_client_get ($url)->cb (sub {
892	...	1486	...
893	$quit->broadcast;	1487	$quit->send;
894	});	1488	});
895		1489
896	$quit->wait;	1490	$quit->recv;
897		1491
898		1492
899	=head1 BENCHMARK	1493	=head1 BENCHMARKS
900		1494
901	To give you an idea of the performance and overheads that AnyEvent adds	1495	To give you an idea of the performance and overheads that AnyEvent adds
902	over the event loops themselves (and to give you an impression of the	1496	over the event loops themselves and to give you an impression of the speed
903	speed of various event loops), here is a benchmark of various supported	1497	of various event loops I prepared some benchmarks.
904	event models natively and with anyevent. The benchmark creates a lot of	1498
905	timers (with a zero timeout) and I/O watchers (watching STDOUT, a pty, to	1499	=head2 BENCHMARKING ANYEVENT OVERHEAD
		1500
		1501	Here is a benchmark of various supported event models used natively and
		1502	through AnyEvent. The benchmark creates a lot of timers (with a zero
		1503	timeout) and I/O watchers (watching STDOUT, a pty, to become writable,
906	become writable, which it is), lets them fire exactly once and destroys	1504	which it is), lets them fire exactly once and destroys them again.
907	them again.
908		1505
909	Rewriting the benchmark to use many different sockets instead of using	1506	Source code for this benchmark is found as F<eg/bench> in the AnyEvent
910	the same filehandle for all I/O watchers results in a much longer runtime	1507	distribution.
911	(socket creation is expensive), but qualitatively the same figures, so it
912	was not used.
913		1508
914	=head2 Explanation of the columns	1509	=head3 Explanation of the columns
915		1510
916	I<watcher> is the number of event watchers created/destroyed. Since	1511	I<watcher> is the number of event watchers created/destroyed. Since
917	different event models feature vastly different performances, each event	1512	different event models feature vastly different performances, each event
918	loop was given a number of watchers so that overall runtime is acceptable	1513	loop was given a number of watchers so that overall runtime is acceptable
919	and similar between tested event loop (and keep them from crashing): Glib	1514	and similar between tested event loop (and keep them from crashing): Glib
…		…
929	all watchers, to avoid adding memory overhead. That means closure creation	1524	all watchers, to avoid adding memory overhead. That means closure creation
930	and memory usage is not included in the figures.	1525	and memory usage is not included in the figures.
931		1526
932	I<invoke> is the time, in microseconds, used to invoke a simple	1527	I<invoke> is the time, in microseconds, used to invoke a simple
933	callback. The callback simply counts down a Perl variable and after it was	1528	callback. The callback simply counts down a Perl variable and after it was
934	invoked "watcher" times, it would C<< ->broadcast >> a condvar once to	1529	invoked "watcher" times, it would C<< ->send >> a condvar once to
935	signal the end of this phase.	1530	signal the end of this phase.
936		1531
937	I<destroy> is the time, in microseconds, that it takes to destroy a single	1532	I<destroy> is the time, in microseconds, that it takes to destroy a single
938	watcher.	1533	watcher.
939		1534
940	=head2 Results	1535	=head3 Results
941		1536
942	name watchers bytes create invoke destroy comment	1537	name watchers bytes create invoke destroy comment
943	EV/EV 400000 244 0.56 0.46 0.31 EV native interface	1538	EV/EV 400000 244 0.56 0.46 0.31 EV native interface
944	EV/Any 100000 244 2.50 0.46 0.29 EV + AnyEvent watchers	1539	EV/Any 100000 244 2.50 0.46 0.29 EV + AnyEvent watchers
945	CoroEV/Any 100000 244 2.49 0.44 0.29 coroutines + Coro::Signal	1540	CoroEV/Any 100000 244 2.49 0.44 0.29 coroutines + Coro::Signal
946	Perl/Any 100000 513 4.92 0.87 1.12 pure perl implementation	1541	Perl/Any 100000 513 4.92 0.87 1.12 pure perl implementation
947	Event/Event 16000 516 31.88 31.30 0.85 Event native interface	1542	Event/Event 16000 516 31.88 31.30 0.85 Event native interface
948	Event/Any 16000 936 39.17 33.63 1.43 Event + AnyEvent watchers	1543	Event/Any 16000 590 35.75 31.42 1.08 Event + AnyEvent watchers
949	Glib/Any 16000 1357 98.22 12.41 54.00 quadratic behaviour	1544	Glib/Any 16000 1357 98.22 12.41 54.00 quadratic behaviour
950	Tk/Any 2000 1860 26.97 67.98 14.00 SEGV with >> 2000 watchers	1545	Tk/Any 2000 1860 26.97 67.98 14.00 SEGV with >> 2000 watchers
951	POE/Event 2000 6644 108.64 736.02 14.73 via POE::Loop::Event	1546	POE/Event 2000 6644 108.64 736.02 14.73 via POE::Loop::Event
952	POE/Select 2000 6343 94.13 809.12 565.96 via POE::Loop::Select	1547	POE/Select 2000 6343 94.13 809.12 565.96 via POE::Loop::Select
953		1548
954	=head2 Discussion	1549	=head3 Discussion
955		1550
956	The benchmark does I<not> measure scalability of the event loop very	1551	The benchmark does I<not> measure scalability of the event loop very
957	well. For example, a select-based event loop (such as the pure perl one)	1552	well. For example, a select-based event loop (such as the pure perl one)
958	can never compete with an event loop that uses epoll when the number of	1553	can never compete with an event loop that uses epoll when the number of
959	file descriptors grows high. In this benchmark, all events become ready at	1554	file descriptors grows high. In this benchmark, all events become ready at
960	the same time, so select/poll-based implementations get an unnatural speed	1555	the same time, so select/poll-based implementations get an unnatural speed
961	boost.	1556	boost.
962		1557
		1558	Also, note that the number of watchers usually has a nonlinear effect on
		1559	overall speed, that is, creating twice as many watchers doesn't take twice
		1560	the time - usually it takes longer. This puts event loops tested with a
		1561	higher number of watchers at a disadvantage.
		1562
		1563	To put the range of results into perspective, consider that on the
		1564	benchmark machine, handling an event takes roughly 1600 CPU cycles with
		1565	EV, 3100 CPU cycles with AnyEvent's pure perl loop and almost 3000000 CPU
		1566	cycles with POE.
		1567
963	C<EV> is the sole leader regarding speed and memory use, which are both	1568	C<EV> is the sole leader regarding speed and memory use, which are both
964	maximal/minimal, respectively. Even when going through AnyEvent, it uses	1569	maximal/minimal, respectively. Even when going through AnyEvent, it uses
965	far less memory than any other event loop and is still faster than Event	1570	far less memory than any other event loop and is still faster than Event
966	natively.	1571	natively.
967		1572
…		…
970	interpreter and the backend itself). Nevertheless this shows that it	1575	interpreter and the backend itself). Nevertheless this shows that it
971	adds very little overhead in itself. Like any select-based backend its	1576	adds very little overhead in itself. Like any select-based backend its
972	performance becomes really bad with lots of file descriptors (and few of	1577	performance becomes really bad with lots of file descriptors (and few of
973	them active), of course, but this was not subject of this benchmark.	1578	them active), of course, but this was not subject of this benchmark.
974		1579
975	The C<Event> module has a relatively high setup and callback invocation cost,	1580	The C<Event> module has a relatively high setup and callback invocation
976	but overall scores on the third place.	1581	cost, but overall scores in on the third place.
977		1582
978	C<Glib>'s memory usage is quite a bit bit higher, but it features a	1583	C<Glib>'s memory usage is quite a bit higher, but it features a
979	faster callback invocation and overall ends up in the same class as	1584	faster callback invocation and overall ends up in the same class as
980	C<Event>. However, Glib scales extremely badly, doubling the number of	1585	C<Event>. However, Glib scales extremely badly, doubling the number of
981	watchers increases the processing time by more than a factor of four,	1586	watchers increases the processing time by more than a factor of four,
982	making it completely unusable when using larger numbers of watchers	1587	making it completely unusable when using larger numbers of watchers
983	(note that only a single file descriptor was used in the benchmark, so	1588	(note that only a single file descriptor was used in the benchmark, so
…		…
989	file descriptor is dup()ed for each watcher. This shows that the dup()	1594	file descriptor is dup()ed for each watcher. This shows that the dup()
990	employed by some adaptors is not a big performance issue (it does incur a	1595	employed by some adaptors is not a big performance issue (it does incur a
991	hidden memory cost inside the kernel which is not reflected in the figures	1596	hidden memory cost inside the kernel which is not reflected in the figures
992	above).	1597	above).
993		1598
994	C<POE>, regardless of underlying event loop (whether using its pure	1599	C<POE>, regardless of underlying event loop (whether using its pure perl
995	perl select-based backend or the Event module, the POE-EV backend	1600	select-based backend or the Event module, the POE-EV backend couldn't
996	couldn't be tested because it wasn't working) shows abysmal performance	1601	be tested because it wasn't working) shows abysmal performance and
997	and memory usage: Watchers use almost 30 times as much memory as	1602	memory usage with AnyEvent: Watchers use almost 30 times as much memory
998	EV watchers, and 10 times as much memory as Event (the high memory	1603	as EV watchers, and 10 times as much memory as Event (the high memory
999	requirements are caused by requiring a session for each watcher). Watcher	1604	requirements are caused by requiring a session for each watcher). Watcher
1000	invocation speed is almost 900 times slower than with AnyEvent's pure perl	1605	invocation speed is almost 900 times slower than with AnyEvent's pure perl
		1606	implementation.
		1607
1001	implementation. The design of the POE adaptor class in AnyEvent can not	1608	The design of the POE adaptor class in AnyEvent can not really account
1002	really account for this, as session creation overhead is small compared	1609	for the performance issues, though, as session creation overhead is
1003	to execution of the state machine, which is coded pretty optimally within	1610	small compared to execution of the state machine, which is coded pretty
1004	L<AnyEvent::Impl::POE>. POE simply seems to be abysmally slow.	1611	optimally within L<AnyEvent::Impl::POE> (and while everybody agrees that
		1612	using multiple sessions is not a good approach, especially regarding
		1613	memory usage, even the author of POE could not come up with a faster
		1614	design).
1005		1615
1006	=head2 Summary	1616	=head3 Summary
1007		1617
1008	=over 4	1618	=over 4
1009		1619
1010	=item * Using EV through AnyEvent is faster than any other event loop	1620	=item * Using EV through AnyEvent is faster than any other event loop
1011	(even when used without AnyEvent), but most event loops have acceptable	1621	(even when used without AnyEvent), but most event loops have acceptable
…		…
1013		1623
1014	=item * The overhead AnyEvent adds is usually much smaller than the overhead of	1624	=item * The overhead AnyEvent adds is usually much smaller than the overhead of
1015	the actual event loop, only with extremely fast event loops such as EV	1625	the actual event loop, only with extremely fast event loops such as EV
1016	adds AnyEvent significant overhead.	1626	adds AnyEvent significant overhead.
1017		1627
1018	=item * You should simply avoid POE like the plague if you want performance or	1628	=item * You should avoid POE like the plague if you want performance or
1019	reasonable memory usage.	1629	reasonable memory usage.
1020		1630
1021	=back	1631	=back
1022		1632
		1633	=head2 BENCHMARKING THE LARGE SERVER CASE
		1634
		1635	This benchmark actually benchmarks the event loop itself. It works by
		1636	creating a number of "servers": each server consists of a socket pair, a
		1637	timeout watcher that gets reset on activity (but never fires), and an I/O
		1638	watcher waiting for input on one side of the socket. Each time the socket
		1639	watcher reads a byte it will write that byte to a random other "server".
		1640
		1641	The effect is that there will be a lot of I/O watchers, only part of which
		1642	are active at any one point (so there is a constant number of active
		1643	fds for each loop iteration, but which fds these are is random). The
		1644	timeout is reset each time something is read because that reflects how
		1645	most timeouts work (and puts extra pressure on the event loops).
		1646
		1647	In this benchmark, we use 10000 socket pairs (20000 sockets), of which 100
		1648	(1%) are active. This mirrors the activity of large servers with many
		1649	connections, most of which are idle at any one point in time.
		1650
		1651	Source code for this benchmark is found as F<eg/bench2> in the AnyEvent
		1652	distribution.
		1653
		1654	=head3 Explanation of the columns
		1655
		1656	I<sockets> is the number of sockets, and twice the number of "servers" (as
		1657	each server has a read and write socket end).
		1658
		1659	I<create> is the time it takes to create a socket pair (which is
		1660	nontrivial) and two watchers: an I/O watcher and a timeout watcher.
		1661
		1662	I<request>, the most important value, is the time it takes to handle a
		1663	single "request", that is, reading the token from the pipe and forwarding
		1664	it to another server. This includes deleting the old timeout and creating
		1665	a new one that moves the timeout into the future.
		1666
		1667	=head3 Results
		1668
		1669	name sockets create request
		1670	EV 20000 69.01 11.16
		1671	Perl 20000 73.32 35.87
		1672	Event 20000 212.62 257.32
		1673	Glib 20000 651.16 1896.30
		1674	POE 20000 349.67 12317.24 uses POE::Loop::Event
		1675
		1676	=head3 Discussion
		1677
		1678	This benchmark I<does> measure scalability and overall performance of the
		1679	particular event loop.
		1680
		1681	EV is again fastest. Since it is using epoll on my system, the setup time
		1682	is relatively high, though.
		1683
		1684	Perl surprisingly comes second. It is much faster than the C-based event
		1685	loops Event and Glib.
		1686
		1687	Event suffers from high setup time as well (look at its code and you will
		1688	understand why). Callback invocation also has a high overhead compared to
		1689	the C<< $_->() for .. >>-style loop that the Perl event loop uses. Event
		1690	uses select or poll in basically all documented configurations.
		1691
		1692	Glib is hit hard by its quadratic behaviour w.r.t. many watchers. It
		1693	clearly fails to perform with many filehandles or in busy servers.
		1694
		1695	POE is still completely out of the picture, taking over 1000 times as long
		1696	as EV, and over 100 times as long as the Perl implementation, even though
		1697	it uses a C-based event loop in this case.
		1698
		1699	=head3 Summary
		1700
		1701	=over 4
		1702
		1703	=item * The pure perl implementation performs extremely well.
		1704
		1705	=item * Avoid Glib or POE in large projects where performance matters.
		1706
		1707	=back
		1708
		1709	=head2 BENCHMARKING SMALL SERVERS
		1710
		1711	While event loops should scale (and select-based ones do not...) even to
		1712	large servers, most programs we (or I :) actually write have only a few
		1713	I/O watchers.
		1714
		1715	In this benchmark, I use the same benchmark program as in the large server
		1716	case, but it uses only eight "servers", of which three are active at any
		1717	one time. This should reflect performance for a small server relatively
		1718	well.
		1719
		1720	The columns are identical to the previous table.
		1721
		1722	=head3 Results
		1723
		1724	name sockets create request
		1725	EV 16 20.00 6.54
		1726	Perl 16 25.75 12.62
		1727	Event 16 81.27 35.86
		1728	Glib 16 32.63 15.48
		1729	POE 16 261.87 276.28 uses POE::Loop::Event
		1730
		1731	=head3 Discussion
		1732
		1733	The benchmark tries to test the performance of a typical small
		1734	server. While knowing how various event loops perform is interesting, keep
		1735	in mind that their overhead in this case is usually not as important, due
		1736	to the small absolute number of watchers (that is, you need efficiency and
		1737	speed most when you have lots of watchers, not when you only have a few of
		1738	them).
		1739
		1740	EV is again fastest.
		1741
		1742	Perl again comes second. It is noticeably faster than the C-based event
		1743	loops Event and Glib, although the difference is too small to really
		1744	matter.
		1745
		1746	POE also performs much better in this case, but is is still far behind the
		1747	others.
		1748
		1749	=head3 Summary
		1750
		1751	=over 4
		1752
		1753	=item * C-based event loops perform very well with small number of
		1754	watchers, as the management overhead dominates.
		1755
		1756	=back
		1757
1023		1758
1024	=head1 FORK	1759	=head1 FORK
1025		1760
1026	Most event libraries are not fork-safe. The ones who are usually are	1761	Most event libraries are not fork-safe. The ones who are usually are
1027	because they are so inefficient. Only L<EV> is fully fork-aware.	1762	because they rely on inefficient but fork-safe C<select> or C<poll>
		1763	calls. Only L<EV> is fully fork-aware.
1028		1764
1029	If you have to fork, you must either do so I<before> creating your first	1765	If you have to fork, you must either do so I<before> creating your first
1030	watcher OR you must not use AnyEvent at all in the child.	1766	watcher OR you must not use AnyEvent at all in the child.
1031		1767
1032		1768
…		…
1040	specified in the variable.	1776	specified in the variable.
1041		1777
1042	You can make AnyEvent completely ignore this variable by deleting it	1778	You can make AnyEvent completely ignore this variable by deleting it
1043	before the first watcher gets created, e.g. with a C<BEGIN> block:	1779	before the first watcher gets created, e.g. with a C<BEGIN> block:
1044		1780
1045	BEGIN { delete $ENV{PERL_ANYEVENT_MODEL} }	1781	BEGIN { delete $ENV{PERL_ANYEVENT_MODEL} }
1046		1782
1047	use AnyEvent;	1783	use AnyEvent;
		1784
		1785	Similar considerations apply to $ENV{PERL_ANYEVENT_VERBOSE}, as that can
		1786	be used to probe what backend is used and gain other information (which is
		1787	probably even less useful to an attacker than PERL_ANYEVENT_MODEL), and
		1788	$ENV{PERL_ANYEGENT_STRICT}.
		1789
		1790
		1791	=head1 BUGS
		1792
		1793	Perl 5.8 has numerous memleaks that sometimes hit this module and are hard
		1794	to work around. If you suffer from memleaks, first upgrade to Perl 5.10
		1795	and check wether the leaks still show up. (Perl 5.10.0 has other annoying
		1796	mamleaks, such as leaking on C<map> and C<grep> but it is usually not as
		1797	pronounced).
1048		1798
1049		1799
1050	=head1 SEE ALSO	1800	=head1 SEE ALSO
1051		1801
1052	Event modules: L<Coro::EV>, L<EV>, L<EV::Glib>, L<Glib::EV>,	1802	Utility functions: L<AnyEvent::Util>.
1053	L<Coro::Event>, L<Event>, L<Glib::Event>, L<Glib>, L<Coro>, L<Tk>,	1803
		1804	Event modules: L<EV>, L<EV::Glib>, L<Glib::EV>, L<Event>, L<Glib::Event>,
1054	L<Event::Lib>, L<Qt>, L<POE>.	1805	L<Glib>, L<Tk>, L<Event::Lib>, L<Qt>, L<POE>.
1055		1806
1056	Implementations: L<AnyEvent::Impl::CoroEV>, L<AnyEvent::Impl::EV>,	1807	Implementations: L<AnyEvent::Impl::EV>, L<AnyEvent::Impl::Event>,
1057	L<AnyEvent::Impl::CoroEvent>, L<AnyEvent::Impl::Event>, L<AnyEvent::Impl::Glib>,	1808	L<AnyEvent::Impl::Glib>, L<AnyEvent::Impl::Tk>, L<AnyEvent::Impl::Perl>,
1058	L<AnyEvent::Impl::Tk>, L<AnyEvent::Impl::Perl>, L<AnyEvent::Impl::EventLib>,	1809	L<AnyEvent::Impl::EventLib>, L<AnyEvent::Impl::Qt>,
1059	L<AnyEvent::Impl::Qt>, L<AnyEvent::Impl::POE>.	1810	L<AnyEvent::Impl::POE>.
1060		1811
		1812	Non-blocking file handles, sockets, TCP clients and
		1813	servers: L<AnyEvent::Handle>, L<AnyEvent::Socket>.
		1814
		1815	Asynchronous DNS: L<AnyEvent::DNS>.
		1816
		1817	Coroutine support: L<Coro>, L<Coro::AnyEvent>, L<Coro::EV>, L<Coro::Event>,
		1818
1061	Nontrivial usage examples: L<Net::FCP>, L<Net::XMPP2>.	1819	Nontrivial usage examples: L<Net::FCP>, L<Net::XMPP2>, L<AnyEvent::DNS>.
1062		1820
1063		1821
1064	=head1 AUTHOR	1822	=head1 AUTHOR
1065		1823
1066	Marc Lehmann <schmorp@schmorp.de>	1824	Marc Lehmann <schmorp@schmorp.de>
1067	http://home.schmorp.de/	1825	http://home.schmorp.de/
1068		1826
1069	=cut	1827	=cut
1070		1828
1071	1	1829	1
1072		1830

Diff Legend

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