ViewVC Help

View File | Revision Log | Show Annotations | Download File

/cvs/AnyEvent/lib/AnyEvent.pm

Comparing AnyEvent/lib/AnyEvent.pm (file contents):
Revision 1.83 by root, Fri Apr 25 13:39:08 2008 UTC vs.
Revision 1.340 by root, Fri Dec 3 18:39:06 2010 UTC

1	=head1 NAME	1	=head1 NAME
2		2
3	AnyEvent - provide framework for multiple event loops	3	AnyEvent - the DBI of event loop programming
4		4
5	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, Irssi, rxvt-unicode, IO::Async, Qt
		6	and POE are various supported event loops/environments.
6		7
7	=head1 SYNOPSIS	8	=head1 SYNOPSIS
8		9
9	use AnyEvent;	10	use AnyEvent;
10		11
		12	# if you prefer function calls, look at the AE manpage for
		13	# an alternative API.
		14
		15	# file handle or descriptor readable
11	my $w = AnyEvent->io (fh => $fh, poll => "r|w", cb => sub {	16	my $w = AnyEvent->io (fh => $fh, poll => "r", cb => sub { ... });
		17
		18	# one-shot or repeating timers
		19	my $w = AnyEvent->timer (after => $seconds, cb => sub { ... });
		20	my $w = AnyEvent->timer (after => $seconds, interval => $seconds, cb => ...);
		21
		22	print AnyEvent->now; # prints current event loop time
		23	print AnyEvent->time; # think Time::HiRes::time or simply CORE::time.
		24
		25	# POSIX signal
		26	my $w = AnyEvent->signal (signal => "TERM", cb => sub { ... });
		27
		28	# child process exit
		29	my $w = AnyEvent->child (pid => $pid, cb => sub {
		30	my ($pid, $status) = @_;
12	...	31	...
13	});	32	});
14		33
15	my $w = AnyEvent->timer (after => $seconds, cb => sub {	34	# called when event loop idle (if applicable)
16	...	35	my $w = AnyEvent->idle (cb => sub { ... });
17	});
18		36
19	my $w = AnyEvent->condvar; # stores whether a condition was flagged	37	my $w = AnyEvent->condvar; # stores whether a condition was flagged
		38	$w->send; # wake up current and all future recv's
20	$w->wait; # enters "main loop" till $condvar gets ->broadcast	39	$w->recv; # enters "main loop" till $condvar gets ->send
21	$w->broadcast; # wake up current and all future wait's	40	# use a condvar in callback mode:
		41	$w->cb (sub { $_[0]->recv });
		42
		43	=head1 INTRODUCTION/TUTORIAL
		44
		45	This manpage is mainly a reference manual. If you are interested
		46	in a tutorial or some gentle introduction, have a look at the
		47	L<AnyEvent::Intro> manpage.
		48
		49	=head1 SUPPORT
		50
		51	An FAQ document is available as L<AnyEvent::FAQ>.
		52
		53	There also is a mailinglist for discussing all things AnyEvent, and an IRC
		54	channel, too.
		55
		56	See the AnyEvent project page at the B<Schmorpforge Ta-Sa Software
		57	Repository>, at L<http://anyevent.schmorp.de>, for more info.
22		58
23	=head1 WHY YOU SHOULD USE THIS MODULE (OR NOT)	59	=head1 WHY YOU SHOULD USE THIS MODULE (OR NOT)
24		60
25	Glib, POE, IO::Async, Event... CPAN offers event models by the dozen	61	Glib, POE, IO::Async, Event... CPAN offers event models by the dozen
26	nowadays. So what is different about AnyEvent?	62	nowadays. So what is different about AnyEvent?
27		63
28	Executive Summary: AnyEvent is I<compatible>, AnyEvent is I<free of	64	Executive Summary: AnyEvent is I<compatible>, AnyEvent is I<free of
29	policy> and AnyEvent is I<small and efficient>.	65	policy> and AnyEvent is I<small and efficient>.
30		66
31	First and foremost, I<AnyEvent is not an event model> itself, it only	67	First and foremost, I<AnyEvent is not an event model> itself, it only
32	interfaces to whatever event model the main program happens to use in a	68	interfaces to whatever event model the main program happens to use, in a
33	pragmatic way. For event models and certain classes of immortals alike,	69	pragmatic way. For event models and certain classes of immortals alike,
34	the statement "there can only be one" is a bitter reality: In general,	70	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	71	only one event loop can be active at the same time in a process. AnyEvent
36	helps hiding the differences between those event loops.	72	cannot change this, but it can hide the differences between those event
		73	loops.
37		74
38	The goal of AnyEvent is to offer module authors the ability to do event	75	The goal of AnyEvent is to offer module authors the ability to do event
39	programming (waiting for I/O or timer events) without subscribing to a	76	programming (waiting for I/O or timer events) without subscribing to a
40	religion, a way of living, and most importantly: without forcing your	77	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	78	module users into the same thing by forcing them to use the same event
42	model you use.	79	model you use.
43		80
44	For modules like POE or IO::Async (which is a total misnomer as it is	81	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	82	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	83	like joining a cult: After you join, you are dependent on them and you
47	cannot use anything else, as it is simply incompatible to everything that	84	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	85	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.	86	module are I<also> forced to use the same event loop you use.
50		87
51	AnyEvent is different: AnyEvent + POE works fine. AnyEvent + Glib works	88	AnyEvent is different: AnyEvent + POE works fine. AnyEvent + Glib works
52	fine. AnyEvent + Tk works fine etc. etc. but none of these work together	89	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	90	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,	91	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	92	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	93	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	94	use one of the supported event loops. It is easy to add new event loops
58	event loops to AnyEvent, too, so it is future-proof).	95	to AnyEvent, too, so it is future-proof).
59		96
60	In addition to being free of having to use I<the one and only true event	97	In addition to being free of having to use I<the one and only true event
61	model>, AnyEvent also is free of bloat and policy: with POE or similar	98	model>, AnyEvent also is free of bloat and policy: with POE or similar
62	modules, you get an enourmous amount of code and strict rules you have to	99	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	100	follow. AnyEvent, on the other hand, is lean and to the point, by only
64	offering the functionality that is necessary, in as thin as a wrapper as	101	offering the functionality that is necessary, in as thin as a wrapper as
65	technically possible.	102	technically possible.
66		103
		104	Of course, AnyEvent comes with a big (and fully optional!) toolbox
		105	of useful functionality, such as an asynchronous DNS resolver, 100%
		106	non-blocking connects (even with TLS/SSL, IPv6 and on broken platforms
		107	such as Windows) and lots of real-world knowledge and workarounds for
		108	platform bugs and differences.
		109
67	Of course, if you want lots of policy (this can arguably be somewhat	110	Now, if you I<do want> lots of policy (this can arguably be somewhat
68	useful) and you want to force your users to use the one and only event	111	useful) and you want to force your users to use the one and only event
69	model, you should I<not> use this module.	112	model, you should I<not> use this module.
70		113
71
72	=head1 DESCRIPTION	114	=head1 DESCRIPTION
73		115
74	L<AnyEvent> provides an identical interface to multiple event loops. This	116	L<AnyEvent> provides a uniform interface to various event loops. This
75	allows module authors to utilise an event loop without forcing module	117	allows module authors to use event loop functionality without forcing
76	users to use the same event loop (as only a single event loop can coexist	118	module users to use a specific event loop implementation (since more
77	peacefully at any one time).	119	than one event loop cannot coexist peacefully).
78		120
79	The interface itself is vaguely similar, but not identical to the L<Event>	121	The interface itself is vaguely similar, but not identical to the L<Event>
80	module.	122	module.
81		123
82	During the first call of any watcher-creation method, the module tries	124	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	125	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>,	126	following modules is already loaded: L<EV>, L<AnyEvent::Impl::Perl>,
85	L<Event>, L<Glib>, L<AnyEvent::Impl::Perl>, L<Tk>, L<Event::Lib>, L<Qt>,	127	L<Event>, L<Glib>, L<Tk>, L<Event::Lib>, L<Qt>, L<POE>. The first one
86	L<POE>. The first one found is used. If none are found, the module tries	128	found is used. If none are detected, the module tries to load the first
87	to load these modules (excluding Tk, Event::Lib, Qt and POE as the pure perl	129	four modules in the order given; but note that if L<EV> is not
88	adaptor should always succeed) in the order given. The first one that can	130	available, the pure-perl L<AnyEvent::Impl::Perl> should always work, so
89	be successfully loaded will be used. If, after this, still none could be	131	the other two are not normally tried.
90	found, AnyEvent will fall back to a pure-perl event loop, which is not
91	very efficient, but should work everywhere.
92		132
93	Because AnyEvent first checks for modules that are already loaded, loading	133	Because AnyEvent first checks for modules that are already loaded, loading
94	an event model explicitly before first using AnyEvent will likely make	134	an event model explicitly before first using AnyEvent will likely make
95	that model the default. For example:	135	that model the default. For example:
96		136
…		…
98	use AnyEvent;	138	use AnyEvent;
99		139
100	# .. AnyEvent will likely default to Tk	140	# .. AnyEvent will likely default to Tk
101		141
102	The I<likely> means that, if any module loads another event model and	142	The I<likely> means that, if any module loads another event model and
103	starts using it, all bets are off. Maybe you should tell their authors to	143	starts using it, all bets are off - this case should be very rare though,
104	use AnyEvent so their modules work together with others seamlessly...	144	as very few modules hardcode event loops without announcing this very
		145	loudly.
105		146
106	The pure-perl implementation of AnyEvent is called	147	The pure-perl implementation of AnyEvent is called
107	C<AnyEvent::Impl::Perl>. Like other event modules you can load it	148	C<AnyEvent::Impl::Perl>. Like other event modules you can load it
108	explicitly.	149	explicitly and enjoy the high availability of that event loop :)
109		150
110	=head1 WATCHERS	151	=head1 WATCHERS
111		152
112	AnyEvent has the central concept of a I<watcher>, which is an object that	153	AnyEvent has the central concept of a I<watcher>, which is an object that
113	stores relevant data for each kind of event you are waiting for, such as	154	stores relevant data for each kind of event you are waiting for, such as
114	the callback to call, the filehandle to watch, etc.	155	the callback to call, the file handle to watch, etc.
115		156
116	These watchers are normal Perl objects with normal Perl lifetime. After	157	These watchers are normal Perl objects with normal Perl lifetime. After
117	creating a watcher it will immediately "watch" for events and invoke the	158	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	159	callback when the event occurs (of course, only when the event model
119	is in control).	160	is in control).
120		161
		162	Note that B<callbacks must not permanently change global variables>
		163	potentially in use by the event loop (such as C<$_> or C<$[>) and that B<<
		164	callbacks must not C<die> >>. The former is good programming practice in
		165	Perl and the latter stems from the fact that exception handling differs
		166	widely between event loops.
		167
121	To disable the watcher you have to destroy it (e.g. by setting the	168	To disable a watcher you have to destroy it (e.g. by setting the
122	variable you store it in to C<undef> or otherwise deleting all references	169	variable you store it in to C<undef> or otherwise deleting all references
123	to it).	170	to it).
124		171
125	All watchers are created by calling a method on the C<AnyEvent> class.	172	All watchers are created by calling a method on the C<AnyEvent> class.
126		173
127	Many watchers either are used with "recursion" (repeating timers for	174	Many watchers either are used with "recursion" (repeating timers for
128	example), or need to refer to their watcher object in other ways.	175	example), or need to refer to their watcher object in other ways.
129		176
130	An any way to achieve that is this pattern:	177	One way to achieve that is this pattern:
131		178
132	my $w; $w = AnyEvent->type (arg => value ..., cb => sub {	179	my $w; $w = AnyEvent->type (arg => value ..., cb => sub {
133	# you can use $w here, for example to undef it	180	# you can use $w here, for example to undef it
134	undef $w;	181	undef $w;
135	});	182	});
136		183
137	Note that C<my $w; $w => combination. This is necessary because in Perl,	184	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	185	my variables are only visible after the statement in which they are
139	declared.	186	declared.
140		187
141	=head2 I/O WATCHERS	188	=head2 I/O WATCHERS
142		189
		190	$w = AnyEvent->io (
		191	fh => <filehandle_or_fileno>,
		192	poll => <"r" or "w">,
		193	cb => <callback>,
		194	);
		195
143	You can create an I/O watcher by calling the C<< AnyEvent->io >> method	196	You can create an I/O watcher by calling the C<< AnyEvent->io >> method
144	with the following mandatory key-value pairs as arguments:	197	with the following mandatory key-value pairs as arguments:
145		198
146	C<fh> the Perl I<file handle> (I<not> file descriptor) to watch for	199	C<fh> is the Perl I<file handle> (or a naked file descriptor) to watch
		200	for events (AnyEvent might or might not keep a reference to this file
		201	handle). Note that only file handles pointing to things for which
		202	non-blocking operation makes sense are allowed. This includes sockets,
		203	most character devices, pipes, fifos and so on, but not for example files
		204	or block devices.
		205
147	events. C<poll> must be a string that is either C<r> or C<w>, which	206	C<poll> must be a string that is either C<r> or C<w>, which creates a
148	creates a watcher waiting for "r"eadable or "w"ritable events,	207	watcher waiting for "r"eadable or "w"ritable events, respectively.
		208
149	respectively. C<cb> is the callback to invoke each time the file handle	209	C<cb> is the callback to invoke each time the file handle becomes ready.
150	becomes ready.	210
		211	Although the callback might get passed parameters, their value and
		212	presence is undefined and you cannot rely on them. Portable AnyEvent
		213	callbacks cannot use arguments passed to I/O watcher callbacks.
151		214
152	The I/O watcher might use the underlying file descriptor or a copy of it.	215	The I/O watcher might use the underlying file descriptor or a copy of it.
153	It is not allowed to close a file handle as long as any watcher is active	216	You must not close a file handle as long as any watcher is active on the
154	on the underlying file descriptor.	217	underlying file descriptor.
155		218
156	Some event loops issue spurious readyness notifications, so you should	219	Some event loops issue spurious readiness notifications, so you should
157	always use non-blocking calls when reading/writing from/to your file	220	always use non-blocking calls when reading/writing from/to your file
158	handles.	221	handles.
159		222
160	Example:
161
162	# wait for readability of STDIN, then read a line and disable the watcher	223	Example: wait for readability of STDIN, then read a line and disable the
		224	watcher.
		225
163	my $w; $w = AnyEvent->io (fh => \*STDIN, poll => 'r', cb => sub {	226	my $w; $w = AnyEvent->io (fh => \*STDIN, poll => 'r', cb => sub {
164	chomp (my $input = <STDIN>);	227	chomp (my $input = <STDIN>);
165	warn "read: $input\n";	228	warn "read: $input\n";
166	undef $w;	229	undef $w;
167	});	230	});
168		231
169	=head2 TIME WATCHERS	232	=head2 TIME WATCHERS
170		233
		234	$w = AnyEvent->timer (after => <seconds>, cb => <callback>);
		235
		236	$w = AnyEvent->timer (
		237	after => <fractional_seconds>,
		238	interval => <fractional_seconds>,
		239	cb => <callback>,
		240	);
		241
171	You can create a time watcher by calling the C<< AnyEvent->timer >>	242	You can create a time watcher by calling the C<< AnyEvent->timer >>
172	method with the following mandatory arguments:	243	method with the following mandatory arguments:
173		244
174	C<after> specifies after how many seconds (fractional values are	245	C<after> specifies after how many seconds (fractional values are
175	supported) should the timer activate. C<cb> the callback to invoke in that	246	supported) the callback should be invoked. C<cb> is the callback to invoke
176	case.	247	in that case.
177		248
178	The timer callback will be invoked at most once: if you want a repeating	249	Although the callback might get passed parameters, their value and
179	timer you have to create a new watcher (this is a limitation by both Tk	250	presence is undefined and you cannot rely on them. Portable AnyEvent
180	and Glib).	251	callbacks cannot use arguments passed to time watcher callbacks.
181		252
182	Example:	253	The callback will normally be invoked only once. If you specify another
		254	parameter, C<interval>, as a strictly positive number (> 0), then the
		255	callback will be invoked regularly at that interval (in fractional
		256	seconds) after the first invocation. If C<interval> is specified with a
		257	false value, then it is treated as if it were not specified at all.
183		258
		259	The callback will be rescheduled before invoking the callback, but no
		260	attempt is made to avoid timer drift in most backends, so the interval is
		261	only approximate.
		262
184	# fire an event after 7.7 seconds	263	Example: fire an event after 7.7 seconds.
		264
185	my $w = AnyEvent->timer (after => 7.7, cb => sub {	265	my $w = AnyEvent->timer (after => 7.7, cb => sub {
186	warn "timeout\n";	266	warn "timeout\n";
187	});	267	});
188		268
189	# to cancel the timer:	269	# to cancel the timer:
190	undef $w;	270	undef $w;
191		271
192	Example 2:
193
194	# fire an event after 0.5 seconds, then roughly every second	272	Example 2: fire an event after 0.5 seconds, then roughly every second.
195	my $w;
196		273
197	my $cb = sub {
198	# cancel the old timer while creating a new one
199	$w = AnyEvent->timer (after => 1, cb => $cb);	274	my $w = AnyEvent->timer (after => 0.5, interval => 1, cb => sub {
		275	warn "timeout\n";
200	};	276	};
201
202	# start the "loop" by creating the first watcher
203	$w = AnyEvent->timer (after => 0.5, cb => $cb);
204		277
205	=head3 TIMING ISSUES	278	=head3 TIMING ISSUES
206		279
207	There are two ways to handle timers: based on real time (relative, "fire	280	There are two ways to handle timers: based on real time (relative, "fire
208	in 10 seconds") and based on wallclock time (absolute, "fire at 12	281	in 10 seconds") and based on wallclock time (absolute, "fire at 12
…		…
210		283
211	While most event loops expect timers to specified in a relative way, they	284	While most event loops expect timers to specified in a relative way, they
212	use absolute time internally. This makes a difference when your clock	285	use absolute time internally. This makes a difference when your clock
213	"jumps", for example, when ntp decides to set your clock backwards from	286	"jumps", for example, when ntp decides to set your clock backwards from
214	the wrong date of 2014-01-01 to 2008-01-01, a watcher that is supposed to	287	the wrong date of 2014-01-01 to 2008-01-01, a watcher that is supposed to
215	fire "after" a second might actually take six years to finally fire.	288	fire "after a second" might actually take six years to finally fire.
216		289
217	AnyEvent cannot compensate for this. The only event loop that is conscious	290	AnyEvent cannot compensate for this. The only event loop that is conscious
218	about these issues is L<EV>, which offers both relative (ev_timer, based	291	of these issues is L<EV>, which offers both relative (ev_timer, based
219	on true relative time) and absolute (ev_periodic, based on wallclock time)	292	on true relative time) and absolute (ev_periodic, based on wallclock time)
220	timers.	293	timers.
221		294
222	AnyEvent always prefers relative timers, if available, matching the	295	AnyEvent always prefers relative timers, if available, matching the
223	AnyEvent API.	296	AnyEvent API.
224		297
		298	AnyEvent has two additional methods that return the "current time":
		299
		300	=over 4
		301
		302	=item AnyEvent->time
		303
		304	This returns the "current wallclock time" as a fractional number of
		305	seconds since the Epoch (the same thing as C<time> or C<Time::HiRes::time>
		306	return, and the result is guaranteed to be compatible with those).
		307
		308	It progresses independently of any event loop processing, i.e. each call
		309	will check the system clock, which usually gets updated frequently.
		310
		311	=item AnyEvent->now
		312
		313	This also returns the "current wallclock time", but unlike C<time>, above,
		314	this value might change only once per event loop iteration, depending on
		315	the event loop (most return the same time as C<time>, above). This is the
		316	time that AnyEvent's timers get scheduled against.
		317
		318	I<In almost all cases (in all cases if you don't care), this is the
		319	function to call when you want to know the current time.>
		320
		321	This function is also often faster then C<< AnyEvent->time >>, and
		322	thus the preferred method if you want some timestamp (for example,
		323	L<AnyEvent::Handle> uses this to update its activity timeouts).
		324
		325	The rest of this section is only of relevance if you try to be very exact
		326	with your timing; you can skip it without a bad conscience.
		327
		328	For a practical example of when these times differ, consider L<Event::Lib>
		329	and L<EV> and the following set-up:
		330
		331	The event loop is running and has just invoked one of your callbacks at
		332	time=500 (assume no other callbacks delay processing). In your callback,
		333	you wait a second by executing C<sleep 1> (blocking the process for a
		334	second) and then (at time=501) you create a relative timer that fires
		335	after three seconds.
		336
		337	With L<Event::Lib>, C<< AnyEvent->time >> and C<< AnyEvent->now >> will
		338	both return C<501>, because that is the current time, and the timer will
		339	be scheduled to fire at time=504 (C<501> + C<3>).
		340
		341	With L<EV>, C<< AnyEvent->time >> returns C<501> (as that is the current
		342	time), but C<< AnyEvent->now >> returns C<500>, as that is the time the
		343	last event processing phase started. With L<EV>, your timer gets scheduled
		344	to run at time=503 (C<500> + C<3>).
		345
		346	In one sense, L<Event::Lib> is more exact, as it uses the current time
		347	regardless of any delays introduced by event processing. However, most
		348	callbacks do not expect large delays in processing, so this causes a
		349	higher drift (and a lot more system calls to get the current time).
		350
		351	In another sense, L<EV> is more exact, as your timer will be scheduled at
		352	the same time, regardless of how long event processing actually took.
		353
		354	In either case, if you care (and in most cases, you don't), then you
		355	can get whatever behaviour you want with any event loop, by taking the
		356	difference between C<< AnyEvent->time >> and C<< AnyEvent->now >> into
		357	account.
		358
		359	=item AnyEvent->now_update
		360
		361	Some event loops (such as L<EV> or L<AnyEvent::Impl::Perl>) cache
		362	the current time for each loop iteration (see the discussion of L<<
		363	AnyEvent->now >>, above).
		364
		365	When a callback runs for a long time (or when the process sleeps), then
		366	this "current" time will differ substantially from the real time, which
		367	might affect timers and time-outs.
		368
		369	When this is the case, you can call this method, which will update the
		370	event loop's idea of "current time".
		371
		372	A typical example would be a script in a web server (e.g. C<mod_perl>) -
		373	when mod_perl executes the script, then the event loop will have the wrong
		374	idea about the "current time" (being potentially far in the past, when the
		375	script ran the last time). In that case you should arrange a call to C<<
		376	AnyEvent->now_update >> each time the web server process wakes up again
		377	(e.g. at the start of your script, or in a handler).
		378
		379	Note that updating the time I<might> cause some events to be handled.
		380
		381	=back
		382
225	=head2 SIGNAL WATCHERS	383	=head2 SIGNAL WATCHERS
226		384
		385	$w = AnyEvent->signal (signal => <uppercase_signal_name>, cb => <callback>);
		386
227	You can watch for signals using a signal watcher, C<signal> is the signal	387	You can watch for signals using a signal watcher, C<signal> is the signal
228	I<name> without any C<SIG> prefix, C<cb> is the Perl callback to	388	I<name> in uppercase and without any C<SIG> prefix, C<cb> is the Perl
229	be invoked whenever a signal occurs.	389	callback to be invoked whenever a signal occurs.
230		390
		391	Although the callback might get passed parameters, their value and
		392	presence is undefined and you cannot rely on them. Portable AnyEvent
		393	callbacks cannot use arguments passed to signal watcher callbacks.
		394
231	Multiple signal occurances can be clumped together into one callback	395	Multiple signal occurrences can be clumped together into one callback
232	invocation, and callback invocation will be synchronous. synchronous means	396	invocation, and callback invocation will be synchronous. Synchronous means
233	that it might take a while until the signal gets handled by the process,	397	that it might take a while until the signal gets handled by the process,
234	but it is guarenteed not to interrupt any other callbacks.	398	but it is guaranteed not to interrupt any other callbacks.
235		399
236	The main advantage of using these watchers is that you can share a signal	400	The main advantage of using these watchers is that you can share a signal
237	between multiple watchers.	401	between multiple watchers, and AnyEvent will ensure that signals will not
		402	interrupt your program at bad times.
238		403
239	This watcher might use C<%SIG>, so programs overwriting those signals	404	This watcher might use C<%SIG> (depending on the event loop used),
240	directly will likely not work correctly.	405	so programs overwriting those signals directly will likely not work
		406	correctly.
241		407
242	Example: exit on SIGINT	408	Example: exit on SIGINT
243		409
244	my $w = AnyEvent->signal (signal => "INT", cb => sub { exit 1 });	410	my $w = AnyEvent->signal (signal => "INT", cb => sub { exit 1 });
245		411
		412	=head3 Restart Behaviour
		413
		414	While restart behaviour is up to the event loop implementation, most will
		415	not restart syscalls (that includes L<Async::Interrupt> and AnyEvent's
		416	pure perl implementation).
		417
		418	=head3 Safe/Unsafe Signals
		419
		420	Perl signals can be either "safe" (synchronous to opcode handling) or
		421	"unsafe" (asynchronous) - the former might get delayed indefinitely, the
		422	latter might corrupt your memory.
		423
		424	AnyEvent signal handlers are, in addition, synchronous to the event loop,
		425	i.e. they will not interrupt your running perl program but will only be
		426	called as part of the normal event handling (just like timer, I/O etc.
		427	callbacks, too).
		428
		429	=head3 Signal Races, Delays and Workarounds
		430
		431	Many event loops (e.g. Glib, Tk, Qt, IO::Async) do not support attaching
		432	callbacks to signals in a generic way, which is a pity, as you cannot
		433	do race-free signal handling in perl, requiring C libraries for
		434	this. AnyEvent will try to do its best, which means in some cases,
		435	signals will be delayed. The maximum time a signal might be delayed is
		436	specified in C<$AnyEvent::MAX_SIGNAL_LATENCY> (default: 10 seconds). This
		437	variable can be changed only before the first signal watcher is created,
		438	and should be left alone otherwise. This variable determines how often
		439	AnyEvent polls for signals (in case a wake-up was missed). Higher values
		440	will cause fewer spurious wake-ups, which is better for power and CPU
		441	saving.
		442
		443	All these problems can be avoided by installing the optional
		444	L<Async::Interrupt> module, which works with most event loops. It will not
		445	work with inherently broken event loops such as L<Event> or L<Event::Lib>
		446	(and not with L<POE> currently, as POE does its own workaround with
		447	one-second latency). For those, you just have to suffer the delays.
		448
246	=head2 CHILD PROCESS WATCHERS	449	=head2 CHILD PROCESS WATCHERS
247		450
		451	$w = AnyEvent->child (pid => <process id>, cb => <callback>);
		452
248	You can also watch on a child process exit and catch its exit status.	453	You can also watch for a child process exit and catch its exit status.
249		454
250	The child process is specified by the C<pid> argument (if set to C<0>, it	455	The child process is specified by the C<pid> argument (on some backends,
251	watches for any child process exit). The watcher will trigger as often	456	using C<0> watches for any child process exit, on others this will
252	as status change for the child are received. This works by installing a	457	croak). The watcher will be triggered only when the child process has
253	signal handler for C<SIGCHLD>. The callback will be called with the pid	458	finished and an exit status is available, not on any trace events
254	and exit status (as returned by waitpid).	459	(stopped/continued).
		460
		461	The callback will be called with the pid and exit status (as returned by
		462	waitpid), so unlike other watcher types, you I<can> rely on child watcher
		463	callback arguments.
		464
		465	This watcher type works by installing a signal handler for C<SIGCHLD>,
		466	and since it cannot be shared, nothing else should use SIGCHLD or reap
		467	random child processes (waiting for specific child processes, e.g. inside
		468	C<system>, is just fine).
255		469
256	There is a slight catch to child watchers, however: you usually start them	470	There is a slight catch to child watchers, however: you usually start them
257	I<after> the child process was created, and this means the process could	471	I<after> the child process was created, and this means the process could
258	have exited already (and no SIGCHLD will be sent anymore).	472	have exited already (and no SIGCHLD will be sent anymore).
259		473
260	Not all event models handle this correctly (POE doesn't), but even for	474	Not all event models handle this correctly (neither POE nor IO::Async do,
		475	see their AnyEvent::Impl manpages for details), but even for event models
261	event models that I<do> handle this correctly, they usually need to be	476	that I<do> handle this correctly, they usually need to be loaded before
262	loaded before the process exits (i.e. before you fork in the first place).	477	the process exits (i.e. before you fork in the first place). AnyEvent's
		478	pure perl event loop handles all cases correctly regardless of when you
		479	start the watcher.
263		480
264	This means you cannot create a child watcher as the very first thing in an	481	This means you cannot create a child watcher as the very first
265	AnyEvent program, you I<have> to create at least one watcher before you	482	thing in an AnyEvent program, you I<have> to create at least one
266	C<fork> the child (alternatively, you can call C<AnyEvent::detect>).	483	watcher before you C<fork> the child (alternatively, you can call
		484	C<AnyEvent::detect>).
		485
		486	As most event loops do not support waiting for child events, they will be
		487	emulated by AnyEvent in most cases, in which the latency and race problems
		488	mentioned in the description of signal watchers apply.
267		489
268	Example: fork a process and wait for it	490	Example: fork a process and wait for it
269		491
270	my $done = AnyEvent->condvar;	492	my $done = AnyEvent->condvar;
271		493
272	AnyEvent::detect; # force event module to be initialised
273
274	my $pid = fork or exit 5;	494	my $pid = fork or exit 5;
275		495
276	my $w = AnyEvent->child (	496	my $w = AnyEvent->child (
277	pid => $pid,	497	pid => $pid,
278	cb => sub {	498	cb => sub {
279	my ($pid, $status) = @_;	499	my ($pid, $status) = @_;
280	warn "pid $pid exited with status $status";	500	warn "pid $pid exited with status $status";
281	$done->broadcast;	501	$done->send;
282	},	502	},
283	);	503	);
284		504
285	# do something else, then wait for process exit	505	# do something else, then wait for process exit
286	$done->wait;	506	$done->recv;
		507
		508	=head2 IDLE WATCHERS
		509
		510	$w = AnyEvent->idle (cb => <callback>);
		511
		512	This will repeatedly invoke the callback after the process becomes idle,
		513	until either the watcher is destroyed or new events have been detected.
		514
		515	Idle watchers are useful when there is a need to do something, but it
		516	is not so important (or wise) to do it instantly. The callback will be
		517	invoked only when there is "nothing better to do", which is usually
		518	defined as "all outstanding events have been handled and no new events
		519	have been detected". That means that idle watchers ideally get invoked
		520	when the event loop has just polled for new events but none have been
		521	detected. Instead of blocking to wait for more events, the idle watchers
		522	will be invoked.
		523
		524	Unfortunately, most event loops do not really support idle watchers (only
		525	EV, Event and Glib do it in a usable fashion) - for the rest, AnyEvent
		526	will simply call the callback "from time to time".
		527
		528	Example: read lines from STDIN, but only process them when the
		529	program is otherwise idle:
		530
		531	my @lines; # read data
		532	my $idle_w;
		533	my $io_w = AnyEvent->io (fh => \*STDIN, poll => 'r', cb => sub {
		534	push @lines, scalar <STDIN>;
		535
		536	# start an idle watcher, if not already done
		537	$idle_w ||= AnyEvent->idle (cb => sub {
		538	# handle only one line, when there are lines left
		539	if (my $line = shift @lines) {
		540	print "handled when idle: $line";
		541	} else {
		542	# otherwise disable the idle watcher again
		543	undef $idle_w;
		544	}
		545	});
		546	});
287		547
288	=head2 CONDITION VARIABLES	548	=head2 CONDITION VARIABLES
289		549
		550	$cv = AnyEvent->condvar;
		551
		552	$cv->send (<list>);
		553	my @res = $cv->recv;
		554
		555	If you are familiar with some event loops you will know that all of them
		556	require you to run some blocking "loop", "run" or similar function that
		557	will actively watch for new events and call your callbacks.
		558
		559	AnyEvent is slightly different: it expects somebody else to run the event
		560	loop and will only block when necessary (usually when told by the user).
		561
		562	The tool to do that is called a "condition variable", so called because
		563	they represent a condition that must become true.
		564
		565	Now is probably a good time to look at the examples further below.
		566
290	Condition variables can be created by calling the C<< AnyEvent->condvar >>	567	Condition variables can be created by calling the C<< AnyEvent->condvar
291	method without any arguments.	568	>> method, usually without arguments. The only argument pair allowed is
		569	C<cb>, which specifies a callback to be called when the condition variable
		570	becomes true, with the condition variable as the first argument (but not
		571	the results).
292		572
293	A condition variable waits for a condition - precisely that the C<<	573	After creation, the condition variable is "false" until it becomes "true"
294	->broadcast >> method has been called.	574	by calling the C<send> method (or calling the condition variable as if it
		575	were a callback, read about the caveats in the description for the C<<
		576	->send >> method).
295		577
296	They are very useful to signal that a condition has been fulfilled, for	578	Since condition variables are the most complex part of the AnyEvent API, here are
		579	some different mental models of what they are - pick the ones you can connect to:
		580
		581	=over 4
		582
		583	=item * Condition variables are like callbacks - you can call them (and pass them instead
		584	of callbacks). Unlike callbacks however, you can also wait for them to be called.
		585
		586	=item * Condition variables are signals - one side can emit or send them,
		587	the other side can wait for them, or install a handler that is called when
		588	the signal fires.
		589
		590	=item * Condition variables are like "Merge Points" - points in your program
		591	where you merge multiple independent results/control flows into one.
		592
		593	=item * Condition variables represent a transaction - functions that start
		594	some kind of transaction can return them, leaving the caller the choice
		595	between waiting in a blocking fashion, or setting a callback.
		596
		597	=item * Condition variables represent future values, or promises to deliver
		598	some result, long before the result is available.
		599
		600	=back
		601
		602	Condition variables are very useful to signal that something has finished,
297	example, if you write a module that does asynchronous http requests,	603	for example, if you write a module that does asynchronous http requests,
298	then a condition variable would be the ideal candidate to signal the	604	then a condition variable would be the ideal candidate to signal the
299	availability of results.	605	availability of results. The user can either act when the callback is
		606	called or can synchronously C<< ->recv >> for the results.
300		607
301	You can also use condition variables to block your main program until	608	You can also use them to simulate traditional event loops - for example,
302	an event occurs - for example, you could C<< ->wait >> in your main	609	you can block your main program until an event occurs - for example, you
303	program until the user clicks the Quit button in your app, which would C<<	610	could C<< ->recv >> in your main program until the user clicks the Quit
304	->broadcast >> the "quit" event.	611	button of your app, which would C<< ->send >> the "quit" event.
305		612
306	Note that condition variables recurse into the event loop - if you have	613	Note that condition variables recurse into the event loop - if you have
307	two pirces of code that call C<< ->wait >> in a round-robbin fashion, you	614	two pieces of code that call C<< ->recv >> in a round-robin fashion, you
308	lose. Therefore, condition variables are good to export to your caller, but	615	lose. Therefore, condition variables are good to export to your caller, but
309	you should avoid making a blocking wait yourself, at least in callbacks,	616	you should avoid making a blocking wait yourself, at least in callbacks,
310	as this asks for trouble.	617	as this asks for trouble.
311		618
312	This object has two methods:	619	Condition variables are represented by hash refs in perl, and the keys
		620	used by AnyEvent itself are all named C<_ae_XXX> to make subclassing
		621	easy (it is often useful to build your own transaction class on top of
		622	AnyEvent). To subclass, use C<AnyEvent::CondVar> as base class and call
		623	its C<new> method in your own C<new> method.
		624
		625	There are two "sides" to a condition variable - the "producer side" which
		626	eventually calls C<< -> send >>, and the "consumer side", which waits
		627	for the send to occur.
		628
		629	Example: wait for a timer.
		630
		631	# condition: "wait till the timer is fired"
		632	my $timer_fired = AnyEvent->condvar;
		633
		634	# create the timer - we could wait for, say
		635	# a handle becomign ready, or even an
		636	# AnyEvent::HTTP request to finish, but
		637	# in this case, we simply use a timer:
		638	my $w = AnyEvent->timer (
		639	after => 1,
		640	cb => sub { $timer_fired->send },
		641	);
		642
		643	# this "blocks" (while handling events) till the callback
		644	# calls ->send
		645	$timer_fired->recv;
		646
		647	Example: wait for a timer, but take advantage of the fact that condition
		648	variables are also callable directly.
		649
		650	my $done = AnyEvent->condvar;
		651	my $delay = AnyEvent->timer (after => 5, cb => $done);
		652	$done->recv;
		653
		654	Example: Imagine an API that returns a condvar and doesn't support
		655	callbacks. This is how you make a synchronous call, for example from
		656	the main program:
		657
		658	use AnyEvent::CouchDB;
		659
		660	...
		661
		662	my @info = $couchdb->info->recv;
		663
		664	And this is how you would just set a callback to be called whenever the
		665	results are available:
		666
		667	$couchdb->info->cb (sub {
		668	my @info = $_[0]->recv;
		669	});
		670
		671	=head3 METHODS FOR PRODUCERS
		672
		673	These methods should only be used by the producing side, i.e. the
		674	code/module that eventually sends the signal. Note that it is also
		675	the producer side which creates the condvar in most cases, but it isn't
		676	uncommon for the consumer to create it as well.
313		677
314	=over 4	678	=over 4
315		679
		680	=item $cv->send (...)
		681
		682	Flag the condition as ready - a running C<< ->recv >> and all further
		683	calls to C<recv> will (eventually) return after this method has been
		684	called. If nobody is waiting the send will be remembered.
		685
		686	If a callback has been set on the condition variable, it is called
		687	immediately from within send.
		688
		689	Any arguments passed to the C<send> call will be returned by all
		690	future C<< ->recv >> calls.
		691
		692	Condition variables are overloaded so one can call them directly (as if
		693	they were a code reference). Calling them directly is the same as calling
		694	C<send>.
		695
		696	=item $cv->croak ($error)
		697
		698	Similar to send, but causes all calls to C<< ->recv >> to invoke
		699	C<Carp::croak> with the given error message/object/scalar.
		700
		701	This can be used to signal any errors to the condition variable
		702	user/consumer. Doing it this way instead of calling C<croak> directly
		703	delays the error detection, but has the overwhelming advantage that it
		704	diagnoses the error at the place where the result is expected, and not
		705	deep in some event callback with no connection to the actual code causing
		706	the problem.
		707
		708	=item $cv->begin ([group callback])
		709
316	=item $cv->wait	710	=item $cv->end
317		711
318	Wait (blocking if necessary) until the C<< ->broadcast >> method has been	712	These two methods can be used to combine many transactions/events into
319	called on c<$cv>, while servicing other watchers normally.	713	one. For example, a function that pings many hosts in parallel might want
		714	to use a condition variable for the whole process.
320		715
		716	Every call to C<< ->begin >> will increment a counter, and every call to
		717	C<< ->end >> will decrement it. If the counter reaches C<0> in C<< ->end
		718	>>, the (last) callback passed to C<begin> will be executed, passing the
		719	condvar as first argument. That callback is I<supposed> to call C<< ->send
		720	>>, but that is not required. If no group callback was set, C<send> will
		721	be called without any arguments.
		722
		723	You can think of C<< $cv->send >> giving you an OR condition (one call
		724	sends), while C<< $cv->begin >> and C<< $cv->end >> giving you an AND
		725	condition (all C<begin> calls must be C<end>'ed before the condvar sends).
		726
		727	Let's start with a simple example: you have two I/O watchers (for example,
		728	STDOUT and STDERR for a program), and you want to wait for both streams to
		729	close before activating a condvar:
		730
		731	my $cv = AnyEvent->condvar;
		732
		733	$cv->begin; # first watcher
		734	my $w1 = AnyEvent->io (fh => $fh1, cb => sub {
		735	defined sysread $fh1, my $buf, 4096
		736	or $cv->end;
		737	});
		738
		739	$cv->begin; # second watcher
		740	my $w2 = AnyEvent->io (fh => $fh2, cb => sub {
		741	defined sysread $fh2, my $buf, 4096
		742	or $cv->end;
		743	});
		744
		745	$cv->recv;
		746
		747	This works because for every event source (EOF on file handle), there is
		748	one call to C<begin>, so the condvar waits for all calls to C<end> before
		749	sending.
		750
		751	The ping example mentioned above is slightly more complicated, as the
		752	there are results to be passwd back, and the number of tasks that are
		753	begun can potentially be zero:
		754
		755	my $cv = AnyEvent->condvar;
		756
		757	my %result;
		758	$cv->begin (sub { shift->send (\%result) });
		759
		760	for my $host (@list_of_hosts) {
		761	$cv->begin;
		762	ping_host_then_call_callback $host, sub {
		763	$result{$host} = ...;
		764	$cv->end;
		765	};
		766	}
		767
		768	$cv->end;
		769
		770	This code fragment supposedly pings a number of hosts and calls
		771	C<send> after results for all then have have been gathered - in any
		772	order. To achieve this, the code issues a call to C<begin> when it starts
		773	each ping request and calls C<end> when it has received some result for
		774	it. Since C<begin> and C<end> only maintain a counter, the order in which
		775	results arrive is not relevant.
		776
		777	There is an additional bracketing call to C<begin> and C<end> outside the
		778	loop, which serves two important purposes: first, it sets the callback
		779	to be called once the counter reaches C<0>, and second, it ensures that
		780	C<send> is called even when C<no> hosts are being pinged (the loop
		781	doesn't execute once).
		782
		783	This is the general pattern when you "fan out" into multiple (but
		784	potentially zero) subrequests: use an outer C<begin>/C<end> pair to set
		785	the callback and ensure C<end> is called at least once, and then, for each
		786	subrequest you start, call C<begin> and for each subrequest you finish,
		787	call C<end>.
		788
		789	=back
		790
		791	=head3 METHODS FOR CONSUMERS
		792
		793	These methods should only be used by the consuming side, i.e. the
		794	code awaits the condition.
		795
		796	=over 4
		797
		798	=item $cv->recv
		799
		800	Wait (blocking if necessary) until the C<< ->send >> or C<< ->croak
		801	>> methods have been called on C<$cv>, while servicing other watchers
		802	normally.
		803
321	You can only wait once on a condition - additional calls will return	804	You can only wait once on a condition - additional calls are valid but
322	immediately.	805	will return immediately.
		806
		807	If an error condition has been set by calling C<< ->croak >>, then this
		808	function will call C<croak>.
		809
		810	In list context, all parameters passed to C<send> will be returned,
		811	in scalar context only the first one will be returned.
		812
		813	Note that doing a blocking wait in a callback is not supported by any
		814	event loop, that is, recursive invocation of a blocking C<< ->recv
		815	>> is not allowed, and the C<recv> call will C<croak> if such a
		816	condition is detected. This condition can be slightly loosened by using
		817	L<Coro::AnyEvent>, which allows you to do a blocking C<< ->recv >> from
		818	any thread that doesn't run the event loop itself.
323		819
324	Not all event models support a blocking wait - some die in that case	820	Not all event models support a blocking wait - some die in that case
325	(programs might want to do that to stay interactive), so I<if you are	821	(programs might want to do that to stay interactive), so I<if you are
326	using this from a module, never require a blocking wait>, but let the	822	using this from a module, never require a blocking wait>. Instead, let the
327	caller decide whether the call will block or not (for example, by coupling	823	caller decide whether the call will block or not (for example, by coupling
328	condition variables with some kind of request results and supporting	824	condition variables with some kind of request results and supporting
329	callbacks so the caller knows that getting the result will not block,	825	callbacks so the caller knows that getting the result will not block,
330	while still suppporting blocking waits if the caller so desires).	826	while still supporting blocking waits if the caller so desires).
331		827
332	Another reason I<never> to C<< ->wait >> in a module is that you cannot	828	You can ensure that C<< ->recv >> never blocks by setting a callback and
333	sensibly have two C<< ->wait >>'s in parallel, as that would require	829	only calling C<< ->recv >> from within that callback (or at a later
334	multiple interpreters or coroutines/threads, none of which C<AnyEvent>	830	time). This will work even when the event loop does not support blocking
335	can supply (the coroutine-aware backends L<AnyEvent::Impl::CoroEV> and	831	waits otherwise.
336	L<AnyEvent::Impl::CoroEvent> explicitly support concurrent C<< ->wait >>'s
337	from different coroutines, however).
338		832
339	=item $cv->broadcast	833	=item $bool = $cv->ready
340		834
341	Flag the condition as ready - a running C<< ->wait >> and all further	835	Returns true when the condition is "true", i.e. whether C<send> or
342	calls to C<wait> will (eventually) return after this method has been	836	C<croak> have been called.
343	called. If nobody is waiting the broadcast will be remembered..	837
		838	=item $cb = $cv->cb ($cb->($cv))
		839
		840	This is a mutator function that returns the callback set and optionally
		841	replaces it before doing so.
		842
		843	The callback will be called when the condition becomes "true", i.e. when
		844	C<send> or C<croak> are called, with the only argument being the
		845	condition variable itself. If the condition is already true, the
		846	callback is called immediately when it is set. Calling C<recv> inside
		847	the callback or at any later time is guaranteed not to block.
344		848
345	=back	849	=back
346		850
347	Example:	851	=head1 SUPPORTED EVENT LOOPS/BACKENDS
348		852
349	# wait till the result is ready	853	The available backend classes are (every class has its own manpage):
350	my $result_ready = AnyEvent->condvar;
351		854
352	# do something such as adding a timer	855	=over 4
353	# or socket watcher the calls $result_ready->broadcast
354	# when the "result" is ready.
355	# in this case, we simply use a timer:
356	my $w = AnyEvent->timer (
357	after => 1,
358	cb => sub { $result_ready->broadcast },
359	);
360		856
361	# this "blocks" (while handling events) till the watcher	857	=item Backends that are autoprobed when no other event loop can be found.
362	# calls broadcast	858
363	$result_ready->wait;	859	EV is the preferred backend when no other event loop seems to be in
		860	use. If EV is not installed, then AnyEvent will fall back to its own
		861	pure-perl implementation, which is available everywhere as it comes with
		862	AnyEvent itself.
		863
		864	AnyEvent::Impl::EV based on EV (interface to libev, best choice).
		865	AnyEvent::Impl::Perl pure-perl implementation, fast and portable.
		866
		867	=item Backends that are transparently being picked up when they are used.
		868
		869	These will be used if they are already loaded when the first watcher
		870	is created, in which case it is assumed that the application is using
		871	them. This means that AnyEvent will automatically pick the right backend
		872	when the main program loads an event module before anything starts to
		873	create watchers. Nothing special needs to be done by the main program.
		874
		875	AnyEvent::Impl::Event based on Event, very stable, few glitches.
		876	AnyEvent::Impl::Glib based on Glib, slow but very stable.
		877	AnyEvent::Impl::Tk based on Tk, very broken.
		878	AnyEvent::Impl::EventLib based on Event::Lib, leaks memory and worse.
		879	AnyEvent::Impl::POE based on POE, very slow, some limitations.
		880	AnyEvent::Impl::Irssi used when running within irssi.
		881
		882	=item Backends with special needs.
		883
		884	Qt requires the Qt::Application to be instantiated first, but will
		885	otherwise be picked up automatically. As long as the main program
		886	instantiates the application before any AnyEvent watchers are created,
		887	everything should just work.
		888
		889	AnyEvent::Impl::Qt based on Qt.
		890
		891	Support for IO::Async can only be partial, as it is too broken and
		892	architecturally limited to even support the AnyEvent API. It also
		893	is the only event loop that needs the loop to be set explicitly, so
		894	it can only be used by a main program knowing about AnyEvent. See
		895	L<AnyEvent::Impl::IOAsync> for the gory details.
		896
		897	AnyEvent::Impl::IOAsync based on IO::Async, cannot be autoprobed.
		898
		899	=item Event loops that are indirectly supported via other backends.
		900
		901	Some event loops can be supported via other modules:
		902
		903	There is no direct support for WxWidgets (L<Wx>) or L<Prima>.
		904
		905	B<WxWidgets> has no support for watching file handles. However, you can
		906	use WxWidgets through the POE adaptor, as POE has a Wx backend that simply
		907	polls 20 times per second, which was considered to be too horrible to even
		908	consider for AnyEvent.
		909
		910	B<Prima> is not supported as nobody seems to be using it, but it has a POE
		911	backend, so it can be supported through POE.
		912
		913	AnyEvent knows about both L<Prima> and L<Wx>, however, and will try to
		914	load L<POE> when detecting them, in the hope that POE will pick them up,
		915	in which case everything will be automatic.
		916
		917	=back
364		918
365	=head1 GLOBAL VARIABLES AND FUNCTIONS	919	=head1 GLOBAL VARIABLES AND FUNCTIONS
366		920
		921	These are not normally required to use AnyEvent, but can be useful to
		922	write AnyEvent extension modules.
		923
367	=over 4	924	=over 4
368		925
369	=item $AnyEvent::MODEL	926	=item $AnyEvent::MODEL
370		927
371	Contains C<undef> until the first watcher is being created. Then it	928	Contains C<undef> until the first watcher is being created, before the
		929	backend has been autodetected.
		930
372	contains the event model that is being used, which is the name of the	931	Afterwards it contains the event model that is being used, which is the
373	Perl class implementing the model. This class is usually one of the	932	name of the Perl class implementing the model. This class is usually one
374	C<AnyEvent::Impl:xxx> modules, but can be any other class in the case	933	of the C<AnyEvent::Impl::xxx> modules, but can be any other class in the
375	AnyEvent has been extended at runtime (e.g. in I<rxvt-unicode>).	934	case AnyEvent has been extended at runtime (e.g. in I<rxvt-unicode> it
376		935	will be C<urxvt::anyevent>).
377	The known classes so far are:
378
379	AnyEvent::Impl::CoroEV based on Coro::EV, best choice.
380	AnyEvent::Impl::CoroEvent based on Coro::Event, second best choice.
381	AnyEvent::Impl::EV based on EV (an interface to libev, best choice).
382	AnyEvent::Impl::Event based on Event, second best choice.
383	AnyEvent::Impl::Glib based on Glib, third-best choice.
384	AnyEvent::Impl::Perl pure-perl implementation, inefficient but portable.
385	AnyEvent::Impl::Tk based on Tk, very bad choice.
386	AnyEvent::Impl::Qt based on Qt, cannot be autoprobed (see its docs).
387	AnyEvent::Impl::EventLib based on Event::Lib, leaks memory and worse.
388	AnyEvent::Impl::POE based on POE, not generic enough for full support.
389
390	There is no support for WxWidgets, as WxWidgets has no support for
391	watching file handles. However, you can use WxWidgets through the
392	POE Adaptor, as POE has a Wx backend that simply polls 20 times per
393	second, which was considered to be too horrible to even consider for
394	AnyEvent. Likewise, other POE backends can be used by AnyEvent by using
395	it's adaptor.
396
397	AnyEvent knows about L<Prima> and L<Wx> and will try to use L<POE> when
398	autodetecting them.
399		936
400	=item AnyEvent::detect	937	=item AnyEvent::detect
401		938
402	Returns C<$AnyEvent::MODEL>, forcing autodetection of the event model	939	Returns C<$AnyEvent::MODEL>, forcing autodetection of the event model
403	if necessary. You should only call this function right before you would	940	if necessary. You should only call this function right before you would
404	have created an AnyEvent watcher anyway, that is, as late as possible at	941	have created an AnyEvent watcher anyway, that is, as late as possible at
405	runtime.	942	runtime, and not e.g. during initialisation of your module.
		943
		944	If you need to do some initialisation before AnyEvent watchers are
		945	created, use C<post_detect>.
		946
		947	=item $guard = AnyEvent::post_detect { BLOCK }
		948
		949	Arranges for the code block to be executed as soon as the event model is
		950	autodetected (or immediately if that has already happened).
		951
		952	The block will be executed I<after> the actual backend has been detected
		953	(C<$AnyEvent::MODEL> is set), but I<before> any watchers have been
		954	created, so it is possible to e.g. patch C<@AnyEvent::ISA> or do
		955	other initialisations - see the sources of L<AnyEvent::Strict> or
		956	L<AnyEvent::AIO> to see how this is used.
		957
		958	The most common usage is to create some global watchers, without forcing
		959	event module detection too early, for example, L<AnyEvent::AIO> creates
		960	and installs the global L<IO::AIO> watcher in a C<post_detect> block to
		961	avoid autodetecting the event module at load time.
		962
		963	If called in scalar or list context, then it creates and returns an object
		964	that automatically removes the callback again when it is destroyed (or
		965	C<undef> when the hook was immediately executed). See L<AnyEvent::AIO> for
		966	a case where this is useful.
		967
		968	Example: Create a watcher for the IO::AIO module and store it in
		969	C<$WATCHER>, but do so only do so after the event loop is initialised.
		970
		971	our WATCHER;
		972
		973	my $guard = AnyEvent::post_detect {
		974	$WATCHER = AnyEvent->io (fh => IO::AIO::poll_fileno, poll => 'r', cb => \&IO::AIO::poll_cb);
		975	};
		976
		977	# the ||= is important in case post_detect immediately runs the block,
		978	# as to not clobber the newly-created watcher. assigning both watcher and
		979	# post_detect guard to the same variable has the advantage of users being
		980	# able to just C<undef $WATCHER> if the watcher causes them grief.
		981
		982	$WATCHER ||= $guard;
		983
		984	=item @AnyEvent::post_detect
		985
		986	If there are any code references in this array (you can C<push> to it
		987	before or after loading AnyEvent), then they will be called directly
		988	after the event loop has been chosen.
		989
		990	You should check C<$AnyEvent::MODEL> before adding to this array, though:
		991	if it is defined then the event loop has already been detected, and the
		992	array will be ignored.
		993
		994	Best use C<AnyEvent::post_detect { BLOCK }> when your application allows
		995	it, as it takes care of these details.
		996
		997	This variable is mainly useful for modules that can do something useful
		998	when AnyEvent is used and thus want to know when it is initialised, but do
		999	not need to even load it by default. This array provides the means to hook
		1000	into AnyEvent passively, without loading it.
		1001
		1002	Example: To load Coro::AnyEvent whenever Coro and AnyEvent are used
		1003	together, you could put this into Coro (this is the actual code used by
		1004	Coro to accomplish this):
		1005
		1006	if (defined $AnyEvent::MODEL) {
		1007	# AnyEvent already initialised, so load Coro::AnyEvent
		1008	require Coro::AnyEvent;
		1009	} else {
		1010	# AnyEvent not yet initialised, so make sure to load Coro::AnyEvent
		1011	# as soon as it is
		1012	push @AnyEvent::post_detect, sub { require Coro::AnyEvent };
		1013	}
406		1014
407	=back	1015	=back
408		1016
409	=head1 WHAT TO DO IN A MODULE	1017	=head1 WHAT TO DO IN A MODULE
410		1018
…		…
414	Be careful when you create watchers in the module body - AnyEvent will	1022	Be careful when you create watchers in the module body - AnyEvent will
415	decide which event module to use as soon as the first method is called, so	1023	decide which event module to use as soon as the first method is called, so
416	by calling AnyEvent in your module body you force the user of your module	1024	by calling AnyEvent in your module body you force the user of your module
417	to load the event module first.	1025	to load the event module first.
418		1026
419	Never call C<< ->wait >> on a condition variable unless you I<know> that	1027	Never call C<< ->recv >> on a condition variable unless you I<know> that
420	the C<< ->broadcast >> method has been called on it already. This is	1028	the C<< ->send >> method has been called on it already. This is
421	because it will stall the whole program, and the whole point of using	1029	because it will stall the whole program, and the whole point of using
422	events is to stay interactive.	1030	events is to stay interactive.
423		1031
424	It is fine, however, to call C<< ->wait >> when the user of your module	1032	It is fine, however, to call C<< ->recv >> when the user of your module
425	requests it (i.e. if you create a http request object ad have a method	1033	requests it (i.e. if you create a http request object ad have a method
426	called C<results> that returns the results, it should call C<< ->wait >>	1034	called C<results> that returns the results, it may call C<< ->recv >>
427	freely, as the user of your module knows what she is doing. always).	1035	freely, as the user of your module knows what she is doing. Always).
428		1036
429	=head1 WHAT TO DO IN THE MAIN PROGRAM	1037	=head1 WHAT TO DO IN THE MAIN PROGRAM
430		1038
431	There will always be a single main program - the only place that should	1039	There will always be a single main program - the only place that should
432	dictate which event model to use.	1040	dictate which event model to use.
433		1041
434	If it doesn't care, it can just "use AnyEvent" and use it itself, or not	1042	If the program is not event-based, it need not do anything special, even
435	do anything special (it does not need to be event-based) and let AnyEvent	1043	when it depends on a module that uses an AnyEvent. If the program itself
436	decide which implementation to chose if some module relies on it.	1044	uses AnyEvent, but does not care which event loop is used, all it needs
		1045	to do is C<use AnyEvent>. In either case, AnyEvent will choose the best
		1046	available loop implementation.
437		1047
438	If the main program relies on a specific event model. For example, in	1048	If the main program relies on a specific event model - for example, in
439	Gtk2 programs you have to rely on the Glib module. You should load the	1049	Gtk2 programs you have to rely on the Glib module - you should load the
440	event module before loading AnyEvent or any module that uses it: generally	1050	event module before loading AnyEvent or any module that uses it: generally
441	speaking, you should load it as early as possible. The reason is that	1051	speaking, you should load it as early as possible. The reason is that
442	modules might create watchers when they are loaded, and AnyEvent will	1052	modules might create watchers when they are loaded, and AnyEvent will
443	decide on the event model to use as soon as it creates watchers, and it	1053	decide on the event model to use as soon as it creates watchers, and it
444	might chose the wrong one unless you load the correct one yourself.	1054	might choose the wrong one unless you load the correct one yourself.
445		1055
446	You can chose to use a rather inefficient pure-perl implementation by	1056	You can chose to use a pure-perl implementation by loading the
447	loading the C<AnyEvent::Impl::Perl> module, which gives you similar	1057	C<AnyEvent::Impl::Perl> module, which gives you similar behaviour
448	behaviour everywhere, but letting AnyEvent chose is generally better.	1058	everywhere, but letting AnyEvent chose the model is generally better.
		1059
		1060	=head2 MAINLOOP EMULATION
		1061
		1062	Sometimes (often for short test scripts, or even standalone programs who
		1063	only want to use AnyEvent), you do not want to run a specific event loop.
		1064
		1065	In that case, you can use a condition variable like this:
		1066
		1067	AnyEvent->condvar->recv;
		1068
		1069	This has the effect of entering the event loop and looping forever.
		1070
		1071	Note that usually your program has some exit condition, in which case
		1072	it is better to use the "traditional" approach of storing a condition
		1073	variable somewhere, waiting for it, and sending it when the program should
		1074	exit cleanly.
		1075
		1076
		1077	=head1 OTHER MODULES
		1078
		1079	The following is a non-exhaustive list of additional modules that use
		1080	AnyEvent as a client and can therefore be mixed easily with other AnyEvent
		1081	modules and other event loops in the same program. Some of the modules
		1082	come as part of AnyEvent, the others are available via CPAN.
		1083
		1084	=over 4
		1085
		1086	=item L<AnyEvent::Util>
		1087
		1088	Contains various utility functions that replace often-used blocking
		1089	functions such as C<inet_aton> with event/callback-based versions.
		1090
		1091	=item L<AnyEvent::Socket>
		1092
		1093	Provides various utility functions for (internet protocol) sockets,
		1094	addresses and name resolution. Also functions to create non-blocking tcp
		1095	connections or tcp servers, with IPv6 and SRV record support and more.
		1096
		1097	=item L<AnyEvent::Handle>
		1098
		1099	Provide read and write buffers, manages watchers for reads and writes,
		1100	supports raw and formatted I/O, I/O queued and fully transparent and
		1101	non-blocking SSL/TLS (via L<AnyEvent::TLS>).
		1102
		1103	=item L<AnyEvent::DNS>
		1104
		1105	Provides rich asynchronous DNS resolver capabilities.
		1106
		1107	=item L<AnyEvent::HTTP>, L<AnyEvent::IRC>, L<AnyEvent::XMPP>, L<AnyEvent::GPSD>, L<AnyEvent::IGS>, L<AnyEvent::FCP>
		1108
		1109	Implement event-based interfaces to the protocols of the same name (for
		1110	the curious, IGS is the International Go Server and FCP is the Freenet
		1111	Client Protocol).
		1112
		1113	=item L<AnyEvent::Handle::UDP>
		1114
		1115	Here be danger!
		1116
		1117	As Pauli would put it, "Not only is it not right, it's not even wrong!" -
		1118	there are so many things wrong with AnyEvent::Handle::UDP, most notably
		1119	its use of a stream-based API with a protocol that isn't streamable, that
		1120	the only way to improve it is to delete it.
		1121
		1122	It features data corruption (but typically only under load) and general
		1123	confusion. On top, the author is not only clueless about UDP but also
		1124	fact-resistant - some gems of his understanding: "connect doesn't work
		1125	with UDP", "UDP packets are not IP packets", "UDP only has datagrams, not
		1126	packets", "I don't need to implement proper error checking as UDP doesn't
		1127	support error checking" and so on - he doesn't even understand what's
		1128	wrong with his module when it is explained to him.
		1129
		1130	=item L<AnyEvent::DBI>
		1131
		1132	Executes L<DBI> requests asynchronously in a proxy process for you,
		1133	notifying you in an event-based way when the operation is finished.
		1134
		1135	=item L<AnyEvent::AIO>
		1136
		1137	Truly asynchronous (as opposed to non-blocking) I/O, should be in the
		1138	toolbox of every event programmer. AnyEvent::AIO transparently fuses
		1139	L<IO::AIO> and AnyEvent together, giving AnyEvent access to event-based
		1140	file I/O, and much more.
		1141
		1142	=item L<AnyEvent::HTTPD>
		1143
		1144	A simple embedded webserver.
		1145
		1146	=item L<AnyEvent::FastPing>
		1147
		1148	The fastest ping in the west.
		1149
		1150	=item L<Coro>
		1151
		1152	Has special support for AnyEvent via L<Coro::AnyEvent>.
		1153
		1154	=back
449		1155
450	=cut	1156	=cut
451		1157
452	package AnyEvent;	1158	package AnyEvent;
453		1159
454	no warnings;	1160	# basically a tuned-down version of common::sense
455	use strict;	1161	sub common_sense {
		1162	# from common:.sense 3.3
		1163	${^WARNING_BITS} ^= ${^WARNING_BITS} ^ "\x3c\x3f\x33\x00\x0f\xf3\x0f\xc0\xf0\xfc\x33\x00";
		1164	# use strict vars subs - NO UTF-8, as Util.pm doesn't like this atm. (uts46data.pl)
		1165	$^H |= 0x00000600;
		1166	}
456		1167
		1168	BEGIN { AnyEvent::common_sense }
		1169
457	use Carp;	1170	use Carp ();
458		1171
459	our $VERSION = '3.3';	1172	our $VERSION = '5.29';
460	our $MODEL;	1173	our $MODEL;
461		1174
462	our $AUTOLOAD;	1175	our $AUTOLOAD;
463	our @ISA;	1176	our @ISA;
464		1177
465	our $verbose = $ENV{PERL_ANYEVENT_VERBOSE}*1;
466
467	our @REGISTRY;	1178	our @REGISTRY;
468		1179
		1180	our $VERBOSE;
		1181
		1182	BEGIN {
		1183	require "AnyEvent/constants.pl";
		1184
		1185	eval "sub TAINT (){" . (${^TAINT}*1) . "}";
		1186
		1187	delete @ENV{grep /^PERL_ANYEVENT_/, keys %ENV}
		1188	if ${^TAINT};
		1189
		1190	$VERBOSE = $ENV{PERL_ANYEVENT_VERBOSE}*1;
		1191
		1192	}
		1193
		1194	our $MAX_SIGNAL_LATENCY = 10;
		1195
		1196	our %PROTOCOL; # (ipv4|ipv6) => (1|2), higher numbers are preferred
		1197
		1198	{
		1199	my $idx;
		1200	$PROTOCOL{$_} = ++$idx
		1201	for reverse split /\s*,\s*/,
		1202	$ENV{PERL_ANYEVENT_PROTOCOLS} || "ipv4,ipv6";
		1203	}
		1204
469	my @models = (	1205	my @models = (
470	[Coro::EV:: => AnyEvent::Impl::CoroEV::],
471	[Coro::Event:: => AnyEvent::Impl::CoroEvent::],
472	[EV:: => AnyEvent::Impl::EV::],	1206	[EV:: => AnyEvent::Impl::EV:: , 1],
		1207	[AnyEvent::Impl::Perl:: => AnyEvent::Impl::Perl:: , 1],
		1208	# everything below here will not (normally) be autoprobed
		1209	# as the pureperl backend should work everywhere
		1210	# and is usually faster
473	[Event:: => AnyEvent::Impl::Event::],	1211	[Event:: => AnyEvent::Impl::Event::, 1],
474	[Glib:: => AnyEvent::Impl::Glib::],	1212	[Glib:: => AnyEvent::Impl::Glib:: , 1], # becomes extremely slow with many watchers
		1213	[Event::Lib:: => AnyEvent::Impl::EventLib::], # too buggy
		1214	[Irssi:: => AnyEvent::Impl::Irssi::], # Irssi has a bogus "Event" package
475	[Tk:: => AnyEvent::Impl::Tk::],	1215	[Tk:: => AnyEvent::Impl::Tk::], # crashes with many handles
		1216	[Qt:: => AnyEvent::Impl::Qt::], # requires special main program
		1217	[POE::Kernel:: => AnyEvent::Impl::POE::], # lasciate ogni speranza
476	[Wx:: => AnyEvent::Impl::POE::],	1218	[Wx:: => AnyEvent::Impl::POE::],
477	[Prima:: => AnyEvent::Impl::POE::],	1219	[Prima:: => AnyEvent::Impl::POE::],
478	[AnyEvent::Impl::Perl:: => AnyEvent::Impl::Perl::],	1220	# IO::Async is just too broken - we would need workarounds for its
479	# everything below here will not be autoprobed as the pureperl backend should work everywhere	1221	# byzantine signal and broken child handling, among others.
480	[Event::Lib:: => AnyEvent::Impl::EventLib::], # too buggy	1222	# IO::Async is rather hard to detect, as it doesn't have any
		1223	# obvious default class.
481	[Qt:: => AnyEvent::Impl::Qt::], # requires special main program	1224	[IO::Async:: => AnyEvent::Impl::IOAsync::], # requires special main program
482	[POE::Kernel:: => AnyEvent::Impl::POE::], # lasciate ogni speranza	1225	[IO::Async::Loop:: => AnyEvent::Impl::IOAsync::], # requires special main program
		1226	[IO::Async::Notifier:: => AnyEvent::Impl::IOAsync::], # requires special main program
		1227	[AnyEvent::Impl::IOAsync:: => AnyEvent::Impl::IOAsync::], # requires special main program
483	);	1228	);
484		1229
485	our %method = map +($_ => 1), qw(io timer signal child condvar broadcast wait one_event DESTROY);	1230	our %method = map +($_ => 1),
		1231	qw(io timer time now now_update signal child idle condvar one_event DESTROY);
		1232
		1233	our @post_detect;
		1234
		1235	sub post_detect(&) {
		1236	my ($cb) = @_;
		1237
		1238	push @post_detect, $cb;
		1239
		1240	defined wantarray
		1241	? bless \$cb, "AnyEvent::Util::postdetect"
		1242	: ()
		1243	}
		1244
		1245	sub AnyEvent::Util::postdetect::DESTROY {
		1246	@post_detect = grep $_ != ${$_[0]}, @post_detect;
		1247	}
486		1248
487	sub detect() {	1249	sub detect() {
		1250	# free some memory
		1251	*detect = sub () { $MODEL };
		1252
		1253	local $!; # for good measure
		1254	local $SIG{__DIE__};
		1255
		1256	if ($ENV{PERL_ANYEVENT_MODEL} =~ /^([a-zA-Z]+)$/) {
		1257	my $model = "AnyEvent::Impl::$1";
		1258	if (eval "require $model") {
		1259	$MODEL = $model;
		1260	warn "AnyEvent: loaded model '$model' (forced by \$ENV{PERL_ANYEVENT_MODEL}), using it.\n" if $VERBOSE >= 2;
		1261	} else {
		1262	warn "AnyEvent: unable to load model '$model' (from \$ENV{PERL_ANYEVENT_MODEL}):\n$@" if $VERBOSE;
		1263	}
		1264	}
		1265
		1266	# check for already loaded models
488	unless ($MODEL) {	1267	unless ($MODEL) {
489	no strict 'refs';	1268	for (@REGISTRY, @models) {
490		1269	my ($package, $model) = @$_;
491	if ($ENV{PERL_ANYEVENT_MODEL} =~ /^([a-zA-Z]+)$/) {	1270	if (${"$package\::VERSION"} > 0) {
492	my $model = "AnyEvent::Impl::$1";
493	if (eval "require $model") {	1271	if (eval "require $model") {
494	$MODEL = $model;	1272	$MODEL = $model;
495	warn "AnyEvent: loaded model '$model' (forced by \$PERL_ANYEVENT_MODEL), using it.\n" if $verbose > 1;	1273	warn "AnyEvent: autodetected model '$model', using it.\n" if $VERBOSE >= 2;
496	} else {	1274	last;
497	warn "AnyEvent: unable to load model '$model' (from \$PERL_ANYEVENT_MODEL):\n$@" if $verbose;	1275	}
498	}	1276	}
499	}	1277	}
500		1278
501	# check for already loaded models
502	unless ($MODEL) {	1279	unless ($MODEL) {
		1280	# try to autoload a model
503	for (@REGISTRY, @models) {	1281	for (@REGISTRY, @models) {
504	my ($package, $model) = @$_;	1282	my ($package, $model, $autoload) = @$_;
		1283	if (
		1284	$autoload
		1285	and eval "require $package"
505	if (${"$package\::VERSION"} > 0) {	1286	and ${"$package\::VERSION"} > 0
506	if (eval "require $model") {	1287	and eval "require $model"
		1288	) {
507	$MODEL = $model;	1289	$MODEL = $model;
508	warn "AnyEvent: autodetected model '$model', using it.\n" if $verbose > 1;	1290	warn "AnyEvent: autoloaded model '$model', using it.\n" if $VERBOSE >= 2;
509	last;	1291	last;
510	}
511	}	1292	}
512	}	1293	}
513		1294
514	unless ($MODEL) {
515	# try to load a model
516
517	for (@REGISTRY, @models) {
518	my ($package, $model) = @$_;
519	if (eval "require $package"
520	and ${"$package\::VERSION"} > 0
521	and eval "require $model") {
522	$MODEL = $model;
523	warn "AnyEvent: autoprobed model '$model', using it.\n" if $verbose > 1;
524	last;
525	}
526	}
527
528	$MODEL	1295	$MODEL
529	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.";	1296	or die "AnyEvent: backend autodetection failed - did you properly install AnyEvent?\n";
530	}
531	}	1297	}
532
533	unshift @ISA, $MODEL;
534	push @{"$MODEL\::ISA"}, "AnyEvent::Base";
535	}	1298	}
		1299
		1300	@models = (); # free probe data
		1301
		1302	push @{"$MODEL\::ISA"}, "AnyEvent::Base";
		1303	unshift @ISA, $MODEL;
		1304
		1305	# now nuke some methods that are overridden by the backend.
		1306	# SUPER is not allowed.
		1307	for (qw(time signal child idle)) {
		1308	undef &{"AnyEvent::Base::$_"}
		1309	if defined &{"$MODEL\::$_"};
		1310	}
		1311
		1312	if ($ENV{PERL_ANYEVENT_STRICT}) {
		1313	eval { require AnyEvent::Strict };
		1314	warn "AnyEvent: cannot load AnyEvent::Strict: $@"
		1315	if $@ && $VERBOSE;
		1316	}
		1317
		1318	(shift @post_detect)->() while @post_detect;
		1319
		1320	*post_detect = sub(&) {
		1321	shift->();
		1322
		1323	undef
		1324	};
536		1325
537	$MODEL	1326	$MODEL
538	}	1327	}
539		1328
540	sub AUTOLOAD {	1329	sub AUTOLOAD {
541	(my $func = $AUTOLOAD) =~ s/.*://;	1330	(my $func = $AUTOLOAD) =~ s/.*://;
542		1331
543	$method{$func}	1332	$method{$func}
544	or croak "$func: not a valid method for AnyEvent objects";	1333	or Carp::croak "$func: not a valid AnyEvent class method";
545		1334
546	detect unless $MODEL;	1335	detect;
547		1336
548	my $class = shift;	1337	my $class = shift;
549	$class->$func (@_);	1338	$class->$func (@_);
550	}	1339	}
551		1340
		1341	# utility function to dup a filehandle. this is used by many backends
		1342	# to support binding more than one watcher per filehandle (they usually
		1343	# allow only one watcher per fd, so we dup it to get a different one).
		1344	sub _dupfh($$;$$) {
		1345	my ($poll, $fh, $r, $w) = @_;
		1346
		1347	# cygwin requires the fh mode to be matching, unix doesn't
		1348	my ($rw, $mode) = $poll eq "r" ? ($r, "<&") : ($w, ">&");
		1349
		1350	open my $fh2, $mode, $fh
		1351	or die "AnyEvent->io: cannot dup() filehandle in mode '$poll': $!,";
		1352
		1353	# we assume CLOEXEC is already set by perl in all important cases
		1354
		1355	($fh2, $rw)
		1356	}
		1357
		1358	=head1 SIMPLIFIED AE API
		1359
		1360	Starting with version 5.0, AnyEvent officially supports a second, much
		1361	simpler, API that is designed to reduce the calling, typing and memory
		1362	overhead by using function call syntax and a fixed number of parameters.
		1363
		1364	See the L<AE> manpage for details.
		1365
		1366	=cut
		1367
		1368	package AE;
		1369
		1370	our $VERSION = $AnyEvent::VERSION;
		1371
		1372	# fall back to the main API by default - backends and AnyEvent::Base
		1373	# implementations can overwrite these.
		1374
		1375	sub io($$$) {
		1376	AnyEvent->io (fh => $_[0], poll => $_[1] ? "w" : "r", cb => $_[2])
		1377	}
		1378
		1379	sub timer($$$) {
		1380	AnyEvent->timer (after => $_[0], interval => $_[1], cb => $_[2])
		1381	}
		1382
		1383	sub signal($$) {
		1384	AnyEvent->signal (signal => $_[0], cb => $_[1])
		1385	}
		1386
		1387	sub child($$) {
		1388	AnyEvent->child (pid => $_[0], cb => $_[1])
		1389	}
		1390
		1391	sub idle($) {
		1392	AnyEvent->idle (cb => $_[0])
		1393	}
		1394
		1395	sub cv(;&) {
		1396	AnyEvent->condvar (@_ ? (cb => $_[0]) : ())
		1397	}
		1398
		1399	sub now() {
		1400	AnyEvent->now
		1401	}
		1402
		1403	sub now_update() {
		1404	AnyEvent->now_update
		1405	}
		1406
		1407	sub time() {
		1408	AnyEvent->time
		1409	}
		1410
552	package AnyEvent::Base;	1411	package AnyEvent::Base;
553		1412
		1413	# default implementations for many methods
		1414
		1415	sub time {
		1416	eval q{ # poor man's autoloading {}
		1417	# probe for availability of Time::HiRes
		1418	if (eval "use Time::HiRes (); Time::HiRes::time (); 1") {
		1419	warn "AnyEvent: using Time::HiRes for sub-second timing accuracy.\n" if $VERBOSE >= 8;
		1420	*AE::time = \&Time::HiRes::time;
		1421	# if (eval "use POSIX (); (POSIX::times())...
		1422	} else {
		1423	warn "AnyEvent: using built-in time(), WARNING, no sub-second resolution!\n" if $VERBOSE;
		1424	*AE::time = sub (){ time }; # epic fail
		1425	}
		1426
		1427	*time = sub { AE::time }; # different prototypes
		1428	};
		1429	die if $@;
		1430
		1431	&time
		1432	}
		1433
		1434	*now = \&time;
		1435
		1436	sub now_update { }
		1437
554	# default implementation for ->condvar, ->wait, ->broadcast	1438	# default implementation for ->condvar
555		1439
556	sub condvar {	1440	sub condvar {
557	bless \my $flag, "AnyEvent::Base::CondVar"	1441	eval q{ # poor man's autoloading {}
558	}	1442	*condvar = sub {
		1443	bless { @_ == 3 ? (_ae_cb => $_[2]) : () }, "AnyEvent::CondVar"
		1444	};
559		1445
560	sub AnyEvent::Base::CondVar::broadcast {	1446	*AE::cv = sub (;&) {
561	${$_[0]}++;	1447	bless { @_ ? (_ae_cb => shift) : () }, "AnyEvent::CondVar"
562	}	1448	};
		1449	};
		1450	die if $@;
563		1451
564	sub AnyEvent::Base::CondVar::wait {	1452	&condvar
565	AnyEvent->one_event while !${$_[0]};
566	}	1453	}
567		1454
568	# default implementation for ->signal	1455	# default implementation for ->signal
569		1456
570	our %SIG_CB;	1457	our $HAVE_ASYNC_INTERRUPT;
		1458
		1459	sub _have_async_interrupt() {
		1460	$HAVE_ASYNC_INTERRUPT = 1*(!$ENV{PERL_ANYEVENT_AVOID_ASYNC_INTERRUPT}
		1461	&& eval "use Async::Interrupt 1.02 (); 1")
		1462	unless defined $HAVE_ASYNC_INTERRUPT;
		1463
		1464	$HAVE_ASYNC_INTERRUPT
		1465	}
		1466
		1467	our ($SIGPIPE_R, $SIGPIPE_W, %SIG_CB, %SIG_EV, $SIG_IO);
		1468	our (%SIG_ASY, %SIG_ASY_W);
		1469	our ($SIG_COUNT, $SIG_TW);
		1470
		1471	# install a dummy wakeup watcher to reduce signal catching latency
		1472	# used by Impls
		1473	sub _sig_add() {
		1474	unless ($SIG_COUNT++) {
		1475	# try to align timer on a full-second boundary, if possible
		1476	my $NOW = AE::now;
		1477
		1478	$SIG_TW = AE::timer
		1479	$MAX_SIGNAL_LATENCY - ($NOW - int $NOW),
		1480	$MAX_SIGNAL_LATENCY,
		1481	sub { } # just for the PERL_ASYNC_CHECK
		1482	;
		1483	}
		1484	}
		1485
		1486	sub _sig_del {
		1487	undef $SIG_TW
		1488	unless --$SIG_COUNT;
		1489	}
		1490
		1491	our $_sig_name_init; $_sig_name_init = sub {
		1492	eval q{ # poor man's autoloading {}
		1493	undef $_sig_name_init;
		1494
		1495	if (_have_async_interrupt) {
		1496	*sig2num = \&Async::Interrupt::sig2num;
		1497	*sig2name = \&Async::Interrupt::sig2name;
		1498	} else {
		1499	require Config;
		1500
		1501	my %signame2num;
		1502	@signame2num{ split ' ', $Config::Config{sig_name} }
		1503	= split ' ', $Config::Config{sig_num};
		1504
		1505	my @signum2name;
		1506	@signum2name[values %signame2num] = keys %signame2num;
		1507
		1508	*sig2num = sub($) {
		1509	$_[0] > 0 ? shift : $signame2num{+shift}
		1510	};
		1511	*sig2name = sub ($) {
		1512	$_[0] > 0 ? $signum2name[+shift] : shift
		1513	};
		1514	}
		1515	};
		1516	die if $@;
		1517	};
		1518
		1519	sub sig2num ($) { &$_sig_name_init; &sig2num }
		1520	sub sig2name($) { &$_sig_name_init; &sig2name }
571		1521
572	sub signal {	1522	sub signal {
		1523	eval q{ # poor man's autoloading {}
		1524	# probe for availability of Async::Interrupt
		1525	if (_have_async_interrupt) {
		1526	warn "AnyEvent: using Async::Interrupt for race-free signal handling.\n" if $VERBOSE >= 8;
		1527
		1528	$SIGPIPE_R = new Async::Interrupt::EventPipe;
		1529	$SIG_IO = AE::io $SIGPIPE_R->fileno, 0, \&_signal_exec;
		1530
		1531	} else {
		1532	warn "AnyEvent: using emulated perl signal handling with latency timer.\n" if $VERBOSE >= 8;
		1533
		1534	if (AnyEvent::WIN32) {
		1535	require AnyEvent::Util;
		1536
		1537	($SIGPIPE_R, $SIGPIPE_W) = AnyEvent::Util::portable_pipe ();
		1538	AnyEvent::Util::fh_nonblocking ($SIGPIPE_R, 1) if $SIGPIPE_R;
		1539	AnyEvent::Util::fh_nonblocking ($SIGPIPE_W, 1) if $SIGPIPE_W; # just in case
		1540	} else {
		1541	pipe $SIGPIPE_R, $SIGPIPE_W;
		1542	fcntl $SIGPIPE_R, AnyEvent::F_SETFL, AnyEvent::O_NONBLOCK if $SIGPIPE_R;
		1543	fcntl $SIGPIPE_W, AnyEvent::F_SETFL, AnyEvent::O_NONBLOCK if $SIGPIPE_W; # just in case
		1544
		1545	# not strictly required, as $^F is normally 2, but let's make sure...
		1546	fcntl $SIGPIPE_R, AnyEvent::F_SETFD, AnyEvent::FD_CLOEXEC;
		1547	fcntl $SIGPIPE_W, AnyEvent::F_SETFD, AnyEvent::FD_CLOEXEC;
		1548	}
		1549
		1550	$SIGPIPE_R
		1551	or Carp::croak "AnyEvent: unable to create a signal reporting pipe: $!\n";
		1552
		1553	$SIG_IO = AE::io $SIGPIPE_R, 0, \&_signal_exec;
		1554	}
		1555
		1556	*signal = $HAVE_ASYNC_INTERRUPT
		1557	? sub {
573	my (undef, %arg) = @_;	1558	my (undef, %arg) = @_;
574		1559
		1560	# async::interrupt
575	my $signal = uc $arg{signal}	1561	my $signal = sig2num $arg{signal};
576	or Carp::croak "required option 'signal' is missing";
577
578	$SIG_CB{$signal}{$arg{cb}} = $arg{cb};	1562	$SIG_CB{$signal}{$arg{cb}} = $arg{cb};
		1563
		1564	$SIG_ASY{$signal} ||= new Async::Interrupt
		1565	cb => sub { undef $SIG_EV{$signal} },
		1566	signal => $signal,
		1567	pipe => [$SIGPIPE_R->filenos],
		1568	pipe_autodrain => 0,
		1569	;
		1570
		1571	bless [$signal, $arg{cb}], "AnyEvent::Base::signal"
		1572	}
		1573	: sub {
		1574	my (undef, %arg) = @_;
		1575
		1576	# pure perl
		1577	my $signal = sig2name $arg{signal};
		1578	$SIG_CB{$signal}{$arg{cb}} = $arg{cb};
		1579
579	$SIG{$signal} ||= sub {	1580	$SIG{$signal} ||= sub {
		1581	local $!;
		1582	syswrite $SIGPIPE_W, "\x00", 1 unless %SIG_EV;
		1583	undef $SIG_EV{$signal};
		1584	};
		1585
		1586	# can't do signal processing without introducing races in pure perl,
		1587	# so limit the signal latency.
		1588	_sig_add;
		1589
		1590	bless [$signal, $arg{cb}], "AnyEvent::Base::signal"
		1591	}
		1592	;
		1593
		1594	*AnyEvent::Base::signal::DESTROY = sub {
		1595	my ($signal, $cb) = @{$_[0]};
		1596
		1597	_sig_del;
		1598
		1599	delete $SIG_CB{$signal}{$cb};
		1600
		1601	$HAVE_ASYNC_INTERRUPT
		1602	? delete $SIG_ASY{$signal}
		1603	: # delete doesn't work with older perls - they then
		1604	# print weird messages, or just unconditionally exit
		1605	# instead of getting the default action.
		1606	undef $SIG{$signal}
		1607	unless keys %{ $SIG_CB{$signal} };
		1608	};
		1609
		1610	*_signal_exec = sub {
		1611	$HAVE_ASYNC_INTERRUPT
		1612	? $SIGPIPE_R->drain
		1613	: sysread $SIGPIPE_R, (my $dummy), 9;
		1614
		1615	while (%SIG_EV) {
		1616	for (keys %SIG_EV) {
		1617	delete $SIG_EV{$_};
580	$_->() for values %{ $SIG_CB{$signal} || {} };	1618	$_->() for values %{ $SIG_CB{$_} || {} };
		1619	}
		1620	}
		1621	};
581	};	1622	};
		1623	die if $@;
582		1624
583	bless [$signal, $arg{cb}], "AnyEvent::Base::Signal"	1625	&signal
584	}
585
586	sub AnyEvent::Base::Signal::DESTROY {
587	my ($signal, $cb) = @{$_[0]};
588
589	delete $SIG_CB{$signal}{$cb};
590
591	$SIG{$signal} = 'DEFAULT' unless keys %{ $SIG_CB{$signal} };
592	}	1626	}
593		1627
594	# default implementation for ->child	1628	# default implementation for ->child
595		1629
596	our %PID_CB;	1630	our %PID_CB;
597	our $CHLD_W;	1631	our $CHLD_W;
598	our $CHLD_DELAY_W;	1632	our $CHLD_DELAY_W;
599	our $PID_IDLE;
600	our $WNOHANG;	1633	our $WNOHANG;
601		1634
602	sub _child_wait {	1635	# used by many Impl's
603	while (0 < (my $pid = waitpid -1, $WNOHANG)) {	1636	sub _emit_childstatus($$) {
		1637	my (undef, $rpid, $rstatus) = @_;
		1638
		1639	$_->($rpid, $rstatus)
604	$_->($pid, $?) for (values %{ $PID_CB{$pid} || {} }),	1640	for values %{ $PID_CB{$rpid} || {} },
605	(values %{ $PID_CB{0} || {} });	1641	values %{ $PID_CB{0} || {} };
606	}
607
608	undef $PID_IDLE;
609	}
610
611	sub _sigchld {
612	# make sure we deliver these changes "synchronous" with the event loop.
613	$CHLD_DELAY_W ||= AnyEvent->timer (after => 0, cb => sub {
614	undef $CHLD_DELAY_W;
615	&_child_wait;
616	});
617	}	1642	}
618		1643
619	sub child {	1644	sub child {
		1645	eval q{ # poor man's autoloading {}
		1646	*_sigchld = sub {
		1647	my $pid;
		1648
		1649	AnyEvent->_emit_childstatus ($pid, $?)
		1650	while ($pid = waitpid -1, $WNOHANG) > 0;
		1651	};
		1652
		1653	*child = sub {
620	my (undef, %arg) = @_;	1654	my (undef, %arg) = @_;
621		1655
622	defined (my $pid = $arg{pid} + 0)	1656	defined (my $pid = $arg{pid} + 0)
623	or Carp::croak "required option 'pid' is missing";	1657	or Carp::croak "required option 'pid' is missing";
624		1658
625	$PID_CB{$pid}{$arg{cb}} = $arg{cb};	1659	$PID_CB{$pid}{$arg{cb}} = $arg{cb};
626		1660
627	unless ($WNOHANG) {	1661	# WNOHANG is almost cetrainly 1 everywhere
628	$WNOHANG = eval { require POSIX; &POSIX::WNOHANG } || 1;	1662	$WNOHANG ||= $^O =~ /^(?:openbsd|netbsd|linux|freebsd|cygwin|MSWin32)$/
629	}	1663	? 1
		1664	: eval { local $SIG{__DIE__}; require POSIX; &POSIX::WNOHANG } || 1;
630		1665
631	unless ($CHLD_W) {	1666	unless ($CHLD_W) {
632	$CHLD_W = AnyEvent->signal (signal => 'CHLD', cb => \&_sigchld);	1667	$CHLD_W = AE::signal CHLD => \&_sigchld;
633	# child could be a zombie already, so make at least one round	1668	# child could be a zombie already, so make at least one round
634	&_sigchld;	1669	&_sigchld;
635	}	1670	}
636		1671
637	bless [$pid, $arg{cb}], "AnyEvent::Base::Child"	1672	bless [$pid, $arg{cb}], "AnyEvent::Base::child"
638	}	1673	};
639		1674
640	sub AnyEvent::Base::Child::DESTROY {	1675	*AnyEvent::Base::child::DESTROY = sub {
641	my ($pid, $cb) = @{$_[0]};	1676	my ($pid, $cb) = @{$_[0]};
642		1677
643	delete $PID_CB{$pid}{$cb};	1678	delete $PID_CB{$pid}{$cb};
644	delete $PID_CB{$pid} unless keys %{ $PID_CB{$pid} };	1679	delete $PID_CB{$pid} unless keys %{ $PID_CB{$pid} };
645		1680
646	undef $CHLD_W unless keys %PID_CB;	1681	undef $CHLD_W unless keys %PID_CB;
		1682	};
		1683	};
		1684	die if $@;
		1685
		1686	&child
647	}	1687	}
		1688
		1689	# idle emulation is done by simply using a timer, regardless
		1690	# of whether the process is idle or not, and not letting
		1691	# the callback use more than 50% of the time.
		1692	sub idle {
		1693	eval q{ # poor man's autoloading {}
		1694	*idle = sub {
		1695	my (undef, %arg) = @_;
		1696
		1697	my ($cb, $w, $rcb) = $arg{cb};
		1698
		1699	$rcb = sub {
		1700	if ($cb) {
		1701	$w = _time;
		1702	&$cb;
		1703	$w = _time - $w;
		1704
		1705	# never use more then 50% of the time for the idle watcher,
		1706	# within some limits
		1707	$w = 0.0001 if $w < 0.0001;
		1708	$w = 5 if $w > 5;
		1709
		1710	$w = AE::timer $w, 0, $rcb;
		1711	} else {
		1712	# clean up...
		1713	undef $w;
		1714	undef $rcb;
		1715	}
		1716	};
		1717
		1718	$w = AE::timer 0.05, 0, $rcb;
		1719
		1720	bless \\$cb, "AnyEvent::Base::idle"
		1721	};
		1722
		1723	*AnyEvent::Base::idle::DESTROY = sub {
		1724	undef $${$_[0]};
		1725	};
		1726	};
		1727	die if $@;
		1728
		1729	&idle
		1730	}
		1731
		1732	package AnyEvent::CondVar;
		1733
		1734	our @ISA = AnyEvent::CondVar::Base::;
		1735
		1736	# only to be used for subclassing
		1737	sub new {
		1738	my $class = shift;
		1739	bless AnyEvent->condvar (@_), $class
		1740	}
		1741
		1742	package AnyEvent::CondVar::Base;
		1743
		1744	#use overload
		1745	# '&{}' => sub { my $self = shift; sub { $self->send (@_) } },
		1746	# fallback => 1;
		1747
		1748	# save 300+ kilobytes by dirtily hardcoding overloading
		1749	${"AnyEvent::CondVar::Base::OVERLOAD"}{dummy}++; # Register with magic by touching.
		1750	*{'AnyEvent::CondVar::Base::()'} = sub { }; # "Make it findable via fetchmethod."
		1751	*{'AnyEvent::CondVar::Base::(&{}'} = sub { my $self = shift; sub { $self->send (@_) } }; # &{}
		1752	${'AnyEvent::CondVar::Base::()'} = 1; # fallback
		1753
		1754	our $WAITING;
		1755
		1756	sub _send {
		1757	# nop
		1758	}
		1759
		1760	sub send {
		1761	my $cv = shift;
		1762	$cv->{_ae_sent} = [@_];
		1763	(delete $cv->{_ae_cb})->($cv) if $cv->{_ae_cb};
		1764	$cv->_send;
		1765	}
		1766
		1767	sub croak {
		1768	$_[0]{_ae_croak} = $_[1];
		1769	$_[0]->send;
		1770	}
		1771
		1772	sub ready {
		1773	$_[0]{_ae_sent}
		1774	}
		1775
		1776	sub _wait {
		1777	$WAITING
		1778	and !$_[0]{_ae_sent}
		1779	and Carp::croak "AnyEvent::CondVar: recursive blocking wait detected";
		1780
		1781	local $WAITING = 1;
		1782	AnyEvent->one_event while !$_[0]{_ae_sent};
		1783	}
		1784
		1785	sub recv {
		1786	$_[0]->_wait;
		1787
		1788	Carp::croak $_[0]{_ae_croak} if $_[0]{_ae_croak};
		1789	wantarray ? @{ $_[0]{_ae_sent} } : $_[0]{_ae_sent}[0]
		1790	}
		1791
		1792	sub cb {
		1793	my $cv = shift;
		1794
		1795	@_
		1796	and $cv->{_ae_cb} = shift
		1797	and $cv->{_ae_sent}
		1798	and (delete $cv->{_ae_cb})->($cv);
		1799
		1800	$cv->{_ae_cb}
		1801	}
		1802
		1803	sub begin {
		1804	++$_[0]{_ae_counter};
		1805	$_[0]{_ae_end_cb} = $_[1] if @_ > 1;
		1806	}
		1807
		1808	sub end {
		1809	return if --$_[0]{_ae_counter};
		1810	&{ $_[0]{_ae_end_cb} || sub { $_[0]->send } };
		1811	}
		1812
		1813	# undocumented/compatibility with pre-3.4
		1814	*broadcast = \&send;
		1815	*wait = \&_wait;
		1816
		1817	=head1 ERROR AND EXCEPTION HANDLING
		1818
		1819	In general, AnyEvent does not do any error handling - it relies on the
		1820	caller to do that if required. The L<AnyEvent::Strict> module (see also
		1821	the C<PERL_ANYEVENT_STRICT> environment variable, below) provides strict
		1822	checking of all AnyEvent methods, however, which is highly useful during
		1823	development.
		1824
		1825	As for exception handling (i.e. runtime errors and exceptions thrown while
		1826	executing a callback), this is not only highly event-loop specific, but
		1827	also not in any way wrapped by this module, as this is the job of the main
		1828	program.
		1829
		1830	The pure perl event loop simply re-throws the exception (usually
		1831	within C<< condvar->recv >>), the L<Event> and L<EV> modules call C<<
		1832	$Event/EV::DIED->() >>, L<Glib> uses C<< install_exception_handler >> and
		1833	so on.
		1834
		1835	=head1 ENVIRONMENT VARIABLES
		1836
		1837	The following environment variables are used by this module or its
		1838	submodules.
		1839
		1840	Note that AnyEvent will remove I<all> environment variables starting with
		1841	C<PERL_ANYEVENT_> from C<%ENV> when it is loaded while taint mode is
		1842	enabled.
		1843
		1844	=over 4
		1845
		1846	=item C<PERL_ANYEVENT_VERBOSE>
		1847
		1848	By default, AnyEvent will be completely silent except in fatal
		1849	conditions. You can set this environment variable to make AnyEvent more
		1850	talkative.
		1851
		1852	When set to C<1> or higher, causes AnyEvent to warn about unexpected
		1853	conditions, such as not being able to load the event model specified by
		1854	C<PERL_ANYEVENT_MODEL>.
		1855
		1856	When set to C<2> or higher, cause AnyEvent to report to STDERR which event
		1857	model it chooses.
		1858
		1859	When set to C<8> or higher, then AnyEvent will report extra information on
		1860	which optional modules it loads and how it implements certain features.
		1861
		1862	=item C<PERL_ANYEVENT_STRICT>
		1863
		1864	AnyEvent does not do much argument checking by default, as thorough
		1865	argument checking is very costly. Setting this variable to a true value
		1866	will cause AnyEvent to load C<AnyEvent::Strict> and then to thoroughly
		1867	check the arguments passed to most method calls. If it finds any problems,
		1868	it will croak.
		1869
		1870	In other words, enables "strict" mode.
		1871
		1872	Unlike C<use strict> (or its modern cousin, C<< use L<common::sense>
		1873	>>, it is definitely recommended to keep it off in production. Keeping
		1874	C<PERL_ANYEVENT_STRICT=1> in your environment while developing programs
		1875	can be very useful, however.
		1876
		1877	=item C<PERL_ANYEVENT_MODEL>
		1878
		1879	This can be used to specify the event model to be used by AnyEvent, before
		1880	auto detection and -probing kicks in. It must be a string consisting
		1881	entirely of ASCII letters. The string C<AnyEvent::Impl::> gets prepended
		1882	and the resulting module name is loaded and if the load was successful,
		1883	used as event model. If it fails to load AnyEvent will proceed with
		1884	auto detection and -probing.
		1885
		1886	This functionality might change in future versions.
		1887
		1888	For example, to force the pure perl model (L<AnyEvent::Impl::Perl>) you
		1889	could start your program like this:
		1890
		1891	PERL_ANYEVENT_MODEL=Perl perl ...
		1892
		1893	=item C<PERL_ANYEVENT_PROTOCOLS>
		1894
		1895	Used by both L<AnyEvent::DNS> and L<AnyEvent::Socket> to determine preferences
		1896	for IPv4 or IPv6. The default is unspecified (and might change, or be the result
		1897	of auto probing).
		1898
		1899	Must be set to a comma-separated list of protocols or address families,
		1900	current supported: C<ipv4> and C<ipv6>. Only protocols mentioned will be
		1901	used, and preference will be given to protocols mentioned earlier in the
		1902	list.
		1903
		1904	This variable can effectively be used for denial-of-service attacks
		1905	against local programs (e.g. when setuid), although the impact is likely
		1906	small, as the program has to handle conenction and other failures anyways.
		1907
		1908	Examples: C<PERL_ANYEVENT_PROTOCOLS=ipv4,ipv6> - prefer IPv4 over IPv6,
		1909	but support both and try to use both. C<PERL_ANYEVENT_PROTOCOLS=ipv4>
		1910	- only support IPv4, never try to resolve or contact IPv6
		1911	addresses. C<PERL_ANYEVENT_PROTOCOLS=ipv6,ipv4> support either IPv4 or
		1912	IPv6, but prefer IPv6 over IPv4.
		1913
		1914	=item C<PERL_ANYEVENT_EDNS0>
		1915
		1916	Used by L<AnyEvent::DNS> to decide whether to use the EDNS0 extension
		1917	for DNS. This extension is generally useful to reduce DNS traffic, but
		1918	some (broken) firewalls drop such DNS packets, which is why it is off by
		1919	default.
		1920
		1921	Setting this variable to C<1> will cause L<AnyEvent::DNS> to announce
		1922	EDNS0 in its DNS requests.
		1923
		1924	=item C<PERL_ANYEVENT_MAX_FORKS>
		1925
		1926	The maximum number of child processes that C<AnyEvent::Util::fork_call>
		1927	will create in parallel.
		1928
		1929	=item C<PERL_ANYEVENT_MAX_OUTSTANDING_DNS>
		1930
		1931	The default value for the C<max_outstanding> parameter for the default DNS
		1932	resolver - this is the maximum number of parallel DNS requests that are
		1933	sent to the DNS server.
		1934
		1935	=item C<PERL_ANYEVENT_RESOLV_CONF>
		1936
		1937	The file to use instead of F</etc/resolv.conf> (or OS-specific
		1938	configuration) in the default resolver. When set to the empty string, no
		1939	default config will be used.
		1940
		1941	=item C<PERL_ANYEVENT_CA_FILE>, C<PERL_ANYEVENT_CA_PATH>.
		1942
		1943	When neither C<ca_file> nor C<ca_path> was specified during
		1944	L<AnyEvent::TLS> context creation, and either of these environment
		1945	variables exist, they will be used to specify CA certificate locations
		1946	instead of a system-dependent default.
		1947
		1948	=item C<PERL_ANYEVENT_AVOID_GUARD> and C<PERL_ANYEVENT_AVOID_ASYNC_INTERRUPT>
		1949
		1950	When these are set to C<1>, then the respective modules are not
		1951	loaded. Mostly good for testing AnyEvent itself.
		1952
		1953	=back
648		1954
649	=head1 SUPPLYING YOUR OWN EVENT MODEL INTERFACE	1955	=head1 SUPPLYING YOUR OWN EVENT MODEL INTERFACE
650		1956
651	This is an advanced topic that you do not normally need to use AnyEvent in	1957	This is an advanced topic that you do not normally need to use AnyEvent in
652	a module. This section is only of use to event loop authors who want to	1958	a module. This section is only of use to event loop authors who want to
…		…
686		1992
687	I<rxvt-unicode> also cheats a bit by not providing blocking access to	1993	I<rxvt-unicode> also cheats a bit by not providing blocking access to
688	condition variables: code blocking while waiting for a condition will	1994	condition variables: code blocking while waiting for a condition will
689	C<die>. This still works with most modules/usages, and blocking calls must	1995	C<die>. This still works with most modules/usages, and blocking calls must
690	not be done in an interactive application, so it makes sense.	1996	not be done in an interactive application, so it makes sense.
691
692	=head1 ENVIRONMENT VARIABLES
693
694	The following environment variables are used by this module:
695
696	=over 4
697
698	=item C<PERL_ANYEVENT_VERBOSE>
699
700	By default, AnyEvent will be completely silent except in fatal
701	conditions. You can set this environment variable to make AnyEvent more
702	talkative.
703
704	When set to C<1> or higher, causes AnyEvent to warn about unexpected
705	conditions, such as not being able to load the event model specified by
706	C<PERL_ANYEVENT_MODEL>.
707
708	When set to C<2> or higher, cause AnyEvent to report to STDERR which event
709	model it chooses.
710
711	=item C<PERL_ANYEVENT_MODEL>
712
713	This can be used to specify the event model to be used by AnyEvent, before
714	autodetection and -probing kicks in. It must be a string consisting
715	entirely of ASCII letters. The string C<AnyEvent::Impl::> gets prepended
716	and the resulting module name is loaded and if the load was successful,
717	used as event model. If it fails to load AnyEvent will proceed with
718	autodetection and -probing.
719
720	This functionality might change in future versions.
721
722	For example, to force the pure perl model (L<AnyEvent::Impl::Perl>) you
723	could start your program like this:
724
725	PERL_ANYEVENT_MODEL=Perl perl ...
726
727	=back
728		1997
729	=head1 EXAMPLE PROGRAM	1998	=head1 EXAMPLE PROGRAM
730		1999
731	The following program uses an I/O watcher to read data from STDIN, a timer	2000	The following program uses an I/O watcher to read data from STDIN, a timer
732	to display a message once per second, and a condition variable to quit the	2001	to display a message once per second, and a condition variable to quit the
…		…
741	poll => 'r',	2010	poll => 'r',
742	cb => sub {	2011	cb => sub {
743	warn "io event <$_[0]>\n"; # will always output <r>	2012	warn "io event <$_[0]>\n"; # will always output <r>
744	chomp (my $input = <STDIN>); # read a line	2013	chomp (my $input = <STDIN>); # read a line
745	warn "read: $input\n"; # output what has been read	2014	warn "read: $input\n"; # output what has been read
746	$cv->broadcast if $input =~ /^q/i; # quit program if /^q/i	2015	$cv->send if $input =~ /^q/i; # quit program if /^q/i
747	},	2016	},
748	);	2017	);
749		2018
750	my $time_watcher; # can only be used once
751
752	sub new_timer {
753	$timer = AnyEvent->timer (after => 1, cb => sub {	2019	my $time_watcher = AnyEvent->timer (after => 1, interval => 1, cb => sub {
754	warn "timeout\n"; # print 'timeout' about every second	2020	warn "timeout\n"; # print 'timeout' at most every second
755	&new_timer; # and restart the time
756	});	2021	});
757	}
758		2022
759	new_timer; # create first timer
760
761	$cv->wait; # wait until user enters /^q/i	2023	$cv->recv; # wait until user enters /^q/i
762		2024
763	=head1 REAL-WORLD EXAMPLE	2025	=head1 REAL-WORLD EXAMPLE
764		2026
765	Consider the L<Net::FCP> module. It features (among others) the following	2027	Consider the L<Net::FCP> module. It features (among others) the following
766	API calls, which are to freenet what HTTP GET requests are to http:	2028	API calls, which are to freenet what HTTP GET requests are to http:
…		…
816	syswrite $txn->{fh}, $txn->{request}	2078	syswrite $txn->{fh}, $txn->{request}
817	or die "connection or write error";	2079	or die "connection or write error";
818	$txn->{w} = AnyEvent->io (fh => $txn->{fh}, poll => 'r', cb => sub { $txn->fh_ready_r });	2080	$txn->{w} = AnyEvent->io (fh => $txn->{fh}, poll => 'r', cb => sub { $txn->fh_ready_r });
819		2081
820	Again, C<fh_ready_r> waits till all data has arrived, and then stores the	2082	Again, C<fh_ready_r> waits till all data has arrived, and then stores the
821	result and signals any possible waiters that the request ahs finished:	2083	result and signals any possible waiters that the request has finished:
822		2084
823	sysread $txn->{fh}, $txn->{buf}, length $txn->{$buf};	2085	sysread $txn->{fh}, $txn->{buf}, length $txn->{$buf};
824		2086
825	if (end-of-file or data complete) {	2087	if (end-of-file or data complete) {
826	$txn->{result} = $txn->{buf};	2088	$txn->{result} = $txn->{buf};
827	$txn->{finished}->broadcast;	2089	$txn->{finished}->send;
828	$txb->{cb}->($txn) of $txn->{cb}; # also call callback	2090	$txb->{cb}->($txn) of $txn->{cb}; # also call callback
829	}	2091	}
830		2092
831	The C<result> method, finally, just waits for the finished signal (if the	2093	The C<result> method, finally, just waits for the finished signal (if the
832	request was already finished, it doesn't wait, of course, and returns the	2094	request was already finished, it doesn't wait, of course, and returns the
833	data:	2095	data:
834		2096
835	$txn->{finished}->wait;	2097	$txn->{finished}->recv;
836	return $txn->{result};	2098	return $txn->{result};
837		2099
838	The actual code goes further and collects all errors (C<die>s, exceptions)	2100	The actual code goes further and collects all errors (C<die>s, exceptions)
839	that occured during request processing. The C<result> method detects	2101	that occurred during request processing. The C<result> method detects
840	whether an exception as thrown (it is stored inside the $txn object)	2102	whether an exception as thrown (it is stored inside the $txn object)
841	and just throws the exception, which means connection errors and other	2103	and just throws the exception, which means connection errors and other
842	problems get reported tot he code that tries to use the result, not in a	2104	problems get reported to the code that tries to use the result, not in a
843	random callback.	2105	random callback.
844		2106
845	All of this enables the following usage styles:	2107	All of this enables the following usage styles:
846		2108
847	1. Blocking:	2109	1. Blocking:
…		…
875		2137
876	my $quit = AnyEvent->condvar;	2138	my $quit = AnyEvent->condvar;
877		2139
878	$fcp->txn_client_get ($url)->cb (sub {	2140	$fcp->txn_client_get ($url)->cb (sub {
879	...	2141	...
880	$quit->broadcast;	2142	$quit->send;
881	});	2143	});
882		2144
883	$quit->wait;	2145	$quit->recv;
884		2146
885		2147
886	=head1 BENCHMARK	2148	=head1 BENCHMARKS
887		2149
888	To give you an idea of the performance and overheads that AnyEvent adds	2150	To give you an idea of the performance and overheads that AnyEvent adds
889	over the event loops themselves (and to give you an impression of the	2151	over the event loops themselves and to give you an impression of the speed
890	speed of various event loops), here is a benchmark of various supported	2152	of various event loops I prepared some benchmarks.
891	event models natively and with anyevent. The benchmark creates a lot of	2153
892	timers (with a zero timeout) and I/O watchers (watching STDOUT, a pty, to	2154	=head2 BENCHMARKING ANYEVENT OVERHEAD
		2155
		2156	Here is a benchmark of various supported event models used natively and
		2157	through AnyEvent. The benchmark creates a lot of timers (with a zero
		2158	timeout) and I/O watchers (watching STDOUT, a pty, to become writable,
893	become writable, which it is), lets them fire exactly once and destroys	2159	which it is), lets them fire exactly once and destroys them again.
894	them again.
895		2160
896	Rewriting the benchmark to use many different sockets instead of using	2161	Source code for this benchmark is found as F<eg/bench> in the AnyEvent
897	the same filehandle for all I/O watchers results in a much longer runtime	2162	distribution. It uses the L<AE> interface, which makes a real difference
898	(socket creation is expensive), but qualitatively the same figures, so it	2163	for the EV and Perl backends only.
899	was not used.
900		2164
901	=head2 Explanation of the columns	2165	=head3 Explanation of the columns
902		2166
903	I<watcher> is the number of event watchers created/destroyed. Since	2167	I<watcher> is the number of event watchers created/destroyed. Since
904	different event models feature vastly different performances, each event	2168	different event models feature vastly different performances, each event
905	loop was given a number of watchers so that overall runtime is acceptable	2169	loop was given a number of watchers so that overall runtime is acceptable
906	and similar between tested event loop (and keep them from crashing): Glib	2170	and similar between tested event loop (and keep them from crashing): Glib
…		…
916	all watchers, to avoid adding memory overhead. That means closure creation	2180	all watchers, to avoid adding memory overhead. That means closure creation
917	and memory usage is not included in the figures.	2181	and memory usage is not included in the figures.
918		2182
919	I<invoke> is the time, in microseconds, used to invoke a simple	2183	I<invoke> is the time, in microseconds, used to invoke a simple
920	callback. The callback simply counts down a Perl variable and after it was	2184	callback. The callback simply counts down a Perl variable and after it was
921	invoked "watcher" times, it would C<< ->broadcast >> a condvar once to	2185	invoked "watcher" times, it would C<< ->send >> a condvar once to
922	signal the end of this phase.	2186	signal the end of this phase.
923		2187
924	I<destroy> is the time, in microseconds, that it takes to destroy a single	2188	I<destroy> is the time, in microseconds, that it takes to destroy a single
925	watcher.	2189	watcher.
926		2190
927	=head2 Results	2191	=head3 Results
928		2192
929	name watchers bytes create invoke destroy comment	2193	name watchers bytes create invoke destroy comment
930	EV/EV 400000 244 0.56 0.46 0.31 EV native interface	2194	EV/EV 100000 223 0.47 0.43 0.27 EV native interface
931	EV/Any 100000 244 2.50 0.46 0.29 EV + AnyEvent watchers	2195	EV/Any 100000 223 0.48 0.42 0.26 EV + AnyEvent watchers
932	CoroEV/Any 100000 244 2.49 0.44 0.29 coroutines + Coro::Signal	2196	Coro::EV/Any 100000 223 0.47 0.42 0.26 coroutines + Coro::Signal
933	Perl/Any 100000 513 4.92 0.87 1.12 pure perl implementation	2197	Perl/Any 100000 431 2.70 0.74 0.92 pure perl implementation
934	Event/Event 16000 516 31.88 31.30 0.85 Event native interface	2198	Event/Event 16000 516 31.16 31.84 0.82 Event native interface
935	Event/Any 16000 936 39.17 33.63 1.43 Event + AnyEvent watchers	2199	Event/Any 16000 1203 42.61 34.79 1.80 Event + AnyEvent watchers
		2200	IOAsync/Any 16000 1911 41.92 27.45 16.81 via IO::Async::Loop::IO_Poll
		2201	IOAsync/Any 16000 1726 40.69 26.37 15.25 via IO::Async::Loop::Epoll
936	Glib/Any 16000 1357 98.22 12.41 54.00 quadratic behaviour	2202	Glib/Any 16000 1118 89.00 12.57 51.17 quadratic behaviour
937	Tk/Any 2000 1860 26.97 67.98 14.00 SEGV with >> 2000 watchers	2203	Tk/Any 2000 1346 20.96 10.75 8.00 SEGV with >> 2000 watchers
938	POE/Event 2000 6644 108.64 736.02 14.73 via POE::Loop::Event	2204	POE/Any 2000 6951 108.97 795.32 14.24 via POE::Loop::Event
939	POE/Select 2000 6343 94.13 809.12 565.96 via POE::Loop::Select	2205	POE/Any 2000 6648 94.79 774.40 575.51 via POE::Loop::Select
940		2206
941	=head2 Discussion	2207	=head3 Discussion
942		2208
943	The benchmark does I<not> measure scalability of the event loop very	2209	The benchmark does I<not> measure scalability of the event loop very
944	well. For example, a select-based event loop (such as the pure perl one)	2210	well. For example, a select-based event loop (such as the pure perl one)
945	can never compete with an event loop that uses epoll when the number of	2211	can never compete with an event loop that uses epoll when the number of
946	file descriptors grows high. In this benchmark, all events become ready at	2212	file descriptors grows high. In this benchmark, all events become ready at
947	the same time, so select/poll-based implementations get an unnatural speed	2213	the same time, so select/poll-based implementations get an unnatural speed
948	boost.	2214	boost.
949		2215
		2216	Also, note that the number of watchers usually has a nonlinear effect on
		2217	overall speed, that is, creating twice as many watchers doesn't take twice
		2218	the time - usually it takes longer. This puts event loops tested with a
		2219	higher number of watchers at a disadvantage.
		2220
		2221	To put the range of results into perspective, consider that on the
		2222	benchmark machine, handling an event takes roughly 1600 CPU cycles with
		2223	EV, 3100 CPU cycles with AnyEvent's pure perl loop and almost 3000000 CPU
		2224	cycles with POE.
		2225
950	C<EV> is the sole leader regarding speed and memory use, which are both	2226	C<EV> is the sole leader regarding speed and memory use, which are both
951	maximal/minimal, respectively. Even when going through AnyEvent, there are	2227	maximal/minimal, respectively. When using the L<AE> API there is zero
952	only two event loops that use slightly less memory (the C<Event> module	2228	overhead (when going through the AnyEvent API create is about 5-6 times
953	natively and the pure perl backend), and no faster event models, not even	2229	slower, with other times being equal, so still uses far less memory than
954	C<Event> natively.	2230	any other event loop and is still faster than Event natively).
955		2231
956	The pure perl implementation is hit in a few sweet spots (both the	2232	The pure perl implementation is hit in a few sweet spots (both the
957	zero timeout and the use of a single fd hit optimisations in the perl	2233	constant timeout and the use of a single fd hit optimisations in the perl
958	interpreter and the backend itself, and all watchers become ready at the	2234	interpreter and the backend itself). Nevertheless this shows that it
959	same time). Nevertheless this shows that it adds very little overhead in	2235	adds very little overhead in itself. Like any select-based backend its
960	itself. Like any select-based backend its performance becomes really bad	2236	performance becomes really bad with lots of file descriptors (and few of
961	with lots of file descriptors (and few of them active), of course, but	2237	them active), of course, but this was not subject of this benchmark.
962	this was not subject of this benchmark.
963		2238
964	The C<Event> module has a relatively high setup and callback invocation cost,	2239	The C<Event> module has a relatively high setup and callback invocation
965	but overall scores on the third place.	2240	cost, but overall scores in on the third place.
966		2241
		2242	C<IO::Async> performs admirably well, about on par with C<Event>, even
		2243	when using its pure perl backend.
		2244
967	C<Glib>'s memory usage is quite a bit bit higher, but it features a	2245	C<Glib>'s memory usage is quite a bit higher, but it features a
968	faster callback invocation and overall ends up in the same class as	2246	faster callback invocation and overall ends up in the same class as
969	C<Event>. However, Glib scales extremely badly, doubling the number of	2247	C<Event>. However, Glib scales extremely badly, doubling the number of
970	watchers increases the processing time by more than a factor of four,	2248	watchers increases the processing time by more than a factor of four,
971	making it completely unusable when using larger numbers of watchers	2249	making it completely unusable when using larger numbers of watchers
972	(note that only a single file descriptor was used in the benchmark, so	2250	(note that only a single file descriptor was used in the benchmark, so
…		…
975	The C<Tk> adaptor works relatively well. The fact that it crashes with	2253	The C<Tk> adaptor works relatively well. The fact that it crashes with
976	more than 2000 watchers is a big setback, however, as correctness takes	2254	more than 2000 watchers is a big setback, however, as correctness takes
977	precedence over speed. Nevertheless, its performance is surprising, as the	2255	precedence over speed. Nevertheless, its performance is surprising, as the
978	file descriptor is dup()ed for each watcher. This shows that the dup()	2256	file descriptor is dup()ed for each watcher. This shows that the dup()
979	employed by some adaptors is not a big performance issue (it does incur a	2257	employed by some adaptors is not a big performance issue (it does incur a
980	hidden memory cost inside the kernel, though, that is not reflected in the	2258	hidden memory cost inside the kernel which is not reflected in the figures
981	figures above).	2259	above).
982		2260
983	C<POE>, regardless of underlying event loop (wether using its pure perl	2261	C<POE>, regardless of underlying event loop (whether using its pure perl
984	select-based backend or the Event module) shows abysmal performance and	2262	select-based backend or the Event module, the POE-EV backend couldn't
		2263	be tested because it wasn't working) shows abysmal performance and
985	memory usage: Watchers use almost 30 times as much memory as EV watchers,	2264	memory usage with AnyEvent: Watchers use almost 30 times as much memory
986	and 10 times as much memory as both Event or EV via AnyEvent. Watcher	2265	as EV watchers, and 10 times as much memory as Event (the high memory
		2266	requirements are caused by requiring a session for each watcher). Watcher
987	invocation is almost 900 times slower than with AnyEvent's pure perl	2267	invocation speed is almost 900 times slower than with AnyEvent's pure perl
		2268	implementation.
		2269
988	implementation. The design of the POE adaptor class in AnyEvent can not	2270	The design of the POE adaptor class in AnyEvent can not really account
989	really account for this, as session creation overhead is small compared	2271	for the performance issues, though, as session creation overhead is
990	to execution of the state machine, which is coded pretty optimally within	2272	small compared to execution of the state machine, which is coded pretty
991	L<AnyEvent::Impl::POE>. POE simply seems to be abysmally slow.	2273	optimally within L<AnyEvent::Impl::POE> (and while everybody agrees that
		2274	using multiple sessions is not a good approach, especially regarding
		2275	memory usage, even the author of POE could not come up with a faster
		2276	design).
992		2277
993	=head2 Summary	2278	=head3 Summary
994		2279
		2280	=over 4
		2281
995	Using EV through AnyEvent is faster than any other event loop, but most	2282	=item * Using EV through AnyEvent is faster than any other event loop
996	event loops have acceptable performance with or without AnyEvent.	2283	(even when used without AnyEvent), but most event loops have acceptable
		2284	performance with or without AnyEvent.
997		2285
998	The overhead AnyEvent adds is usually much smaller than the overhead of	2286	=item * The overhead AnyEvent adds is usually much smaller than the overhead of
999	the actual event loop, only with extremely fast event loops such as the EV	2287	the actual event loop, only with extremely fast event loops such as EV
1000	adds AnyEvent significant overhead.	2288	adds AnyEvent significant overhead.
1001		2289
1002	And you should simply avoid POE like the plague if you want performance or	2290	=item * You should avoid POE like the plague if you want performance or
1003	reasonable memory usage.	2291	reasonable memory usage.
1004		2292
		2293	=back
		2294
		2295	=head2 BENCHMARKING THE LARGE SERVER CASE
		2296
		2297	This benchmark actually benchmarks the event loop itself. It works by
		2298	creating a number of "servers": each server consists of a socket pair, a
		2299	timeout watcher that gets reset on activity (but never fires), and an I/O
		2300	watcher waiting for input on one side of the socket. Each time the socket
		2301	watcher reads a byte it will write that byte to a random other "server".
		2302
		2303	The effect is that there will be a lot of I/O watchers, only part of which
		2304	are active at any one point (so there is a constant number of active
		2305	fds for each loop iteration, but which fds these are is random). The
		2306	timeout is reset each time something is read because that reflects how
		2307	most timeouts work (and puts extra pressure on the event loops).
		2308
		2309	In this benchmark, we use 10000 socket pairs (20000 sockets), of which 100
		2310	(1%) are active. This mirrors the activity of large servers with many
		2311	connections, most of which are idle at any one point in time.
		2312
		2313	Source code for this benchmark is found as F<eg/bench2> in the AnyEvent
		2314	distribution. It uses the L<AE> interface, which makes a real difference
		2315	for the EV and Perl backends only.
		2316
		2317	=head3 Explanation of the columns
		2318
		2319	I<sockets> is the number of sockets, and twice the number of "servers" (as
		2320	each server has a read and write socket end).
		2321
		2322	I<create> is the time it takes to create a socket pair (which is
		2323	nontrivial) and two watchers: an I/O watcher and a timeout watcher.
		2324
		2325	I<request>, the most important value, is the time it takes to handle a
		2326	single "request", that is, reading the token from the pipe and forwarding
		2327	it to another server. This includes deleting the old timeout and creating
		2328	a new one that moves the timeout into the future.
		2329
		2330	=head3 Results
		2331
		2332	name sockets create request
		2333	EV 20000 62.66 7.99
		2334	Perl 20000 68.32 32.64
		2335	IOAsync 20000 174.06 101.15 epoll
		2336	IOAsync 20000 174.67 610.84 poll
		2337	Event 20000 202.69 242.91
		2338	Glib 20000 557.01 1689.52
		2339	POE 20000 341.54 12086.32 uses POE::Loop::Event
		2340
		2341	=head3 Discussion
		2342
		2343	This benchmark I<does> measure scalability and overall performance of the
		2344	particular event loop.
		2345
		2346	EV is again fastest. Since it is using epoll on my system, the setup time
		2347	is relatively high, though.
		2348
		2349	Perl surprisingly comes second. It is much faster than the C-based event
		2350	loops Event and Glib.
		2351
		2352	IO::Async performs very well when using its epoll backend, and still quite
		2353	good compared to Glib when using its pure perl backend.
		2354
		2355	Event suffers from high setup time as well (look at its code and you will
		2356	understand why). Callback invocation also has a high overhead compared to
		2357	the C<< $_->() for .. >>-style loop that the Perl event loop uses. Event
		2358	uses select or poll in basically all documented configurations.
		2359
		2360	Glib is hit hard by its quadratic behaviour w.r.t. many watchers. It
		2361	clearly fails to perform with many filehandles or in busy servers.
		2362
		2363	POE is still completely out of the picture, taking over 1000 times as long
		2364	as EV, and over 100 times as long as the Perl implementation, even though
		2365	it uses a C-based event loop in this case.
		2366
		2367	=head3 Summary
		2368
		2369	=over 4
		2370
		2371	=item * The pure perl implementation performs extremely well.
		2372
		2373	=item * Avoid Glib or POE in large projects where performance matters.
		2374
		2375	=back
		2376
		2377	=head2 BENCHMARKING SMALL SERVERS
		2378
		2379	While event loops should scale (and select-based ones do not...) even to
		2380	large servers, most programs we (or I :) actually write have only a few
		2381	I/O watchers.
		2382
		2383	In this benchmark, I use the same benchmark program as in the large server
		2384	case, but it uses only eight "servers", of which three are active at any
		2385	one time. This should reflect performance for a small server relatively
		2386	well.
		2387
		2388	The columns are identical to the previous table.
		2389
		2390	=head3 Results
		2391
		2392	name sockets create request
		2393	EV 16 20.00 6.54
		2394	Perl 16 25.75 12.62
		2395	Event 16 81.27 35.86
		2396	Glib 16 32.63 15.48
		2397	POE 16 261.87 276.28 uses POE::Loop::Event
		2398
		2399	=head3 Discussion
		2400
		2401	The benchmark tries to test the performance of a typical small
		2402	server. While knowing how various event loops perform is interesting, keep
		2403	in mind that their overhead in this case is usually not as important, due
		2404	to the small absolute number of watchers (that is, you need efficiency and
		2405	speed most when you have lots of watchers, not when you only have a few of
		2406	them).
		2407
		2408	EV is again fastest.
		2409
		2410	Perl again comes second. It is noticeably faster than the C-based event
		2411	loops Event and Glib, although the difference is too small to really
		2412	matter.
		2413
		2414	POE also performs much better in this case, but is is still far behind the
		2415	others.
		2416
		2417	=head3 Summary
		2418
		2419	=over 4
		2420
		2421	=item * C-based event loops perform very well with small number of
		2422	watchers, as the management overhead dominates.
		2423
		2424	=back
		2425
		2426	=head2 THE IO::Lambda BENCHMARK
		2427
		2428	Recently I was told about the benchmark in the IO::Lambda manpage, which
		2429	could be misinterpreted to make AnyEvent look bad. In fact, the benchmark
		2430	simply compares IO::Lambda with POE, and IO::Lambda looks better (which
		2431	shouldn't come as a surprise to anybody). As such, the benchmark is
		2432	fine, and mostly shows that the AnyEvent backend from IO::Lambda isn't
		2433	very optimal. But how would AnyEvent compare when used without the extra
		2434	baggage? To explore this, I wrote the equivalent benchmark for AnyEvent.
		2435
		2436	The benchmark itself creates an echo-server, and then, for 500 times,
		2437	connects to the echo server, sends a line, waits for the reply, and then
		2438	creates the next connection. This is a rather bad benchmark, as it doesn't
		2439	test the efficiency of the framework or much non-blocking I/O, but it is a
		2440	benchmark nevertheless.
		2441
		2442	name runtime
		2443	Lambda/select 0.330 sec
		2444	+ optimized 0.122 sec
		2445	Lambda/AnyEvent 0.327 sec
		2446	+ optimized 0.138 sec
		2447	Raw sockets/select 0.077 sec
		2448	POE/select, components 0.662 sec
		2449	POE/select, raw sockets 0.226 sec
		2450	POE/select, optimized 0.404 sec
		2451
		2452	AnyEvent/select/nb 0.085 sec
		2453	AnyEvent/EV/nb 0.068 sec
		2454	+state machine 0.134 sec
		2455
		2456	The benchmark is also a bit unfair (my fault): the IO::Lambda/POE
		2457	benchmarks actually make blocking connects and use 100% blocking I/O,
		2458	defeating the purpose of an event-based solution. All of the newly
		2459	written AnyEvent benchmarks use 100% non-blocking connects (using
		2460	AnyEvent::Socket::tcp_connect and the asynchronous pure perl DNS
		2461	resolver), so AnyEvent is at a disadvantage here, as non-blocking connects
		2462	generally require a lot more bookkeeping and event handling than blocking
		2463	connects (which involve a single syscall only).
		2464
		2465	The last AnyEvent benchmark additionally uses L<AnyEvent::Handle>, which
		2466	offers similar expressive power as POE and IO::Lambda, using conventional
		2467	Perl syntax. This means that both the echo server and the client are 100%
		2468	non-blocking, further placing it at a disadvantage.
		2469
		2470	As you can see, the AnyEvent + EV combination even beats the
		2471	hand-optimised "raw sockets benchmark", while AnyEvent + its pure perl
		2472	backend easily beats IO::Lambda and POE.
		2473
		2474	And even the 100% non-blocking version written using the high-level (and
		2475	slow :) L<AnyEvent::Handle> abstraction beats both POE and IO::Lambda
		2476	higher level ("unoptimised") abstractions by a large margin, even though
		2477	it does all of DNS, tcp-connect and socket I/O in a non-blocking way.
		2478
		2479	The two AnyEvent benchmarks programs can be found as F<eg/ae0.pl> and
		2480	F<eg/ae2.pl> in the AnyEvent distribution, the remaining benchmarks are
		2481	part of the IO::Lambda distribution and were used without any changes.
		2482
		2483
		2484	=head1 SIGNALS
		2485
		2486	AnyEvent currently installs handlers for these signals:
		2487
		2488	=over 4
		2489
		2490	=item SIGCHLD
		2491
		2492	A handler for C<SIGCHLD> is installed by AnyEvent's child watcher
		2493	emulation for event loops that do not support them natively. Also, some
		2494	event loops install a similar handler.
		2495
		2496	Additionally, when AnyEvent is loaded and SIGCHLD is set to IGNORE, then
		2497	AnyEvent will reset it to default, to avoid losing child exit statuses.
		2498
		2499	=item SIGPIPE
		2500
		2501	A no-op handler is installed for C<SIGPIPE> when C<$SIG{PIPE}> is C<undef>
		2502	when AnyEvent gets loaded.
		2503
		2504	The rationale for this is that AnyEvent users usually do not really depend
		2505	on SIGPIPE delivery (which is purely an optimisation for shell use, or
		2506	badly-written programs), but C<SIGPIPE> can cause spurious and rare
		2507	program exits as a lot of people do not expect C<SIGPIPE> when writing to
		2508	some random socket.
		2509
		2510	The rationale for installing a no-op handler as opposed to ignoring it is
		2511	that this way, the handler will be restored to defaults on exec.
		2512
		2513	Feel free to install your own handler, or reset it to defaults.
		2514
		2515	=back
		2516
		2517	=cut
		2518
		2519	undef $SIG{CHLD}
		2520	if $SIG{CHLD} eq 'IGNORE';
		2521
		2522	$SIG{PIPE} = sub { }
		2523	unless defined $SIG{PIPE};
		2524
		2525	=head1 RECOMMENDED/OPTIONAL MODULES
		2526
		2527	One of AnyEvent's main goals is to be 100% Pure-Perl(tm): only perl (and
		2528	its built-in modules) are required to use it.
		2529
		2530	That does not mean that AnyEvent won't take advantage of some additional
		2531	modules if they are installed.
		2532
		2533	This section explains which additional modules will be used, and how they
		2534	affect AnyEvent's operation.
		2535
		2536	=over 4
		2537
		2538	=item L<Async::Interrupt>
		2539
		2540	This slightly arcane module is used to implement fast signal handling: To
		2541	my knowledge, there is no way to do completely race-free and quick
		2542	signal handling in pure perl. To ensure that signals still get
		2543	delivered, AnyEvent will start an interval timer to wake up perl (and
		2544	catch the signals) with some delay (default is 10 seconds, look for
		2545	C<$AnyEvent::MAX_SIGNAL_LATENCY>).
		2546
		2547	If this module is available, then it will be used to implement signal
		2548	catching, which means that signals will not be delayed, and the event loop
		2549	will not be interrupted regularly, which is more efficient (and good for
		2550	battery life on laptops).
		2551
		2552	This affects not just the pure-perl event loop, but also other event loops
		2553	that have no signal handling on their own (e.g. Glib, Tk, Qt).
		2554
		2555	Some event loops (POE, Event, Event::Lib) offer signal watchers natively,
		2556	and either employ their own workarounds (POE) or use AnyEvent's workaround
		2557	(using C<$AnyEvent::MAX_SIGNAL_LATENCY>). Installing L<Async::Interrupt>
		2558	does nothing for those backends.
		2559
		2560	=item L<EV>
		2561
		2562	This module isn't really "optional", as it is simply one of the backend
		2563	event loops that AnyEvent can use. However, it is simply the best event
		2564	loop available in terms of features, speed and stability: It supports
		2565	the AnyEvent API optimally, implements all the watcher types in XS, does
		2566	automatic timer adjustments even when no monotonic clock is available,
		2567	can take avdantage of advanced kernel interfaces such as C<epoll> and
		2568	C<kqueue>, and is the fastest backend I<by far>. You can even embed
		2569	L<Glib>/L<Gtk2> in it (or vice versa, see L<EV::Glib> and L<Glib::EV>).
		2570
		2571	If you only use backends that rely on another event loop (e.g. C<Tk>),
		2572	then this module will do nothing for you.
		2573
		2574	=item L<Guard>
		2575
		2576	The guard module, when used, will be used to implement
		2577	C<AnyEvent::Util::guard>. This speeds up guards considerably (and uses a
		2578	lot less memory), but otherwise doesn't affect guard operation much. It is
		2579	purely used for performance.
		2580
		2581	=item L<JSON> and L<JSON::XS>
		2582
		2583	One of these modules is required when you want to read or write JSON data
		2584	via L<AnyEvent::Handle>. L<JSON> is also written in pure-perl, but can take
		2585	advantage of the ultra-high-speed L<JSON::XS> module when it is installed.
		2586
		2587	=item L<Net::SSLeay>
		2588
		2589	Implementing TLS/SSL in Perl is certainly interesting, but not very
		2590	worthwhile: If this module is installed, then L<AnyEvent::Handle> (with
		2591	the help of L<AnyEvent::TLS>), gains the ability to do TLS/SSL.
		2592
		2593	=item L<Time::HiRes>
		2594
		2595	This module is part of perl since release 5.008. It will be used when the
		2596	chosen event library does not come with a timing source of its own. The
		2597	pure-perl event loop (L<AnyEvent::Impl::Perl>) will additionally use it to
		2598	try to use a monotonic clock for timing stability.
		2599
		2600	=back
		2601
1005		2602
1006	=head1 FORK	2603	=head1 FORK
1007		2604
1008	Most event libraries are not fork-safe. The ones who are usually are	2605	Most event libraries are not fork-safe. The ones who are usually are
1009	because they are so inefficient. Only L<EV> is fully fork-aware.	2606	because they rely on inefficient but fork-safe C<select> or C<poll> calls
		2607	- higher performance APIs such as BSD's kqueue or the dreaded Linux epoll
		2608	are usually badly thought-out hacks that are incompatible with fork in
		2609	one way or another. Only L<EV> is fully fork-aware and ensures that you
		2610	continue event-processing in both parent and child (or both, if you know
		2611	what you are doing).
		2612
		2613	This means that, in general, you cannot fork and do event processing in
		2614	the child if the event library was initialised before the fork (which
		2615	usually happens when the first AnyEvent watcher is created, or the library
		2616	is loaded).
1010		2617
1011	If you have to fork, you must either do so I<before> creating your first	2618	If you have to fork, you must either do so I<before> creating your first
1012	watcher OR you must not use AnyEvent at all in the child.	2619	watcher OR you must not use AnyEvent at all in the child OR you must do
		2620	something completely out of the scope of AnyEvent.
		2621
		2622	The problem of doing event processing in the parent I<and> the child
		2623	is much more complicated: even for backends that I<are> fork-aware or
		2624	fork-safe, their behaviour is not usually what you want: fork clones all
		2625	watchers, that means all timers, I/O watchers etc. are active in both
		2626	parent and child, which is almost never what you want. USing C<exec>
		2627	to start worker children from some kind of manage rprocess is usually
		2628	preferred, because it is much easier and cleaner, at the expense of having
		2629	to have another binary.
1013		2630
1014		2631
1015	=head1 SECURITY CONSIDERATIONS	2632	=head1 SECURITY CONSIDERATIONS
1016		2633
1017	AnyEvent can be forced to load any event model via	2634	AnyEvent can be forced to load any event model via
…		…
1022	specified in the variable.	2639	specified in the variable.
1023		2640
1024	You can make AnyEvent completely ignore this variable by deleting it	2641	You can make AnyEvent completely ignore this variable by deleting it
1025	before the first watcher gets created, e.g. with a C<BEGIN> block:	2642	before the first watcher gets created, e.g. with a C<BEGIN> block:
1026		2643
1027	BEGIN { delete $ENV{PERL_ANYEVENT_MODEL} }	2644	BEGIN { delete $ENV{PERL_ANYEVENT_MODEL} }
1028		2645
1029	use AnyEvent;	2646	use AnyEvent;
		2647
		2648	Similar considerations apply to $ENV{PERL_ANYEVENT_VERBOSE}, as that can
		2649	be used to probe what backend is used and gain other information (which is
		2650	probably even less useful to an attacker than PERL_ANYEVENT_MODEL), and
		2651	$ENV{PERL_ANYEVENT_STRICT}.
		2652
		2653	Note that AnyEvent will remove I<all> environment variables starting with
		2654	C<PERL_ANYEVENT_> from C<%ENV> when it is loaded while taint mode is
		2655	enabled.
		2656
		2657
		2658	=head1 BUGS
		2659
		2660	Perl 5.8 has numerous memleaks that sometimes hit this module and are hard
		2661	to work around. If you suffer from memleaks, first upgrade to Perl 5.10
		2662	and check wether the leaks still show up. (Perl 5.10.0 has other annoying
		2663	memleaks, such as leaking on C<map> and C<grep> but it is usually not as
		2664	pronounced).
1030		2665
1031		2666
1032	=head1 SEE ALSO	2667	=head1 SEE ALSO
1033		2668
1034	Event modules: L<Coro::EV>, L<EV>, L<EV::Glib>, L<Glib::EV>,	2669	Tutorial/Introduction: L<AnyEvent::Intro>.
1035	L<Coro::Event>, L<Event>, L<Glib::Event>, L<Glib>, L<Coro>, L<Tk>,	2670
		2671	FAQ: L<AnyEvent::FAQ>.
		2672
		2673	Utility functions: L<AnyEvent::Util>.
		2674
		2675	Event modules: L<EV>, L<EV::Glib>, L<Glib::EV>, L<Event>, L<Glib::Event>,
1036	L<Event::Lib>, L<Qt>, L<POE>.	2676	L<Glib>, L<Tk>, L<Event::Lib>, L<Qt>, L<POE>.
1037		2677
1038	Implementations: L<AnyEvent::Impl::CoroEV>, L<AnyEvent::Impl::EV>,	2678	Implementations: L<AnyEvent::Impl::EV>, L<AnyEvent::Impl::Event>,
		2679	L<AnyEvent::Impl::Glib>, L<AnyEvent::Impl::Tk>, L<AnyEvent::Impl::Perl>,
		2680	L<AnyEvent::Impl::EventLib>, L<AnyEvent::Impl::Qt>,
1039	L<AnyEvent::Impl::CoroEvent>, L<AnyEvent::Impl::Event>, L<AnyEvent::Impl::Glib>,	2681	L<AnyEvent::Impl::POE>, L<AnyEvent::Impl::IOAsync>, L<Anyevent::Impl::Irssi>.
1040	L<AnyEvent::Impl::Tk>, L<AnyEvent::Impl::Perl>, L<AnyEvent::Impl::EventLib>,
1041	L<AnyEvent::Impl::Qt>, L<AnyEvent::Impl::POE>.
1042		2682
		2683	Non-blocking file handles, sockets, TCP clients and
		2684	servers: L<AnyEvent::Handle>, L<AnyEvent::Socket>, L<AnyEvent::TLS>.
		2685
		2686	Asynchronous DNS: L<AnyEvent::DNS>.
		2687
		2688	Thread support: L<Coro>, L<Coro::AnyEvent>, L<Coro::EV>, L<Coro::Event>.
		2689
1043	Nontrivial usage examples: L<Net::FCP>, L<Net::XMPP2>.	2690	Nontrivial usage examples: L<AnyEvent::GPSD>, L<AnyEvent::IRC>,
		2691	L<AnyEvent::HTTP>.
1044		2692
1045		2693
1046	=head1 AUTHOR	2694	=head1 AUTHOR
1047		2695
1048	Marc Lehmann <schmorp@schmorp.de>	2696	Marc Lehmann <schmorp@schmorp.de>
1049	http://home.schmorp.de/	2697	http://home.schmorp.de/
1050		2698
1051	=cut	2699	=cut
1052		2700
1053	1	2701	1
1054		2702

Diff Legend

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