ViewVC Help

View File | Revision Log | Show Annotations | Download File

/cvs/AnyEvent/lib/AnyEvent.pm

Comparing AnyEvent/lib/AnyEvent.pm (file contents):
Revision 1.176 by root, Wed Aug 20 12:37:21 2008 UTC vs.
Revision 1.408 by root, Thu Dec 6 12:04:23 2012 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, 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	FLTK 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 { ... });
12		17
		18	# one-shot or repeating timers
13	my $w = AnyEvent->timer (after => $seconds, cb => sub { ... });	19	my $w = AnyEvent->timer (after => $seconds, cb => sub { ... });
14	my $w = AnyEvent->timer (after => $seconds, interval => $seconds, cb => ...	20	my $w = AnyEvent->timer (after => $seconds, interval => $seconds, cb => ...);
15		21
16	print AnyEvent->now; # prints current event loop time	22	print AnyEvent->now; # prints current event loop time
17	print AnyEvent->time; # think Time::HiRes::time or simply CORE::time.	23	print AnyEvent->time; # think Time::HiRes::time or simply CORE::time.
18		24
		25	# POSIX signal
19	my $w = AnyEvent->signal (signal => "TERM", cb => sub { ... });	26	my $w = AnyEvent->signal (signal => "TERM", cb => sub { ... });
20		27
		28	# child process exit
21	my $w = AnyEvent->child (pid => $pid, cb => sub {	29	my $w = AnyEvent->child (pid => $pid, cb => sub {
22	my ($pid, $status) = @_;	30	my ($pid, $status) = @_;
23	...	31	...
24	});	32	});
		33
		34	# called when event loop idle (if applicable)
		35	my $w = AnyEvent->idle (cb => sub { ... });
25		36
26	my $w = AnyEvent->condvar; # stores whether a condition was flagged	37	my $w = AnyEvent->condvar; # stores whether a condition was flagged
27	$w->send; # wake up current and all future recv's	38	$w->send; # wake up current and all future recv's
28	$w->recv; # enters "main loop" till $condvar gets ->send	39	$w->recv; # enters "main loop" till $condvar gets ->send
29	# use a condvar in callback mode:	40	# use a condvar in callback mode:
…		…
32	=head1 INTRODUCTION/TUTORIAL	43	=head1 INTRODUCTION/TUTORIAL
33		44
34	This manpage is mainly a reference manual. If you are interested	45	This manpage is mainly a reference manual. If you are interested
35	in a tutorial or some gentle introduction, have a look at the	46	in a tutorial or some gentle introduction, have a look at the
36	L<AnyEvent::Intro> manpage.	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.
37		58
38	=head1 WHY YOU SHOULD USE THIS MODULE (OR NOT)	59	=head1 WHY YOU SHOULD USE THIS MODULE (OR NOT)
39		60
40	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
41	nowadays. So what is different about AnyEvent?	62	nowadays. So what is different about AnyEvent?
…		…
57	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
58	model you use.	79	model you use.
59		80
60	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
61	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
62	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
63	cannot use anything else, as they are simply incompatible to everything	84	cannot use anything else, as they are simply incompatible to everything
64	that isn't them. What's worse, all the potential users of your	85	that isn't them. What's worse, all the potential users of your
65	module are 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.
66		87
67	AnyEvent is different: AnyEvent + POE works fine. AnyEvent + Glib works	88	AnyEvent is different: AnyEvent + POE works fine. AnyEvent + Glib works
68	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
69	with the rest: POE + IO::Async? No go. Tk + Event? No go. Again: if	90	with the rest: POE + EV? No go. Tk + Event? No go. Again: if your module
70	your module uses one of those, every user of your module has to use it,	91	uses one of those, every user of your module has to use it, too. But if
71	too. But if your module uses AnyEvent, it works transparently with all	92	your module uses AnyEvent, it works transparently with all event models it
72	event models it supports (including stuff like IO::Async, as long as those	93	supports (including stuff like IO::Async, as long as those use one of the
73	use one of the supported event loops. It is trivial to add new event loops	94	supported event loops. It is easy to add new event loops to AnyEvent, too,
74	to AnyEvent, too, so it is future-proof).	95	so it is future-proof).
75		96
76	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
77	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
78	modules, you get an enormous amount of code and strict rules you have to	99	modules, you get an enormous amount of code and strict rules you have to
79	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
80	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
81	technically possible.	102	technically possible.
82		103
83	Of course, AnyEvent comes with a big (and fully optional!) toolbox	104	Of course, AnyEvent comes with a big (and fully optional!) toolbox
84	of useful functionality, such as an asynchronous DNS resolver, 100%	105	of useful functionality, such as an asynchronous DNS resolver, 100%
…		…
90	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
91	model, you should I<not> use this module.	112	model, you should I<not> use this module.
92		113
93	=head1 DESCRIPTION	114	=head1 DESCRIPTION
94		115
95	L<AnyEvent> provides an identical interface to multiple event loops. This	116	L<AnyEvent> provides a uniform interface to various event loops. This
96	allows module authors to utilise an event loop without forcing module	117	allows module authors to use event loop functionality without forcing
97	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
98	peacefully at any one time).	119	than one event loop cannot coexist peacefully).
99		120
100	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>
101	module.	122	module.
102		123
103	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
104	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
105	following modules is already loaded: L<EV>,	126	following modules is already loaded: L<EV>, L<AnyEvent::Loop>,
106	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
107	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
108	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
109	adaptor should always succeed) in the order given. The first one that can	130	available, the pure-perl L<AnyEvent::Loop> should always work, so
110	be successfully loaded will be used. If, after this, still none could be	131	the other two are not normally tried.
111	found, AnyEvent will fall back to a pure-perl event loop, which is not
112	very efficient, but should work everywhere.
113		132
114	Because AnyEvent first checks for modules that are already loaded, loading	133	Because AnyEvent first checks for modules that are already loaded, loading
115	an event model explicitly before first using AnyEvent will likely make	134	an event model explicitly before first using AnyEvent will likely make
116	that model the default. For example:	135	that model the default. For example:
117		136
…		…
119	use AnyEvent;	138	use AnyEvent;
120		139
121	# .. AnyEvent will likely default to Tk	140	# .. AnyEvent will likely default to Tk
122		141
123	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
124	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,
125	use AnyEvent so their modules work together with others seamlessly...	144	as very few modules hardcode event loops without announcing this very
		145	loudly.
126		146
127	The pure-perl implementation of AnyEvent is called	147	The pure-perl implementation of AnyEvent is called C<AnyEvent::Loop>. Like
128	C<AnyEvent::Impl::Perl>. Like other event modules you can load it	148	other event modules you can load it explicitly and enjoy the high
129	explicitly and enjoy the high availability of that event loop :)	149	availability of that event loop :)
130		150
131	=head1 WATCHERS	151	=head1 WATCHERS
132		152
133	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
134	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
…		…
137	These watchers are normal Perl objects with normal Perl lifetime. After	157	These watchers are normal Perl objects with normal Perl lifetime. After
138	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
139	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
140	is in control).	160	is in control).
141		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
142	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
143	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
144	to it).	170	to it).
145		171
146	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.
147		173
148	Many watchers either are used with "recursion" (repeating timers for	174	Many watchers either are used with "recursion" (repeating timers for
149	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.
150		176
151	An any way to achieve that is this pattern:	177	One way to achieve that is this pattern:
152		178
153	my $w; $w = AnyEvent->type (arg => value ..., cb => sub {	179	my $w; $w = AnyEvent->type (arg => value ..., cb => sub {
154	# you can use $w here, for example to undef it	180	# you can use $w here, for example to undef it
155	undef $w;	181	undef $w;
156	});	182	});
…		…
159	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
160	declared.	186	declared.
161		187
162	=head2 I/O WATCHERS	188	=head2 I/O WATCHERS
163		189
		190	$w = AnyEvent->io (
		191	fh => <filehandle_or_fileno>,
		192	poll => <"r" or "w">,
		193	cb => <callback>,
		194	);
		195
164	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
165	with the following mandatory key-value pairs as arguments:	197	with the following mandatory key-value pairs as arguments:
166		198
167	C<fh> the Perl I<file handle> (I<not> file descriptor) to watch for events	199	C<fh> is the Perl I<file handle> (or a naked file descriptor) to watch
168	(AnyEvent might or might not keep a reference to this file handle). C<poll>	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
169	must be a string that is either C<r> or C<w>, which creates a watcher	206	C<poll> must be a string that is either C<r> or C<w>, which creates a
170	waiting for "r"eadable or "w"ritable events, respectively. C<cb> is the	207	watcher waiting for "r"eadable or "w"ritable events, respectively.
		208
171	callback to invoke each time the file handle becomes ready.	209	C<cb> is the callback to invoke each time the file handle becomes ready.
172		210
173	Although the callback might get passed parameters, their value and	211	Although the callback might get passed parameters, their value and
174	presence is undefined and you cannot rely on them. Portable AnyEvent	212	presence is undefined and you cannot rely on them. Portable AnyEvent
175	callbacks cannot use arguments passed to I/O watcher callbacks.	213	callbacks cannot use arguments passed to I/O watcher callbacks.
176		214
177	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.
178	You must not close a file handle as long as any watcher is active on the	216	You must not close a file handle as long as any watcher is active on the
179	underlying file descriptor.	217	underlying file descriptor.
180		218
181	Some event loops issue spurious readyness notifications, so you should	219	Some event loops issue spurious readiness notifications, so you should
182	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
183	handles.	221	handles.
184		222
185	Example: wait for readability of STDIN, then read a line and disable the	223	Example: wait for readability of STDIN, then read a line and disable the
186	watcher.	224	watcher.
…		…
191	undef $w;	229	undef $w;
192	});	230	});
193		231
194	=head2 TIME WATCHERS	232	=head2 TIME WATCHERS
195		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
196	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 >>
197	method with the following mandatory arguments:	243	method with the following mandatory arguments:
198		244
199	C<after> specifies after how many seconds (fractional values are	245	C<after> specifies after how many seconds (fractional values are
200	supported) the callback should be invoked. C<cb> is the callback to invoke	246	supported) the callback should be invoked. C<cb> is the callback to invoke
…		…
202		248
203	Although the callback might get passed parameters, their value and	249	Although the callback might get passed parameters, their value and
204	presence is undefined and you cannot rely on them. Portable AnyEvent	250	presence is undefined and you cannot rely on them. Portable AnyEvent
205	callbacks cannot use arguments passed to time watcher callbacks.	251	callbacks cannot use arguments passed to time watcher callbacks.
206		252
207	The callback will normally be invoked once only. If you specify another	253	The callback will normally be invoked only once. If you specify another
208	parameter, C<interval>, as a strictly positive number (> 0), then the	254	parameter, C<interval>, as a strictly positive number (> 0), then the
209	callback will be invoked regularly at that interval (in fractional	255	callback will be invoked regularly at that interval (in fractional
210	seconds) after the first invocation. If C<interval> is specified with a	256	seconds) after the first invocation. If C<interval> is specified with a
211	false value, then it is treated as if it were missing.	257	false value, then it is treated as if it were not specified at all.
212		258
213	The callback will be rescheduled before invoking the callback, but no	259	The callback will be rescheduled before invoking the callback, but no
214	attempt is done to avoid timer drift in most backends, so the interval is	260	attempt is made to avoid timer drift in most backends, so the interval is
215	only approximate.	261	only approximate.
216		262
217	Example: fire an event after 7.7 seconds.	263	Example: fire an event after 7.7 seconds.
218		264
219	my $w = AnyEvent->timer (after => 7.7, cb => sub {	265	my $w = AnyEvent->timer (after => 7.7, cb => sub {
…		…
237		283
238	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
239	use absolute time internally. This makes a difference when your clock	285	use absolute time internally. This makes a difference when your clock
240	"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
241	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
242	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.
243		289
244	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
245	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
246	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)
247	timers.	293	timers.
248		294
249	AnyEvent always prefers relative timers, if available, matching the	295	AnyEvent always prefers relative timers, if available, matching the
250	AnyEvent API.	296	AnyEvent API.
…		…
272	I<In almost all cases (in all cases if you don't care), this is the	318	I<In almost all cases (in all cases if you don't care), this is the
273	function to call when you want to know the current time.>	319	function to call when you want to know the current time.>
274		320
275	This function is also often faster then C<< AnyEvent->time >>, and	321	This function is also often faster then C<< AnyEvent->time >>, and
276	thus the preferred method if you want some timestamp (for example,	322	thus the preferred method if you want some timestamp (for example,
277	L<AnyEvent::Handle> uses this to update it's activity timeouts).	323	L<AnyEvent::Handle> uses this to update its activity timeouts).
278		324
279	The rest of this section is only of relevance if you try to be very exact	325	The rest of this section is only of relevance if you try to be very exact
280	with your timing, you can skip it without bad conscience.	326	with your timing; you can skip it without a bad conscience.
281		327
282	For a practical example of when these times differ, consider L<Event::Lib>	328	For a practical example of when these times differ, consider L<Event::Lib>
283	and L<EV> and the following set-up:	329	and L<EV> and the following set-up:
284		330
285	The event loop is running and has just invoked one of your callback at	331	The event loop is running and has just invoked one of your callbacks at
286	time=500 (assume no other callbacks delay processing). In your callback,	332	time=500 (assume no other callbacks delay processing). In your callback,
287	you wait a second by executing C<sleep 1> (blocking the process for a	333	you wait a second by executing C<sleep 1> (blocking the process for a
288	second) and then (at time=501) you create a relative timer that fires	334	second) and then (at time=501) you create a relative timer that fires
289	after three seconds.	335	after three seconds.
290		336
…		…
308	In either case, if you care (and in most cases, you don't), then you	354	In either case, if you care (and in most cases, you don't), then you
309	can get whatever behaviour you want with any event loop, by taking the	355	can get whatever behaviour you want with any event loop, by taking the
310	difference between C<< AnyEvent->time >> and C<< AnyEvent->now >> into	356	difference between C<< AnyEvent->time >> and C<< AnyEvent->now >> into
311	account.	357	account.
312		358
		359	=item AnyEvent->now_update
		360
		361	Some event loops (such as L<EV> or L<AnyEvent::Loop>) cache the current
		362	time for each loop iteration (see the discussion of L<< AnyEvent->now >>,
		363	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
313	=back	381	=back
314		382
315	=head2 SIGNAL WATCHERS	383	=head2 SIGNAL WATCHERS
		384
		385	$w = AnyEvent->signal (signal => <uppercase_signal_name>, cb => <callback>);
316		386
317	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
318	I<name> in uppercase and without any C<SIG> prefix, C<cb> is the Perl	388	I<name> in uppercase and without any C<SIG> prefix, C<cb> is the Perl
319	callback to be invoked whenever a signal occurs.	389	callback to be invoked whenever a signal occurs.
320		390
…		…
326	invocation, and callback invocation will be synchronous. Synchronous means	396	invocation, and callback invocation will be synchronous. Synchronous means
327	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,
328	but it is guaranteed not to interrupt any other callbacks.	398	but it is guaranteed not to interrupt any other callbacks.
329		399
330	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
331	between multiple watchers.	401	between multiple watchers, and AnyEvent will ensure that signals will not
		402	interrupt your program at bad times.
332		403
333	This watcher might use C<%SIG>, so programs overwriting those signals	404	This watcher might use C<%SIG> (depending on the event loop used),
334	directly will likely not work correctly.	405	so programs overwriting those signals directly will likely not work
		406	correctly.
335		407
336	Example: exit on SIGINT	408	Example: exit on SIGINT
337		409
338	my $w = AnyEvent->signal (signal => "INT", cb => sub { exit 1 });	410	my $w = AnyEvent->signal (signal => "INT", cb => sub { exit 1 });
339		411
		412	=head3 Restart Behaviour
		413
		414	While restart behaviour is up to the event loop implementation, most will
		415	not restart syscalls (that includes L<Async::Interrupt> and AnyEvent's
		416	pure perl implementation).
		417
		418	=head3 Safe/Unsafe Signals
		419
		420	Perl signals can be either "safe" (synchronous to opcode handling)
		421	or "unsafe" (asynchronous) - the former might delay signal delivery
		422	indefinitely, the latter might corrupt your memory.
		423
		424	AnyEvent signal handlers are, in addition, synchronous to the event loop,
		425	i.e. they will not interrupt your running perl program but will only be
		426	called as part of the normal event handling (just like timer, I/O etc.
		427	callbacks, too).
		428
		429	=head3 Signal Races, Delays and Workarounds
		430
		431	Many event loops (e.g. Glib, Tk, Qt, IO::Async) do not support
		432	attaching callbacks to signals in a generic way, which is a pity,
		433	as you cannot do race-free signal handling in perl, requiring
		434	C libraries for this. AnyEvent will try to do its best, which
		435	means in some cases, signals will be delayed. The maximum time
		436	a signal might be delayed is 10 seconds by default, but can
		437	be overriden via C<$ENV{PERL_ANYEVENT_MAX_SIGNAL_LATENCY}> or
		438	C<$AnyEvent::MAX_SIGNAL_LATENCY> - see the L<ENVIRONMENT VARIABLES>
		439	section for details.
		440
		441	All these problems can be avoided by installing the optional
		442	L<Async::Interrupt> module, which works with most event loops. It will not
		443	work with inherently broken event loops such as L<Event> or L<Event::Lib>
		444	(and not with L<POE> currently). For those, you just have to suffer the
		445	delays.
		446
340	=head2 CHILD PROCESS WATCHERS	447	=head2 CHILD PROCESS WATCHERS
341		448
		449	$w = AnyEvent->child (pid => <process id>, cb => <callback>);
		450
342	You can also watch on a child process exit and catch its exit status.	451	You can also watch for a child process exit and catch its exit status.
343		452
344	The child process is specified by the C<pid> argument (if set to C<0>, it	453	The child process is specified by the C<pid> argument (on some backends,
345	watches for any child process exit). The watcher will trigger as often	454	using C<0> watches for any child process exit, on others this will
346	as status change for the child are received. This works by installing a	455	croak). The watcher will be triggered only when the child process has
347	signal handler for C<SIGCHLD>. The callback will be called with the pid	456	finished and an exit status is available, not on any trace events
348	and exit status (as returned by waitpid), so unlike other watcher types,	457	(stopped/continued).
349	you I<can> rely on child watcher callback arguments.	458
		459	The callback will be called with the pid and exit status (as returned by
		460	waitpid), so unlike other watcher types, you I<can> rely on child watcher
		461	callback arguments.
		462
		463	This watcher type works by installing a signal handler for C<SIGCHLD>,
		464	and since it cannot be shared, nothing else should use SIGCHLD or reap
		465	random child processes (waiting for specific child processes, e.g. inside
		466	C<system>, is just fine).
350		467
351	There is a slight catch to child watchers, however: you usually start them	468	There is a slight catch to child watchers, however: you usually start them
352	I<after> the child process was created, and this means the process could	469	I<after> the child process was created, and this means the process could
353	have exited already (and no SIGCHLD will be sent anymore).	470	have exited already (and no SIGCHLD will be sent anymore).
354		471
355	Not all event models handle this correctly (POE doesn't), but even for	472	Not all event models handle this correctly (neither POE nor IO::Async do,
		473	see their AnyEvent::Impl manpages for details), but even for event models
356	event models that I<do> handle this correctly, they usually need to be	474	that I<do> handle this correctly, they usually need to be loaded before
357	loaded before the process exits (i.e. before you fork in the first place).	475	the process exits (i.e. before you fork in the first place). AnyEvent's
		476	pure perl event loop handles all cases correctly regardless of when you
		477	start the watcher.
358		478
359	This means you cannot create a child watcher as the very first thing in an	479	This means you cannot create a child watcher as the very first
360	AnyEvent program, you I<have> to create at least one watcher before you	480	thing in an AnyEvent program, you I<have> to create at least one
361	C<fork> the child (alternatively, you can call C<AnyEvent::detect>).	481	watcher before you C<fork> the child (alternatively, you can call
		482	C<AnyEvent::detect>).
		483
		484	As most event loops do not support waiting for child events, they will be
		485	emulated by AnyEvent in most cases, in which case the latency and race
		486	problems mentioned in the description of signal watchers apply.
362		487
363	Example: fork a process and wait for it	488	Example: fork a process and wait for it
364		489
365	my $done = AnyEvent->condvar;	490	my $done = AnyEvent->condvar;
366		491
…		…
376	);	501	);
377		502
378	# do something else, then wait for process exit	503	# do something else, then wait for process exit
379	$done->recv;	504	$done->recv;
380		505
		506	=head2 IDLE WATCHERS
		507
		508	$w = AnyEvent->idle (cb => <callback>);
		509
		510	This will repeatedly invoke the callback after the process becomes idle,
		511	until either the watcher is destroyed or new events have been detected.
		512
		513	Idle watchers are useful when there is a need to do something, but it
		514	is not so important (or wise) to do it instantly. The callback will be
		515	invoked only when there is "nothing better to do", which is usually
		516	defined as "all outstanding events have been handled and no new events
		517	have been detected". That means that idle watchers ideally get invoked
		518	when the event loop has just polled for new events but none have been
		519	detected. Instead of blocking to wait for more events, the idle watchers
		520	will be invoked.
		521
		522	Unfortunately, most event loops do not really support idle watchers (only
		523	EV, Event and Glib do it in a usable fashion) - for the rest, AnyEvent
		524	will simply call the callback "from time to time".
		525
		526	Example: read lines from STDIN, but only process them when the
		527	program is otherwise idle:
		528
		529	my @lines; # read data
		530	my $idle_w;
		531	my $io_w = AnyEvent->io (fh => \*STDIN, poll => 'r', cb => sub {
		532	push @lines, scalar <STDIN>;
		533
		534	# start an idle watcher, if not already done
		535	$idle_w ||= AnyEvent->idle (cb => sub {
		536	# handle only one line, when there are lines left
		537	if (my $line = shift @lines) {
		538	print "handled when idle: $line";
		539	} else {
		540	# otherwise disable the idle watcher again
		541	undef $idle_w;
		542	}
		543	});
		544	});
		545
381	=head2 CONDITION VARIABLES	546	=head2 CONDITION VARIABLES
		547
		548	$cv = AnyEvent->condvar;
		549
		550	$cv->send (<list>);
		551	my @res = $cv->recv;
382		552
383	If you are familiar with some event loops you will know that all of them	553	If you are familiar with some event loops you will know that all of them
384	require you to run some blocking "loop", "run" or similar function that	554	require you to run some blocking "loop", "run" or similar function that
385	will actively watch for new events and call your callbacks.	555	will actively watch for new events and call your callbacks.
386		556
387	AnyEvent is different, it expects somebody else to run the event loop and	557	AnyEvent is slightly different: it expects somebody else to run the event
388	will only block when necessary (usually when told by the user).	558	loop and will only block when necessary (usually when told by the user).
389		559
390	The instrument to do that is called a "condition variable", so called	560	The tool to do that is called a "condition variable", so called because
391	because they represent a condition that must become true.	561	they represent a condition that must become true.
		562
		563	Now is probably a good time to look at the examples further below.
392		564
393	Condition variables can be created by calling the C<< AnyEvent->condvar	565	Condition variables can be created by calling the C<< AnyEvent->condvar
394	>> method, usually without arguments. The only argument pair allowed is	566	>> method, usually without arguments. The only argument pair allowed is
395
396	C<cb>, which specifies a callback to be called when the condition variable	567	C<cb>, which specifies a callback to be called when the condition variable
397	becomes true, with the condition variable as the first argument (but not	568	becomes true, with the condition variable as the first argument (but not
398	the results).	569	the results).
399		570
400	After creation, the condition variable is "false" until it becomes "true"	571	After creation, the condition variable is "false" until it becomes "true"
401	by calling the C<send> method (or calling the condition variable as if it	572	by calling the C<send> method (or calling the condition variable as if it
402	were a callback, read about the caveats in the description for the C<<	573	were a callback, read about the caveats in the description for the C<<
403	->send >> method).	574	->send >> method).
404		575
405	Condition variables are similar to callbacks, except that you can	576	Since condition variables are the most complex part of the AnyEvent API, here are
406	optionally wait for them. They can also be called merge points - points	577	some different mental models of what they are - pick the ones you can connect to:
407	in time where multiple outstanding events have been processed. And yet	578
408	another way to call them is transactions - each condition variable can be	579	=over 4
409	used to represent a transaction, which finishes at some point and delivers	580
410	a result.	581	=item * Condition variables are like callbacks - you can call them (and pass them instead
		582	of callbacks). Unlike callbacks however, you can also wait for them to be called.
		583
		584	=item * Condition variables are signals - one side can emit or send them,
		585	the other side can wait for them, or install a handler that is called when
		586	the signal fires.
		587
		588	=item * Condition variables are like "Merge Points" - points in your program
		589	where you merge multiple independent results/control flows into one.
		590
		591	=item * Condition variables represent a transaction - functions that start
		592	some kind of transaction can return them, leaving the caller the choice
		593	between waiting in a blocking fashion, or setting a callback.
		594
		595	=item * Condition variables represent future values, or promises to deliver
		596	some result, long before the result is available.
		597
		598	=back
411		599
412	Condition variables are very useful to signal that something has finished,	600	Condition variables are very useful to signal that something has finished,
413	for example, if you write a module that does asynchronous http requests,	601	for example, if you write a module that does asynchronous http requests,
414	then a condition variable would be the ideal candidate to signal the	602	then a condition variable would be the ideal candidate to signal the
415	availability of results. The user can either act when the callback is	603	availability of results. The user can either act when the callback is
…		…
428		616
429	Condition variables are represented by hash refs in perl, and the keys	617	Condition variables are represented by hash refs in perl, and the keys
430	used by AnyEvent itself are all named C<_ae_XXX> to make subclassing	618	used by AnyEvent itself are all named C<_ae_XXX> to make subclassing
431	easy (it is often useful to build your own transaction class on top of	619	easy (it is often useful to build your own transaction class on top of
432	AnyEvent). To subclass, use C<AnyEvent::CondVar> as base class and call	620	AnyEvent). To subclass, use C<AnyEvent::CondVar> as base class and call
433	it's C<new> method in your own C<new> method.	621	its C<new> method in your own C<new> method.
434		622
435	There are two "sides" to a condition variable - the "producer side" which	623	There are two "sides" to a condition variable - the "producer side" which
436	eventually calls C<< -> send >>, and the "consumer side", which waits	624	eventually calls C<< -> send >>, and the "consumer side", which waits
437	for the send to occur.	625	for the send to occur.
438		626
439	Example: wait for a timer.	627	Example: wait for a timer.
440		628
441	# wait till the result is ready	629	# condition: "wait till the timer is fired"
442	my $result_ready = AnyEvent->condvar;	630	my $timer_fired = AnyEvent->condvar;
443		631
444	# do something such as adding a timer	632	# create the timer - we could wait for, say
445	# or socket watcher the calls $result_ready->send	633	# a handle becomign ready, or even an
446	# when the "result" is ready.	634	# AnyEvent::HTTP request to finish, but
447	# in this case, we simply use a timer:	635	# in this case, we simply use a timer:
448	my $w = AnyEvent->timer (	636	my $w = AnyEvent->timer (
449	after => 1,	637	after => 1,
450	cb => sub { $result_ready->send },	638	cb => sub { $timer_fired->send },
451	);	639	);
452		640
453	# this "blocks" (while handling events) till the callback	641	# this "blocks" (while handling events) till the callback
454	# calls send	642	# calls ->send
455	$result_ready->recv;	643	$timer_fired->recv;
456		644
457	Example: wait for a timer, but take advantage of the fact that	645	Example: wait for a timer, but take advantage of the fact that condition
458	condition variables are also code references.	646	variables are also callable directly.
459		647
460	my $done = AnyEvent->condvar;	648	my $done = AnyEvent->condvar;
461	my $delay = AnyEvent->timer (after => 5, cb => $done);	649	my $delay = AnyEvent->timer (after => 5, cb => $done);
462	$done->recv;	650	$done->recv;
463		651
…		…
469		657
470	...	658	...
471		659
472	my @info = $couchdb->info->recv;	660	my @info = $couchdb->info->recv;
473		661
474	And this is how you would just ste a callback to be called whenever the	662	And this is how you would just set a callback to be called whenever the
475	results are available:	663	results are available:
476		664
477	$couchdb->info->cb (sub {	665	$couchdb->info->cb (sub {
478	my @info = $_[0]->recv;	666	my @info = $_[0]->recv;
479	});	667	});
…		…
497	immediately from within send.	685	immediately from within send.
498		686
499	Any arguments passed to the C<send> call will be returned by all	687	Any arguments passed to the C<send> call will be returned by all
500	future C<< ->recv >> calls.	688	future C<< ->recv >> calls.
501		689
502	Condition variables are overloaded so one can call them directly	690	Condition variables are overloaded so one can call them directly (as if
503	(as a code reference). Calling them directly is the same as calling	691	they were a code reference). Calling them directly is the same as calling
504	C<send>. Note, however, that many C-based event loops do not handle	692	C<send>.
505	overloading, so as tempting as it may be, passing a condition variable
506	instead of a callback does not work. Both the pure perl and EV loops
507	support overloading, however, as well as all functions that use perl to
508	invoke a callback (as in L<AnyEvent::Socket> and L<AnyEvent::DNS> for
509	example).
510		693
511	=item $cv->croak ($error)	694	=item $cv->croak ($error)
512		695
513	Similar to send, but causes all call's to C<< ->recv >> to invoke	696	Similar to send, but causes all calls to C<< ->recv >> to invoke
514	C<Carp::croak> with the given error message/object/scalar.	697	C<Carp::croak> with the given error message/object/scalar.
515		698
516	This can be used to signal any errors to the condition variable	699	This can be used to signal any errors to the condition variable
517	user/consumer.	700	user/consumer. Doing it this way instead of calling C<croak> directly
		701	delays the error detection, but has the overwhelming advantage that it
		702	diagnoses the error at the place where the result is expected, and not
		703	deep in some event callback with no connection to the actual code causing
		704	the problem.
518		705
519	=item $cv->begin ([group callback])	706	=item $cv->begin ([group callback])
520		707
521	=item $cv->end	708	=item $cv->end
522
523	These two methods are EXPERIMENTAL and MIGHT CHANGE.
524		709
525	These two methods can be used to combine many transactions/events into	710	These two methods can be used to combine many transactions/events into
526	one. For example, a function that pings many hosts in parallel might want	711	one. For example, a function that pings many hosts in parallel might want
527	to use a condition variable for the whole process.	712	to use a condition variable for the whole process.
528		713
529	Every call to C<< ->begin >> will increment a counter, and every call to	714	Every call to C<< ->begin >> will increment a counter, and every call to
530	C<< ->end >> will decrement it. If the counter reaches C<0> in C<< ->end	715	C<< ->end >> will decrement it. If the counter reaches C<0> in C<< ->end
531	>>, the (last) callback passed to C<begin> will be executed. That callback	716	>>, the (last) callback passed to C<begin> will be executed, passing the
532	is I<supposed> to call C<< ->send >>, but that is not required. If no	717	condvar as first argument. That callback is I<supposed> to call C<< ->send
533	callback was set, C<send> will be called without any arguments.	718	>>, but that is not required. If no group callback was set, C<send> will
		719	be called without any arguments.
534		720
535	Let's clarify this with the ping example:	721	You can think of C<< $cv->send >> giving you an OR condition (one call
		722	sends), while C<< $cv->begin >> and C<< $cv->end >> giving you an AND
		723	condition (all C<begin> calls must be C<end>'ed before the condvar sends).
		724
		725	Let's start with a simple example: you have two I/O watchers (for example,
		726	STDOUT and STDERR for a program), and you want to wait for both streams to
		727	close before activating a condvar:
536		728
537	my $cv = AnyEvent->condvar;	729	my $cv = AnyEvent->condvar;
538		730
		731	$cv->begin; # first watcher
		732	my $w1 = AnyEvent->io (fh => $fh1, cb => sub {
		733	defined sysread $fh1, my $buf, 4096
		734	or $cv->end;
		735	});
		736
		737	$cv->begin; # second watcher
		738	my $w2 = AnyEvent->io (fh => $fh2, cb => sub {
		739	defined sysread $fh2, my $buf, 4096
		740	or $cv->end;
		741	});
		742
		743	$cv->recv;
		744
		745	This works because for every event source (EOF on file handle), there is
		746	one call to C<begin>, so the condvar waits for all calls to C<end> before
		747	sending.
		748
		749	The ping example mentioned above is slightly more complicated, as the
		750	there are results to be passwd back, and the number of tasks that are
		751	begun can potentially be zero:
		752
		753	my $cv = AnyEvent->condvar;
		754
539	my %result;	755	my %result;
540	$cv->begin (sub { $cv->send (\%result) });	756	$cv->begin (sub { shift->send (\%result) });
541		757
542	for my $host (@list_of_hosts) {	758	for my $host (@list_of_hosts) {
543	$cv->begin;	759	$cv->begin;
544	ping_host_then_call_callback $host, sub {	760	ping_host_then_call_callback $host, sub {
545	$result{$host} = ...;	761	$result{$host} = ...;
…		…
547	};	763	};
548	}	764	}
549		765
550	$cv->end;	766	$cv->end;
551		767
		768	...
		769
		770	my $results = $cv->recv;
		771
552	This code fragment supposedly pings a number of hosts and calls	772	This code fragment supposedly pings a number of hosts and calls
553	C<send> after results for all then have have been gathered - in any	773	C<send> after results for all then have have been gathered - in any
554	order. To achieve this, the code issues a call to C<begin> when it starts	774	order. To achieve this, the code issues a call to C<begin> when it starts
555	each ping request and calls C<end> when it has received some result for	775	each ping request and calls C<end> when it has received some result for
556	it. Since C<begin> and C<end> only maintain a counter, the order in which	776	it. Since C<begin> and C<end> only maintain a counter, the order in which
…		…
560	loop, which serves two important purposes: first, it sets the callback	780	loop, which serves two important purposes: first, it sets the callback
561	to be called once the counter reaches C<0>, and second, it ensures that	781	to be called once the counter reaches C<0>, and second, it ensures that
562	C<send> is called even when C<no> hosts are being pinged (the loop	782	C<send> is called even when C<no> hosts are being pinged (the loop
563	doesn't execute once).	783	doesn't execute once).
564		784
565	This is the general pattern when you "fan out" into multiple subrequests:	785	This is the general pattern when you "fan out" into multiple (but
566	use an outer C<begin>/C<end> pair to set the callback and ensure C<end>	786	potentially zero) subrequests: use an outer C<begin>/C<end> pair to set
567	is called at least once, and then, for each subrequest you start, call	787	the callback and ensure C<end> is called at least once, and then, for each
568	C<begin> and for each subrequest you finish, call C<end>.	788	subrequest you start, call C<begin> and for each subrequest you finish,
		789	call C<end>.
569		790
570	=back	791	=back
571		792
572	=head3 METHODS FOR CONSUMERS	793	=head3 METHODS FOR CONSUMERS
573		794
…		…
577	=over 4	798	=over 4
578		799
579	=item $cv->recv	800	=item $cv->recv
580		801
581	Wait (blocking if necessary) until the C<< ->send >> or C<< ->croak	802	Wait (blocking if necessary) until the C<< ->send >> or C<< ->croak
582	>> methods have been called on c<$cv>, while servicing other watchers	803	>> methods have been called on C<$cv>, while servicing other watchers
583	normally.	804	normally.
584		805
585	You can only wait once on a condition - additional calls are valid but	806	You can only wait once on a condition - additional calls are valid but
586	will return immediately.	807	will return immediately.
587		808
…		…
589	function will call C<croak>.	810	function will call C<croak>.
590		811
591	In list context, all parameters passed to C<send> will be returned,	812	In list context, all parameters passed to C<send> will be returned,
592	in scalar context only the first one will be returned.	813	in scalar context only the first one will be returned.
593		814
		815	Note that doing a blocking wait in a callback is not supported by any
		816	event loop, that is, recursive invocation of a blocking C<< ->recv
		817	>> is not allowed, and the C<recv> call will C<croak> if such a
		818	condition is detected. This condition can be slightly loosened by using
		819	L<Coro::AnyEvent>, which allows you to do a blocking C<< ->recv >> from
		820	any thread that doesn't run the event loop itself.
		821
594	Not all event models support a blocking wait - some die in that case	822	Not all event models support a blocking wait - some die in that case
595	(programs might want to do that to stay interactive), so I<if you are	823	(programs might want to do that to stay interactive), so I<if you are
596	using this from a module, never require a blocking wait>, but let the	824	using this from a module, never require a blocking wait>. Instead, let the
597	caller decide whether the call will block or not (for example, by coupling	825	caller decide whether the call will block or not (for example, by coupling
598	condition variables with some kind of request results and supporting	826	condition variables with some kind of request results and supporting
599	callbacks so the caller knows that getting the result will not block,	827	callbacks so the caller knows that getting the result will not block,
600	while still supporting blocking waits if the caller so desires).	828	while still supporting blocking waits if the caller so desires).
601		829
602	Another reason I<never> to C<< ->recv >> in a module is that you cannot
603	sensibly have two C<< ->recv >>'s in parallel, as that would require
604	multiple interpreters or coroutines/threads, none of which C<AnyEvent>
605	can supply.
606
607	The L<Coro> module, however, I<can> and I<does> supply coroutines and, in
608	fact, L<Coro::AnyEvent> replaces AnyEvent's condvars by coroutine-safe
609	versions and also integrates coroutines into AnyEvent, making blocking
610	C<< ->recv >> calls perfectly safe as long as they are done from another
611	coroutine (one that doesn't run the event loop).
612
613	You can ensure that C<< -recv >> never blocks by setting a callback and	830	You can ensure that C<< ->recv >> never blocks by setting a callback and
614	only calling C<< ->recv >> from within that callback (or at a later	831	only calling C<< ->recv >> from within that callback (or at a later
615	time). This will work even when the event loop does not support blocking	832	time). This will work even when the event loop does not support blocking
616	waits otherwise.	833	waits otherwise.
617		834
618	=item $bool = $cv->ready	835	=item $bool = $cv->ready
…		…
624		841
625	This is a mutator function that returns the callback set and optionally	842	This is a mutator function that returns the callback set and optionally
626	replaces it before doing so.	843	replaces it before doing so.
627		844
628	The callback will be called when the condition becomes "true", i.e. when	845	The callback will be called when the condition becomes "true", i.e. when
629	C<send> or C<croak> are called, with the only argument being the condition	846	C<send> or C<croak> are called, with the only argument being the
630	variable itself. Calling C<recv> inside the callback or at any later time	847	condition variable itself. If the condition is already true, the
631	is guaranteed not to block.	848	callback is called immediately when it is set. Calling C<recv> inside
		849	the callback or at any later time is guaranteed not to block.
632		850
633	=back	851	=back
634		852
		853	=head1 SUPPORTED EVENT LOOPS/BACKENDS
		854
		855	The available backend classes are (every class has its own manpage):
		856
		857	=over 4
		858
		859	=item Backends that are autoprobed when no other event loop can be found.
		860
		861	EV is the preferred backend when no other event loop seems to be in
		862	use. If EV is not installed, then AnyEvent will fall back to its own
		863	pure-perl implementation, which is available everywhere as it comes with
		864	AnyEvent itself.
		865
		866	AnyEvent::Impl::EV based on EV (interface to libev, best choice).
		867	AnyEvent::Impl::Perl pure-perl AnyEvent::Loop, fast and portable.
		868
		869	=item Backends that are transparently being picked up when they are used.
		870
		871	These will be used if they are already loaded when the first watcher
		872	is created, in which case it is assumed that the application is using
		873	them. This means that AnyEvent will automatically pick the right backend
		874	when the main program loads an event module before anything starts to
		875	create watchers. Nothing special needs to be done by the main program.
		876
		877	AnyEvent::Impl::Event based on Event, very stable, few glitches.
		878	AnyEvent::Impl::Glib based on Glib, slow but very stable.
		879	AnyEvent::Impl::Tk based on Tk, very broken.
		880	AnyEvent::Impl::EventLib based on Event::Lib, leaks memory and worse.
		881	AnyEvent::Impl::POE based on POE, very slow, some limitations.
		882	AnyEvent::Impl::Irssi used when running within irssi.
		883	AnyEvent::Impl::IOAsync based on IO::Async.
		884	AnyEvent::Impl::Cocoa based on Cocoa::EventLoop.
		885	AnyEvent::Impl::FLTK based on FLTK (fltk 2 binding).
		886
		887	=item Backends with special needs.
		888
		889	Qt requires the Qt::Application to be instantiated first, but will
		890	otherwise be picked up automatically. As long as the main program
		891	instantiates the application before any AnyEvent watchers are created,
		892	everything should just work.
		893
		894	AnyEvent::Impl::Qt based on Qt.
		895
		896	=item Event loops that are indirectly supported via other backends.
		897
		898	Some event loops can be supported via other modules:
		899
		900	There is no direct support for WxWidgets (L<Wx>) or L<Prima>.
		901
		902	B<WxWidgets> has no support for watching file handles. However, you can
		903	use WxWidgets through the POE adaptor, as POE has a Wx backend that simply
		904	polls 20 times per second, which was considered to be too horrible to even
		905	consider for AnyEvent.
		906
		907	B<Prima> is not supported as nobody seems to be using it, but it has a POE
		908	backend, so it can be supported through POE.
		909
		910	AnyEvent knows about both L<Prima> and L<Wx>, however, and will try to
		911	load L<POE> when detecting them, in the hope that POE will pick them up,
		912	in which case everything will be automatic.
		913
		914	=back
		915
635	=head1 GLOBAL VARIABLES AND FUNCTIONS	916	=head1 GLOBAL VARIABLES AND FUNCTIONS
636		917
		918	These are not normally required to use AnyEvent, but can be useful to
		919	write AnyEvent extension modules.
		920
637	=over 4	921	=over 4
638		922
639	=item $AnyEvent::MODEL	923	=item $AnyEvent::MODEL
640		924
641	Contains C<undef> until the first watcher is being created. Then it	925	Contains C<undef> until the first watcher is being created, before the
		926	backend has been autodetected.
		927
642	contains the event model that is being used, which is the name of the	928	Afterwards it contains the event model that is being used, which is the
643	Perl class implementing the model. This class is usually one of the	929	name of the Perl class implementing the model. This class is usually one
644	C<AnyEvent::Impl:xxx> modules, but can be any other class in the case	930	of the C<AnyEvent::Impl::xxx> modules, but can be any other class in the
645	AnyEvent has been extended at runtime (e.g. in I<rxvt-unicode>).	931	case AnyEvent has been extended at runtime (e.g. in I<rxvt-unicode> it
646		932	will be C<urxvt::anyevent>).
647	The known classes so far are:
648
649	AnyEvent::Impl::EV based on EV (an interface to libev, best choice).
650	AnyEvent::Impl::Event based on Event, second best choice.
651	AnyEvent::Impl::Perl pure-perl implementation, fast and portable.
652	AnyEvent::Impl::Glib based on Glib, third-best choice.
653	AnyEvent::Impl::Tk based on Tk, very bad choice.
654	AnyEvent::Impl::Qt based on Qt, cannot be autoprobed (see its docs).
655	AnyEvent::Impl::EventLib based on Event::Lib, leaks memory and worse.
656	AnyEvent::Impl::POE based on POE, not generic enough for full support.
657
658	There is no support for WxWidgets, as WxWidgets has no support for
659	watching file handles. However, you can use WxWidgets through the
660	POE Adaptor, as POE has a Wx backend that simply polls 20 times per
661	second, which was considered to be too horrible to even consider for
662	AnyEvent. Likewise, other POE backends can be used by AnyEvent by using
663	it's adaptor.
664
665	AnyEvent knows about L<Prima> and L<Wx> and will try to use L<POE> when
666	autodetecting them.
667		933
668	=item AnyEvent::detect	934	=item AnyEvent::detect
669		935
670	Returns C<$AnyEvent::MODEL>, forcing autodetection of the event model	936	Returns C<$AnyEvent::MODEL>, forcing autodetection of the event model
671	if necessary. You should only call this function right before you would	937	if necessary. You should only call this function right before you would
672	have created an AnyEvent watcher anyway, that is, as late as possible at	938	have created an AnyEvent watcher anyway, that is, as late as possible at
673	runtime.	939	runtime, and not e.g. during initialisation of your module.
		940
		941	The effect of calling this function is as if a watcher had been created
		942	(specifically, actions that happen "when the first watcher is created"
		943	happen when calling detetc as well).
		944
		945	If you need to do some initialisation before AnyEvent watchers are
		946	created, use C<post_detect>.
674		947
675	=item $guard = AnyEvent::post_detect { BLOCK }	948	=item $guard = AnyEvent::post_detect { BLOCK }
676		949
677	Arranges for the code block to be executed as soon as the event model is	950	Arranges for the code block to be executed as soon as the event model is
678	autodetected (or immediately if this has already happened).	951	autodetected (or immediately if that has already happened).
		952
		953	The block will be executed I<after> the actual backend has been detected
		954	(C<$AnyEvent::MODEL> is set), but I<before> any watchers have been
		955	created, so it is possible to e.g. patch C<@AnyEvent::ISA> or do
		956	other initialisations - see the sources of L<AnyEvent::Strict> or
		957	L<AnyEvent::AIO> to see how this is used.
		958
		959	The most common usage is to create some global watchers, without forcing
		960	event module detection too early, for example, L<AnyEvent::AIO> creates
		961	and installs the global L<IO::AIO> watcher in a C<post_detect> block to
		962	avoid autodetecting the event module at load time.
679		963
680	If called in scalar or list context, then it creates and returns an object	964	If called in scalar or list context, then it creates and returns an object
681	that automatically removes the callback again when it is destroyed. See	965	that automatically removes the callback again when it is destroyed (or
		966	C<undef> when the hook was immediately executed). See L<AnyEvent::AIO> for
682	L<Coro::BDB> for a case where this is useful.	967	a case where this is useful.
		968
		969	Example: Create a watcher for the IO::AIO module and store it in
		970	C<$WATCHER>, but do so only do so after the event loop is initialised.
		971
		972	our WATCHER;
		973
		974	my $guard = AnyEvent::post_detect {
		975	$WATCHER = AnyEvent->io (fh => IO::AIO::poll_fileno, poll => 'r', cb => \&IO::AIO::poll_cb);
		976	};
		977
		978	# the ||= is important in case post_detect immediately runs the block,
		979	# as to not clobber the newly-created watcher. assigning both watcher and
		980	# post_detect guard to the same variable has the advantage of users being
		981	# able to just C<undef $WATCHER> if the watcher causes them grief.
		982
		983	$WATCHER ||= $guard;
683		984
684	=item @AnyEvent::post_detect	985	=item @AnyEvent::post_detect
685		986
686	If there are any code references in this array (you can C<push> to it	987	If there are any code references in this array (you can C<push> to it
687	before or after loading AnyEvent), then they will called directly after	988	before or after loading AnyEvent), then they will be called directly
688	the event loop has been chosen.	989	after the event loop has been chosen.
689		990
690	You should check C<$AnyEvent::MODEL> before adding to this array, though:	991	You should check C<$AnyEvent::MODEL> before adding to this array, though:
691	if it contains a true value then the event loop has already been detected,	992	if it is defined then the event loop has already been detected, and the
692	and the array will be ignored.	993	array will be ignored.
693		994
694	Best use C<AnyEvent::post_detect { BLOCK }> instead.	995	Best use C<AnyEvent::post_detect { BLOCK }> when your application allows
		996	it, as it takes care of these details.
		997
		998	This variable is mainly useful for modules that can do something useful
		999	when AnyEvent is used and thus want to know when it is initialised, but do
		1000	not need to even load it by default. This array provides the means to hook
		1001	into AnyEvent passively, without loading it.
		1002
		1003	Example: To load Coro::AnyEvent whenever Coro and AnyEvent are used
		1004	together, you could put this into Coro (this is the actual code used by
		1005	Coro to accomplish this):
		1006
		1007	if (defined $AnyEvent::MODEL) {
		1008	# AnyEvent already initialised, so load Coro::AnyEvent
		1009	require Coro::AnyEvent;
		1010	} else {
		1011	# AnyEvent not yet initialised, so make sure to load Coro::AnyEvent
		1012	# as soon as it is
		1013	push @AnyEvent::post_detect, sub { require Coro::AnyEvent };
		1014	}
		1015
		1016	=item AnyEvent::postpone { BLOCK }
		1017
		1018	Arranges for the block to be executed as soon as possible, but not before
		1019	the call itself returns. In practise, the block will be executed just
		1020	before the event loop polls for new events, or shortly afterwards.
		1021
		1022	This function never returns anything (to make the C<return postpone { ...
		1023	}> idiom more useful.
		1024
		1025	To understand the usefulness of this function, consider a function that
		1026	asynchronously does something for you and returns some transaction
		1027	object or guard to let you cancel the operation. For example,
		1028	C<AnyEvent::Socket::tcp_connect>:
		1029
		1030	# start a conenction attempt unless one is active
		1031	$self->{connect_guard} ||= AnyEvent::Socket::tcp_connect "www.example.net", 80, sub {
		1032	delete $self->{connect_guard};
		1033	...
		1034	};
		1035
		1036	Imagine that this function could instantly call the callback, for
		1037	example, because it detects an obvious error such as a negative port
		1038	number. Invoking the callback before the function returns causes problems
		1039	however: the callback will be called and will try to delete the guard
		1040	object. But since the function hasn't returned yet, there is nothing to
		1041	delete. When the function eventually returns it will assign the guard
		1042	object to C<< $self->{connect_guard} >>, where it will likely never be
		1043	deleted, so the program thinks it is still trying to connect.
		1044
		1045	This is where C<AnyEvent::postpone> should be used. Instead of calling the
		1046	callback directly on error:
		1047
		1048	$cb->(undef), return # signal error to callback, BAD!
		1049	if $some_error_condition;
		1050
		1051	It should use C<postpone>:
		1052
		1053	AnyEvent::postpone { $cb->(undef) }, return # signal error to callback, later
		1054	if $some_error_condition;
		1055
		1056	=item AnyEvent::log $level, $msg[, @args]
		1057
		1058	Log the given C<$msg> at the given C<$level>.
		1059
		1060	If L<AnyEvent::Log> is not loaded then this function makes a simple test
		1061	to see whether the message will be logged. If the test succeeds it will
		1062	load AnyEvent::Log and call C<AnyEvent::Log::log> - consequently, look at
		1063	the L<AnyEvent::Log> documentation for details.
		1064
		1065	If the test fails it will simply return. Right now this happens when a
		1066	numerical loglevel is used and it is larger than the level specified via
		1067	C<$ENV{PERL_ANYEVENT_VERBOSE}>.
		1068
		1069	If you want to sprinkle loads of logging calls around your code, consider
		1070	creating a logger callback with the C<AnyEvent::Log::logger> function,
		1071	which can reduce typing, codesize and can reduce the logging overhead
		1072	enourmously.
695		1073
696	=back	1074	=back
697		1075
698	=head1 WHAT TO DO IN A MODULE	1076	=head1 WHAT TO DO IN A MODULE
699		1077
…		…
710	because it will stall the whole program, and the whole point of using	1088	because it will stall the whole program, and the whole point of using
711	events is to stay interactive.	1089	events is to stay interactive.
712		1090
713	It is fine, however, to call C<< ->recv >> when the user of your module	1091	It is fine, however, to call C<< ->recv >> when the user of your module
714	requests it (i.e. if you create a http request object ad have a method	1092	requests it (i.e. if you create a http request object ad have a method
715	called C<results> that returns the results, it should call C<< ->recv >>	1093	called C<results> that returns the results, it may call C<< ->recv >>
716	freely, as the user of your module knows what she is doing. always).	1094	freely, as the user of your module knows what she is doing. Always).
717		1095
718	=head1 WHAT TO DO IN THE MAIN PROGRAM	1096	=head1 WHAT TO DO IN THE MAIN PROGRAM
719		1097
720	There will always be a single main program - the only place that should	1098	There will always be a single main program - the only place that should
721	dictate which event model to use.	1099	dictate which event model to use.
722		1100
723	If it doesn't care, it can just "use AnyEvent" and use it itself, or not	1101	If the program is not event-based, it need not do anything special, even
724	do anything special (it does not need to be event-based) and let AnyEvent	1102	when it depends on a module that uses an AnyEvent. If the program itself
725	decide which implementation to chose if some module relies on it.	1103	uses AnyEvent, but does not care which event loop is used, all it needs
		1104	to do is C<use AnyEvent>. In either case, AnyEvent will choose the best
		1105	available loop implementation.
726		1106
727	If the main program relies on a specific event model - for example, in	1107	If the main program relies on a specific event model - for example, in
728	Gtk2 programs you have to rely on the Glib module - you should load the	1108	Gtk2 programs you have to rely on the Glib module - you should load the
729	event module before loading AnyEvent or any module that uses it: generally	1109	event module before loading AnyEvent or any module that uses it: generally
730	speaking, you should load it as early as possible. The reason is that	1110	speaking, you should load it as early as possible. The reason is that
731	modules might create watchers when they are loaded, and AnyEvent will	1111	modules might create watchers when they are loaded, and AnyEvent will
732	decide on the event model to use as soon as it creates watchers, and it	1112	decide on the event model to use as soon as it creates watchers, and it
733	might chose the wrong one unless you load the correct one yourself.	1113	might choose the wrong one unless you load the correct one yourself.
734		1114
735	You can chose to use a pure-perl implementation by loading the	1115	You can chose to use a pure-perl implementation by loading the
736	C<AnyEvent::Impl::Perl> module, which gives you similar behaviour	1116	C<AnyEvent::Loop> module, which gives you similar behaviour
737	everywhere, but letting AnyEvent chose the model is generally better.	1117	everywhere, but letting AnyEvent chose the model is generally better.
738		1118
739	=head2 MAINLOOP EMULATION	1119	=head2 MAINLOOP EMULATION
740		1120
741	Sometimes (often for short test scripts, or even standalone programs who	1121	Sometimes (often for short test scripts, or even standalone programs who
…		…
754		1134
755		1135
756	=head1 OTHER MODULES	1136	=head1 OTHER MODULES
757		1137
758	The following is a non-exhaustive list of additional modules that use	1138	The following is a non-exhaustive list of additional modules that use
759	AnyEvent and can therefore be mixed easily with other AnyEvent modules	1139	AnyEvent as a client and can therefore be mixed easily with other
760	in the same program. Some of the modules come with AnyEvent, some are	1140	AnyEvent modules and other event loops in the same program. Some of the
761	available via CPAN.	1141	modules come as part of AnyEvent, the others are available via CPAN (see
		1142	L<http://search.cpan.org/search?m=module&q=anyevent%3A%3A*> for
		1143	a longer non-exhaustive list), and the list is heavily biased towards
		1144	modules of the AnyEvent author himself :)
762		1145
763	=over 4	1146	=over 4
764		1147
765	=item L<AnyEvent::Util>	1148	=item L<AnyEvent::Util>
766		1149
767	Contains various utility functions that replace often-used but blocking	1150	Contains various utility functions that replace often-used blocking
768	functions such as C<inet_aton> by event-/callback-based versions.	1151	functions such as C<inet_aton> with event/callback-based versions.
769		1152
770	=item L<AnyEvent::Socket>	1153	=item L<AnyEvent::Socket>
771		1154
772	Provides various utility functions for (internet protocol) sockets,	1155	Provides various utility functions for (internet protocol) sockets,
773	addresses and name resolution. Also functions to create non-blocking tcp	1156	addresses and name resolution. Also functions to create non-blocking tcp
…		…
775		1158
776	=item L<AnyEvent::Handle>	1159	=item L<AnyEvent::Handle>
777		1160
778	Provide read and write buffers, manages watchers for reads and writes,	1161	Provide read and write buffers, manages watchers for reads and writes,
779	supports raw and formatted I/O, I/O queued and fully transparent and	1162	supports raw and formatted I/O, I/O queued and fully transparent and
780	non-blocking SSL/TLS.	1163	non-blocking SSL/TLS (via L<AnyEvent::TLS>).
781		1164
782	=item L<AnyEvent::DNS>	1165	=item L<AnyEvent::DNS>
783		1166
784	Provides rich asynchronous DNS resolver capabilities.	1167	Provides rich asynchronous DNS resolver capabilities.
785		1168
		1169	=item L<AnyEvent::HTTP>, L<AnyEvent::IRC>, L<AnyEvent::XMPP>, L<AnyEvent::GPSD>, L<AnyEvent::IGS>, L<AnyEvent::FCP>
		1170
		1171	Implement event-based interfaces to the protocols of the same name (for
		1172	the curious, IGS is the International Go Server and FCP is the Freenet
		1173	Client Protocol).
		1174
786	=item L<AnyEvent::HTTP>	1175	=item L<AnyEvent::AIO>
787		1176
788	A simple-to-use HTTP library that is capable of making a lot of concurrent	1177	Truly asynchronous (as opposed to non-blocking) I/O, should be in the
789	HTTP requests.	1178	toolbox of every event programmer. AnyEvent::AIO transparently fuses
		1179	L<IO::AIO> and AnyEvent together, giving AnyEvent access to event-based
		1180	file I/O, and much more.
		1181
		1182	=item L<AnyEvent::Filesys::Notify>
		1183
		1184	AnyEvent is good for non-blocking stuff, but it can't detect file or
		1185	path changes (e.g. "watch this directory for new files", "watch this
		1186	file for changes"). The L<AnyEvent::Filesys::Notify> module promises to
		1187	do just that in a portbale fashion, supporting inotify on GNU/Linux and
		1188	some weird, without doubt broken, stuff on OS X to monitor files. It can
		1189	fall back to blocking scans at regular intervals transparently on other
		1190	platforms, so it's about as portable as it gets.
		1191
		1192	(I haven't used it myself, but I haven't heard anybody complaining about
		1193	it yet).
		1194
		1195	=item L<AnyEvent::DBI>
		1196
		1197	Executes L<DBI> requests asynchronously in a proxy process for you,
		1198	notifying you in an event-based way when the operation is finished.
790		1199
791	=item L<AnyEvent::HTTPD>	1200	=item L<AnyEvent::HTTPD>
792		1201
793	Provides a simple web application server framework.	1202	A simple embedded webserver.
794		1203
795	=item L<AnyEvent::FastPing>	1204	=item L<AnyEvent::FastPing>
796		1205
797	The fastest ping in the west.	1206	The fastest ping in the west.
798		1207
799	=item L<AnyEvent::DBI>
800
801	Executes L<DBI> requests asynchronously in a proxy process.
802
803	=item L<AnyEvent::AIO>
804
805	Truly asynchronous I/O, should be in the toolbox of every event
806	programmer. AnyEvent::AIO transparently fuses L<IO::AIO> and AnyEvent
807	together.
808
809	=item L<AnyEvent::BDB>
810
811	Truly asynchronous Berkeley DB access. AnyEvent::BDB transparently fuses
812	L<BDB> and AnyEvent together.
813
814	=item L<AnyEvent::GPSD>
815
816	A non-blocking interface to gpsd, a daemon delivering GPS information.
817
818	=item L<AnyEvent::IGS>
819
820	A non-blocking interface to the Internet Go Server protocol (used by
821	L<App::IGS>).
822
823	=item L<Net::IRC3>
824
825	AnyEvent based IRC client module family.
826
827	=item L<Net::XMPP2>
828
829	AnyEvent based XMPP (Jabber protocol) module family.
830
831	=item L<Net::FCP>
832
833	AnyEvent-based implementation of the Freenet Client Protocol, birthplace
834	of AnyEvent.
835
836	=item L<Event::ExecFlow>
837
838	High level API for event-based execution flow control.
839
840	=item L<Coro>	1208	=item L<Coro>
841		1209
842	Has special support for AnyEvent via L<Coro::AnyEvent>.	1210	Has special support for AnyEvent via L<Coro::AnyEvent>, which allows you
		1211	to simply invert the flow control - don't call us, we will call you:
843		1212
844	=item L<IO::Lambda>	1213	async {
		1214	Coro::AnyEvent::sleep 5; # creates a 5s timer and waits for it
		1215	print "5 seconds later!\n";
845		1216
846	The lambda approach to I/O - don't ask, look there. Can use AnyEvent.	1217	Coro::AnyEvent::readable *STDIN; # uses an I/O watcher
		1218	my $line = <STDIN>; # works for ttys
		1219
		1220	AnyEvent::HTTP::http_get "url", Coro::rouse_cb;
		1221	my ($body, $hdr) = Coro::rouse_wait;
		1222	};
847		1223
848	=back	1224	=back
849		1225
850	=cut	1226	=cut
851		1227
852	package AnyEvent;	1228	package AnyEvent;
853		1229
854	no warnings;	1230	# basically a tuned-down version of common::sense
855	use strict;	1231	sub common_sense {
		1232	# from common:.sense 3.5
		1233	local $^W;
		1234	${^WARNING_BITS} ^= ${^WARNING_BITS} ^ "\x3c\x3f\x33\x00\x0f\xf0\x0f\xc0\xf0\xfc\x33\x00";
		1235	# use strict vars subs - NO UTF-8, as Util.pm doesn't like this atm. (uts46data.pl)
		1236	$^H |= 0x00000600;
		1237	}
856		1238
		1239	BEGIN { AnyEvent::common_sense }
		1240
857	use Carp;	1241	use Carp ();
858		1242
859	our $VERSION = 4.231;	1243	our $VERSION = '7.04';
860	our $MODEL;	1244	our $MODEL;
861
862	our $AUTOLOAD;
863	our @ISA;	1245	our @ISA;
864
865	our @REGISTRY;	1246	our @REGISTRY;
866		1247	our $VERBOSE;
867	our $WIN32;	1248	our %PROTOCOL; # (ipv4|ipv6) => (1|2), higher numbers are preferred
		1249	our $MAX_SIGNAL_LATENCY = $ENV{PERL_ANYEVENT_MAX_SIGNAL_LATENCY} || 10; # executes after the BEGIN block below (tainting!)
868		1250
869	BEGIN {	1251	BEGIN {
870	my $win32 = ! ! ($^O =~ /mswin32/i);	1252	require "AnyEvent/constants.pl";
871	eval "sub WIN32(){ $win32 }";
872	}
873		1253
874	our $verbose = $ENV{PERL_ANYEVENT_VERBOSE}*1;	1254	eval "sub TAINT (){" . (${^TAINT}*1) . "}";
875		1255
876	our %PROTOCOL; # (ipv4|ipv6) => (1|2), higher numbers are preferred	1256	delete @ENV{grep /^PERL_ANYEVENT_/, keys %ENV}
		1257	if ${^TAINT};
877		1258
878	{	1259	$ENV{"PERL_ANYEVENT_$_"} = $ENV{"AE_$_"}
		1260	for grep s/^AE_// && !exists $ENV{"PERL_ANYEVENT_$_"}, keys %ENV;
		1261
		1262	@ENV{grep /^PERL_ANYEVENT_/, keys %ENV} = ()
		1263	if ${^TAINT};
		1264
		1265	# $ENV{PERL_ANYEVENT_xxx} now valid
		1266
		1267	$VERBOSE = length $ENV{PERL_ANYEVENT_VERBOSE} ? $ENV{PERL_ANYEVENT_VERBOSE}*1 : 4;
		1268
879	my $idx;	1269	my $idx;
880	$PROTOCOL{$_} = ++$idx	1270	$PROTOCOL{$_} = ++$idx
881	for reverse split /\s*,\s*/,	1271	for reverse split /\s*,\s*/,
882	$ENV{PERL_ANYEVENT_PROTOCOLS} || "ipv4,ipv6";	1272	$ENV{PERL_ANYEVENT_PROTOCOLS} || "ipv4,ipv6";
883	}	1273	}
884		1274
		1275	our @post_detect;
		1276
		1277	sub post_detect(&) {
		1278	my ($cb) = @_;
		1279
		1280	push @post_detect, $cb;
		1281
		1282	defined wantarray
		1283	? bless \$cb, "AnyEvent::Util::postdetect"
		1284	: ()
		1285	}
		1286
		1287	sub AnyEvent::Util::postdetect::DESTROY {
		1288	@post_detect = grep $_ != ${$_[0]}, @post_detect;
		1289	}
		1290
		1291	our $POSTPONE_W;
		1292	our @POSTPONE;
		1293
		1294	sub _postpone_exec {
		1295	undef $POSTPONE_W;
		1296
		1297	&{ shift @POSTPONE }
		1298	while @POSTPONE;
		1299	}
		1300
		1301	sub postpone(&) {
		1302	push @POSTPONE, shift;
		1303
		1304	$POSTPONE_W ||= AE::timer (0, 0, \&_postpone_exec);
		1305
		1306	()
		1307	}
		1308
		1309	sub log($$;@) {
		1310	# only load the big bloated module when we actually are about to log something
		1311	if ($_[0] <= ($VERBOSE || 1)) { # also catches non-numeric levels(!) and fatal
		1312	local ($!, $@);
		1313	require AnyEvent::Log; # among other things, sets $VERBOSE to 9
		1314	# AnyEvent::Log overwrites this function
		1315	goto &log;
		1316	}
		1317
		1318	0 # not logged
		1319	}
		1320
		1321	sub _logger($;$) {
		1322	my ($level, $renabled) = @_;
		1323
		1324	$$renabled = $level <= $VERBOSE;
		1325
		1326	my $logger = [(caller)[0], $level, $renabled];
		1327
		1328	$AnyEvent::Log::LOGGER{$logger+0} = $logger;
		1329
		1330	# return unless defined wantarray;
		1331	#
		1332	# require AnyEvent::Util;
		1333	# my $guard = AnyEvent::Util::guard (sub {
		1334	# # "clean up"
		1335	# delete $LOGGER{$logger+0};
		1336	# });
		1337	#
		1338	# sub {
		1339	# return 0 unless $$renabled;
		1340	#
		1341	# $guard if 0; # keep guard alive, but don't cause runtime overhead
		1342	# require AnyEvent::Log unless $AnyEvent::Log::VERSION;
		1343	# package AnyEvent::Log;
		1344	# _log ($logger->[0], $level, @_) # logger->[0] has been converted at load time
		1345	# }
		1346	}
		1347
		1348	if (length $ENV{PERL_ANYEVENT_LOG}) {
		1349	require AnyEvent::Log; # AnyEvent::Log does the thing for us
		1350	}
		1351
885	my @models = (	1352	our @models = (
886	[EV:: => AnyEvent::Impl::EV::],	1353	[EV:: => AnyEvent::Impl::EV::],
887	[Event:: => AnyEvent::Impl::Event::],
888	[AnyEvent::Impl::Perl:: => AnyEvent::Impl::Perl::],	1354	[AnyEvent::Loop:: => AnyEvent::Impl::Perl::],
889	# everything below here will not be autoprobed	1355	# everything below here will not (normally) be autoprobed
890	# as the pureperl backend should work everywhere	1356	# as the pure perl backend should work everywhere
891	# and is usually faster	1357	# and is usually faster
		1358	[Irssi:: => AnyEvent::Impl::Irssi::], # Irssi has a bogus "Event" package, so msut be near the top
		1359	[Event:: => AnyEvent::Impl::Event::], # slow, stable
		1360	[Glib:: => AnyEvent::Impl::Glib::], # becomes extremely slow with many watchers
		1361	# everything below here should not be autoloaded
		1362	[Event::Lib:: => AnyEvent::Impl::EventLib::], # too buggy
892	[Tk:: => AnyEvent::Impl::Tk::], # crashes with many handles	1363	[Tk:: => AnyEvent::Impl::Tk::], # crashes with many handles
893	[Glib:: => AnyEvent::Impl::Glib::], # becomes extremely slow with many watchers
894	[Event::Lib:: => AnyEvent::Impl::EventLib::], # too buggy
895	[Qt:: => AnyEvent::Impl::Qt::], # requires special main program	1364	[Qt:: => AnyEvent::Impl::Qt::], # requires special main program
896	[POE::Kernel:: => AnyEvent::Impl::POE::], # lasciate ogni speranza	1365	[POE::Kernel:: => AnyEvent::Impl::POE::], # lasciate ogni speranza
897	[Wx:: => AnyEvent::Impl::POE::],	1366	[Wx:: => AnyEvent::Impl::POE::],
898	[Prima:: => AnyEvent::Impl::POE::],	1367	[Prima:: => AnyEvent::Impl::POE::],
		1368	[IO::Async::Loop:: => AnyEvent::Impl::IOAsync::], # a bitch to autodetect
		1369	[Cocoa::EventLoop:: => AnyEvent::Impl::Cocoa::],
		1370	[FLTK:: => AnyEvent::Impl::FLTK::],
899	);	1371	);
900		1372
901	our %method = map +($_ => 1), qw(io timer time now signal child condvar one_event DESTROY);	1373	our @isa_hook;
902		1374
903	our @post_detect;	1375	sub _isa_set {
		1376	my @pkg = ("AnyEvent", (map $_->[0], grep defined, @isa_hook), $MODEL);
904		1377
		1378	@{"$pkg[$_-1]::ISA"} = $pkg[$_]
		1379	for 1 .. $#pkg;
		1380
		1381	grep $_ && $_->[1], @isa_hook
		1382	and AE::_reset ();
		1383	}
		1384
		1385	# used for hooking AnyEvent::Strict and AnyEvent::Debug::Wrap into the class hierarchy
		1386	sub _isa_hook($$;$) {
		1387	my ($i, $pkg, $reset_ae) = @_;
		1388
		1389	$isa_hook[$i] = $pkg ? [$pkg, $reset_ae] : undef;
		1390
		1391	_isa_set;
		1392	}
		1393
		1394	# all autoloaded methods reserve the complete glob, not just the method slot.
		1395	# due to bugs in perls method cache implementation.
		1396	our @methods = qw(io timer time now now_update signal child idle condvar);
		1397
905	sub post_detect(&) {	1398	sub detect() {
906	my ($cb) = @_;	1399	return $MODEL if $MODEL; # some programs keep references to detect
907		1400
908	if ($MODEL) {	1401	# IO::Async::Loop::AnyEvent is extremely evil, refuse to work with it
909	$cb->();	1402	# the author knows about the problems and what it does to AnyEvent as a whole
		1403	# (and the ability of others to use AnyEvent), but simply wants to abuse AnyEvent
		1404	# anyway.
		1405	AnyEvent::log fatal => "IO::Async::Loop::AnyEvent detected - that module is broken by\n"
		1406	. "design, abuses internals and breaks AnyEvent - will not continue."
		1407	if exists $INC{"IO/Async/Loop/AnyEvent.pm"};
910		1408
911	1	1409	local $!; # for good measure
		1410	local $SIG{__DIE__}; # we use eval
		1411
		1412	# free some memory
		1413	*detect = sub () { $MODEL };
		1414	# undef &func doesn't correctly update the method cache. grmbl.
		1415	# so we delete the whole glob. grmbl.
		1416	# otoh, perl doesn't let me undef an active usb, but it lets me free
		1417	# a glob with an active sub. hrm. i hope it works, but perl is
		1418	# usually buggy in this department. sigh.
		1419	delete @{"AnyEvent::"}{@methods};
		1420	undef @methods;
		1421
		1422	if ($ENV{PERL_ANYEVENT_MODEL} =~ /^([a-zA-Z0-9:]+)$/) {
		1423	my $model = $1;
		1424	$model = "AnyEvent::Impl::$model" unless $model =~ s/::$//;
		1425	if (eval "require $model") {
		1426	AnyEvent::log 7 => "Loaded model '$model' (forced by \$ENV{PERL_ANYEVENT_MODEL}), using it.";
		1427	$MODEL = $model;
912	} else {	1428	} else {
913	push @post_detect, $cb;	1429	AnyEvent::log 4 => "Unable to load model '$model' (from \$ENV{PERL_ANYEVENT_MODEL}):\n$@";
914		1430	}
915	defined wantarray
916	? bless \$cb, "AnyEvent::Util::PostDetect"
917	: ()
918	}	1431	}
919	}
920		1432
921	sub AnyEvent::Util::PostDetect::DESTROY {	1433	# check for already loaded models
922	@post_detect = grep $_ != ${$_[0]}, @post_detect;
923	}
924
925	sub detect() {
926	unless ($MODEL) {	1434	unless ($MODEL) {
927	no strict 'refs';	1435	for (@REGISTRY, @models) {
928	local $SIG{__DIE__};	1436	my ($package, $model) = @$_;
929		1437	if (${"$package\::VERSION"} > 0) {
930	if ($ENV{PERL_ANYEVENT_MODEL} =~ /^([a-zA-Z]+)$/) {
931	my $model = "AnyEvent::Impl::$1";
932	if (eval "require $model") {	1438	if (eval "require $model") {
		1439	AnyEvent::log 7 => "Autodetected model '$model', using it.";
933	$MODEL = $model;	1440	$MODEL = $model;
934	warn "AnyEvent: loaded model '$model' (forced by \$PERL_ANYEVENT_MODEL), using it.\n" if $verbose > 1;	1441	last;
935	} else {	1442	} else {
936	warn "AnyEvent: unable to load model '$model' (from \$PERL_ANYEVENT_MODEL):\n$@" if $verbose;	1443	AnyEvent::log 8 => "Detected event loop $package, but cannot load '$model', skipping: $@";
		1444	}
937	}	1445	}
938	}	1446	}
939		1447
940	# check for already loaded models
941	unless ($MODEL) {	1448	unless ($MODEL) {
		1449	# try to autoload a model
942	for (@REGISTRY, @models) {	1450	for (@REGISTRY, @models) {
943	my ($package, $model) = @$_;	1451	my ($package, $model) = @$_;
		1452	if (
		1453	eval "require $package"
944	if (${"$package\::VERSION"} > 0) {	1454	and ${"$package\::VERSION"} > 0
945	if (eval "require $model") {	1455	and eval "require $model"
		1456	) {
		1457	AnyEvent::log 7 => "Autoloaded model '$model', using it.";
946	$MODEL = $model;	1458	$MODEL = $model;
947	warn "AnyEvent: autodetected model '$model', using it.\n" if $verbose > 1;
948	last;	1459	last;
949	}
950	}	1460	}
951	}	1461	}
952		1462
953	unless ($MODEL) {
954	# try to load a model
955
956	for (@REGISTRY, @models) {
957	my ($package, $model) = @$_;
958	if (eval "require $package"
959	and ${"$package\::VERSION"} > 0
960	and eval "require $model") {
961	$MODEL = $model;
962	warn "AnyEvent: autoprobed model '$model', using it.\n" if $verbose > 1;
963	last;
964	}
965	}
966
967	$MODEL	1463	$MODEL
968	or die "No event module selected for AnyEvent and autodetect failed. Install any one of these modules: EV, Event or Glib.";	1464	or AnyEvent::log fatal => "Backend autodetection failed - did you properly install AnyEvent?";
969	}
970	}	1465	}
971
972	push @{"$MODEL\::ISA"}, "AnyEvent::Base";
973
974	unshift @ISA, $MODEL;
975
976	require AnyEvent::Strict if $ENV{PERL_ANYEVENT_STRICT};
977
978	(shift @post_detect)->() while @post_detect;
979	}	1466	}
980		1467
		1468	# free memory only needed for probing
		1469	undef @models;
		1470	undef @REGISTRY;
		1471
		1472	push @{"$MODEL\::ISA"}, "AnyEvent::Base";
		1473
		1474	# now nuke some methods that are overridden by the backend.
		1475	# SUPER usage is not allowed in these.
		1476	for (qw(time signal child idle)) {
		1477	undef &{"AnyEvent::Base::$_"}
		1478	if defined &{"$MODEL\::$_"};
		1479	}
		1480
		1481	_isa_set;
		1482
		1483	# we're officially open!
		1484
		1485	if ($ENV{PERL_ANYEVENT_STRICT}) {
		1486	require AnyEvent::Strict;
		1487	}
		1488
		1489	if ($ENV{PERL_ANYEVENT_DEBUG_WRAP}) {
		1490	require AnyEvent::Debug;
		1491	AnyEvent::Debug::wrap ($ENV{PERL_ANYEVENT_DEBUG_WRAP});
		1492	}
		1493
		1494	if (length $ENV{PERL_ANYEVENT_DEBUG_SHELL}) {
		1495	require AnyEvent::Socket;
		1496	require AnyEvent::Debug;
		1497
		1498	my $shell = $ENV{PERL_ANYEVENT_DEBUG_SHELL};
		1499	$shell =~ s/\$\$/$$/g;
		1500
		1501	my ($host, $service) = AnyEvent::Socket::parse_hostport ($shell);
		1502	$AnyEvent::Debug::SHELL = AnyEvent::Debug::shell ($host, $service);
		1503	}
		1504
		1505	# now the anyevent environment is set up as the user told us to, so
		1506	# call the actual user code - post detects
		1507
		1508	(shift @post_detect)->() while @post_detect;
		1509	undef @post_detect;
		1510
		1511	*post_detect = sub(&) {
		1512	shift->();
		1513
		1514	undef
		1515	};
		1516
981	$MODEL	1517	$MODEL
982	}	1518	}
983		1519
984	sub AUTOLOAD {	1520	for my $name (@methods) {
985	(my $func = $AUTOLOAD) =~ s/.*://;	1521	*$name = sub {
986		1522	detect;
987	$method{$func}	1523	# we use goto because
988	or croak "$func: not a valid method for AnyEvent objects";	1524	# a) it makes the thunk more transparent
989		1525	# b) it allows us to delete the thunk later
990	detect unless $MODEL;	1526	goto &{ UNIVERSAL::can AnyEvent => "SUPER::$name" }
991		1527	};
992	my $class = shift;
993	$class->$func (@_);
994	}	1528	}
995		1529
996	# utility function to dup a filehandle. this is used by many backends	1530	# utility function to dup a filehandle. this is used by many backends
997	# to support binding more than one watcher per filehandle (they usually	1531	# to support binding more than one watcher per filehandle (they usually
998	# allow only one watcher per fd, so we dup it to get a different one).	1532	# allow only one watcher per fd, so we dup it to get a different one).
999	sub _dupfh($$$$) {	1533	sub _dupfh($$;$$) {
1000	my ($poll, $fh, $r, $w) = @_;	1534	my ($poll, $fh, $r, $w) = @_;
1001		1535
1002	require Fcntl;
1003
1004	# cygwin requires the fh mode to be matching, unix doesn't	1536	# cygwin requires the fh mode to be matching, unix doesn't
1005	my ($rw, $mode) = $poll eq "r" ? ($r, "<")	1537	my ($rw, $mode) = $poll eq "r" ? ($r, "<&") : ($w, ">&");
1006	: $poll eq "w" ? ($w, ">")
1007	: Carp::croak "AnyEvent->io requires poll set to either 'r' or 'w'";
1008		1538
1009	open my $fh2, "$mode&" . fileno $fh	1539	open my $fh2, $mode, $fh
1010	or die "cannot dup() filehandle: $!";	1540	or die "AnyEvent->io: cannot dup() filehandle in mode '$poll': $!,";
1011		1541
1012	# we assume CLOEXEC is already set by perl in all important cases	1542	# we assume CLOEXEC is already set by perl in all important cases
1013		1543
1014	($fh2, $rw)	1544	($fh2, $rw)
1015	}	1545	}
1016		1546
		1547	=head1 SIMPLIFIED AE API
		1548
		1549	Starting with version 5.0, AnyEvent officially supports a second, much
		1550	simpler, API that is designed to reduce the calling, typing and memory
		1551	overhead by using function call syntax and a fixed number of parameters.
		1552
		1553	See the L<AE> manpage for details.
		1554
		1555	=cut
		1556
		1557	package AE;
		1558
		1559	our $VERSION = $AnyEvent::VERSION;
		1560
		1561	sub _reset() {
		1562	eval q{
		1563	# fall back to the main API by default - backends and AnyEvent::Base
		1564	# implementations can overwrite these.
		1565
		1566	sub io($$$) {
		1567	AnyEvent->io (fh => $_[0], poll => $_[1] ? "w" : "r", cb => $_[2])
		1568	}
		1569
		1570	sub timer($$$) {
		1571	AnyEvent->timer (after => $_[0], interval => $_[1], cb => $_[2])
		1572	}
		1573
		1574	sub signal($$) {
		1575	AnyEvent->signal (signal => $_[0], cb => $_[1])
		1576	}
		1577
		1578	sub child($$) {
		1579	AnyEvent->child (pid => $_[0], cb => $_[1])
		1580	}
		1581
		1582	sub idle($) {
		1583	AnyEvent->idle (cb => $_[0]);
		1584	}
		1585
		1586	sub cv(;&) {
		1587	AnyEvent->condvar (@_ ? (cb => $_[0]) : ())
		1588	}
		1589
		1590	sub now() {
		1591	AnyEvent->now
		1592	}
		1593
		1594	sub now_update() {
		1595	AnyEvent->now_update
		1596	}
		1597
		1598	sub time() {
		1599	AnyEvent->time
		1600	}
		1601
		1602	*postpone = \&AnyEvent::postpone;
		1603	*log = \&AnyEvent::log;
		1604	};
		1605	die if $@;
		1606	}
		1607
		1608	BEGIN { _reset }
		1609
1017	package AnyEvent::Base;	1610	package AnyEvent::Base;
1018		1611
1019	# default implementation for now and time	1612	# default implementations for many methods
1020		1613
1021	use Time::HiRes ();	1614	sub time {
		1615	eval q{ # poor man's autoloading {}
		1616	# probe for availability of Time::HiRes
		1617	if (eval "use Time::HiRes (); Time::HiRes::time (); 1") {
		1618	*time = sub { Time::HiRes::time () };
		1619	*AE::time = \& Time::HiRes::time ;
		1620	*now = \&time;
		1621	AnyEvent::log 8 => "using Time::HiRes for sub-second timing accuracy.";
		1622	# if (eval "use POSIX (); (POSIX::times())...
		1623	} else {
		1624	*time = sub { CORE::time };
		1625	*AE::time = sub (){ CORE::time };
		1626	*now = \&time;
		1627	AnyEvent::log 3 => "Using built-in time(), no sub-second resolution!";
		1628	}
		1629	};
		1630	die if $@;
1022		1631
1023	sub time { Time::HiRes::time }	1632	&time
1024	sub now { Time::HiRes::time }	1633	}
		1634
		1635	*now = \&time;
		1636	sub now_update { }
		1637
		1638	sub _poll {
		1639	Carp::croak "$AnyEvent::MODEL does not support blocking waits. Caught";
		1640	}
1025		1641
1026	# default implementation for ->condvar	1642	# default implementation for ->condvar
		1643	# in fact, the default should not be overwritten
1027		1644
1028	sub condvar {	1645	sub condvar {
		1646	eval q{ # poor man's autoloading {}
		1647	*condvar = sub {
1029	bless { @_ == 3 ? (_ae_cb => $_[2]) : () }, AnyEvent::CondVar::	1648	bless { @_ == 3 ? (_ae_cb => $_[2]) : () }, "AnyEvent::CondVar"
		1649	};
		1650
		1651	*AE::cv = sub (;&) {
		1652	bless { @_ ? (_ae_cb => shift) : () }, "AnyEvent::CondVar"
		1653	};
		1654	};
		1655	die if $@;
		1656
		1657	&condvar
1030	}	1658	}
1031		1659
1032	# default implementation for ->signal	1660	# default implementation for ->signal
1033		1661
1034	our %SIG_CB;	1662	our $HAVE_ASYNC_INTERRUPT;
		1663
		1664	sub _have_async_interrupt() {
		1665	$HAVE_ASYNC_INTERRUPT = 1*(!$ENV{PERL_ANYEVENT_AVOID_ASYNC_INTERRUPT}
		1666	&& eval "use Async::Interrupt 1.02 (); 1")
		1667	unless defined $HAVE_ASYNC_INTERRUPT;
		1668
		1669	$HAVE_ASYNC_INTERRUPT
		1670	}
		1671
		1672	our ($SIGPIPE_R, $SIGPIPE_W, %SIG_CB, %SIG_EV, $SIG_IO);
		1673	our (%SIG_ASY, %SIG_ASY_W);
		1674	our ($SIG_COUNT, $SIG_TW);
		1675
		1676	# install a dummy wakeup watcher to reduce signal catching latency
		1677	# used by Impls
		1678	sub _sig_add() {
		1679	unless ($SIG_COUNT++) {
		1680	# try to align timer on a full-second boundary, if possible
		1681	my $NOW = AE::now;
		1682
		1683	$SIG_TW = AE::timer
		1684	$MAX_SIGNAL_LATENCY - ($NOW - int $NOW),
		1685	$MAX_SIGNAL_LATENCY,
		1686	sub { } # just for the PERL_ASYNC_CHECK
		1687	;
		1688	}
		1689	}
		1690
		1691	sub _sig_del {
		1692	undef $SIG_TW
		1693	unless --$SIG_COUNT;
		1694	}
		1695
		1696	our $_sig_name_init; $_sig_name_init = sub {
		1697	eval q{ # poor man's autoloading {}
		1698	undef $_sig_name_init;
		1699
		1700	if (_have_async_interrupt) {
		1701	*sig2num = \&Async::Interrupt::sig2num;
		1702	*sig2name = \&Async::Interrupt::sig2name;
		1703	} else {
		1704	require Config;
		1705
		1706	my %signame2num;
		1707	@signame2num{ split ' ', $Config::Config{sig_name} }
		1708	= split ' ', $Config::Config{sig_num};
		1709
		1710	my @signum2name;
		1711	@signum2name[values %signame2num] = keys %signame2num;
		1712
		1713	*sig2num = sub($) {
		1714	$_[0] > 0 ? shift : $signame2num{+shift}
		1715	};
		1716	*sig2name = sub ($) {
		1717	$_[0] > 0 ? $signum2name[+shift] : shift
		1718	};
		1719	}
		1720	};
		1721	die if $@;
		1722	};
		1723
		1724	sub sig2num ($) { &$_sig_name_init; &sig2num }
		1725	sub sig2name($) { &$_sig_name_init; &sig2name }
1035		1726
1036	sub signal {	1727	sub signal {
		1728	eval q{ # poor man's autoloading {}
		1729	# probe for availability of Async::Interrupt
		1730	if (_have_async_interrupt) {
		1731	AnyEvent::log 8 => "Using Async::Interrupt for race-free signal handling.";
		1732
		1733	$SIGPIPE_R = new Async::Interrupt::EventPipe;
		1734	$SIG_IO = AE::io $SIGPIPE_R->fileno, 0, \&_signal_exec;
		1735
		1736	} else {
		1737	AnyEvent::log 8 => "Using emulated perl signal handling with latency timer.";
		1738
		1739	if (AnyEvent::WIN32) {
		1740	require AnyEvent::Util;
		1741
		1742	($SIGPIPE_R, $SIGPIPE_W) = AnyEvent::Util::portable_pipe ();
		1743	AnyEvent::Util::fh_nonblocking ($SIGPIPE_R, 1) if $SIGPIPE_R;
		1744	AnyEvent::Util::fh_nonblocking ($SIGPIPE_W, 1) if $SIGPIPE_W; # just in case
		1745	} else {
		1746	pipe $SIGPIPE_R, $SIGPIPE_W;
		1747	fcntl $SIGPIPE_R, AnyEvent::F_SETFL, AnyEvent::O_NONBLOCK if $SIGPIPE_R;
		1748	fcntl $SIGPIPE_W, AnyEvent::F_SETFL, AnyEvent::O_NONBLOCK if $SIGPIPE_W; # just in case
		1749
		1750	# not strictly required, as $^F is normally 2, but let's make sure...
		1751	fcntl $SIGPIPE_R, AnyEvent::F_SETFD, AnyEvent::FD_CLOEXEC;
		1752	fcntl $SIGPIPE_W, AnyEvent::F_SETFD, AnyEvent::FD_CLOEXEC;
		1753	}
		1754
		1755	$SIGPIPE_R
		1756	or Carp::croak "AnyEvent: unable to create a signal reporting pipe: $!\n";
		1757
		1758	$SIG_IO = AE::io $SIGPIPE_R, 0, \&_signal_exec;
		1759	}
		1760
		1761	*signal = $HAVE_ASYNC_INTERRUPT
		1762	? sub {
1037	my (undef, %arg) = @_;	1763	my (undef, %arg) = @_;
1038		1764
		1765	# async::interrupt
1039	my $signal = uc $arg{signal}	1766	my $signal = sig2num $arg{signal};
1040	or Carp::croak "required option 'signal' is missing";
1041
1042	$SIG_CB{$signal}{$arg{cb}} = $arg{cb};	1767	$SIG_CB{$signal}{$arg{cb}} = $arg{cb};
		1768
		1769	$SIG_ASY{$signal} ||= new Async::Interrupt
		1770	cb => sub { undef $SIG_EV{$signal} },
		1771	signal => $signal,
		1772	pipe => [$SIGPIPE_R->filenos],
		1773	pipe_autodrain => 0,
		1774	;
		1775
		1776	bless [$signal, $arg{cb}], "AnyEvent::Base::signal"
		1777	}
		1778	: sub {
		1779	my (undef, %arg) = @_;
		1780
		1781	# pure perl
		1782	my $signal = sig2name $arg{signal};
		1783	$SIG_CB{$signal}{$arg{cb}} = $arg{cb};
		1784
1043	$SIG{$signal} ||= sub {	1785	$SIG{$signal} ||= sub {
		1786	local $!;
		1787	syswrite $SIGPIPE_W, "\x00", 1 unless %SIG_EV;
		1788	undef $SIG_EV{$signal};
		1789	};
		1790
		1791	# can't do signal processing without introducing races in pure perl,
		1792	# so limit the signal latency.
		1793	_sig_add;
		1794
		1795	bless [$signal, $arg{cb}], "AnyEvent::Base::signal"
		1796	}
		1797	;
		1798
		1799	*AnyEvent::Base::signal::DESTROY = sub {
		1800	my ($signal, $cb) = @{$_[0]};
		1801
		1802	_sig_del;
		1803
		1804	delete $SIG_CB{$signal}{$cb};
		1805
		1806	$HAVE_ASYNC_INTERRUPT
		1807	? delete $SIG_ASY{$signal}
		1808	: # delete doesn't work with older perls - they then
		1809	# print weird messages, or just unconditionally exit
		1810	# instead of getting the default action.
		1811	undef $SIG{$signal}
		1812	unless keys %{ $SIG_CB{$signal} };
		1813	};
		1814
		1815	*_signal_exec = sub {
		1816	$HAVE_ASYNC_INTERRUPT
		1817	? $SIGPIPE_R->drain
		1818	: sysread $SIGPIPE_R, (my $dummy), 9;
		1819
		1820	while (%SIG_EV) {
		1821	for (keys %SIG_EV) {
		1822	delete $SIG_EV{$_};
1044	$_->() for values %{ $SIG_CB{$signal} || {} };	1823	&$_ for values %{ $SIG_CB{$_} || {} };
		1824	}
		1825	}
		1826	};
1045	};	1827	};
		1828	die if $@;
1046		1829
1047	bless [$signal, $arg{cb}], "AnyEvent::Base::Signal"	1830	&signal
1048	}
1049
1050	sub AnyEvent::Base::Signal::DESTROY {
1051	my ($signal, $cb) = @{$_[0]};
1052
1053	delete $SIG_CB{$signal}{$cb};
1054
1055	delete $SIG{$signal} unless keys %{ $SIG_CB{$signal} };
1056	}	1831	}
1057		1832
1058	# default implementation for ->child	1833	# default implementation for ->child
1059		1834
1060	our %PID_CB;	1835	our %PID_CB;
1061	our $CHLD_W;	1836	our $CHLD_W;
1062	our $CHLD_DELAY_W;	1837	our $CHLD_DELAY_W;
1063	our $PID_IDLE;
1064	our $WNOHANG;
1065		1838
1066	sub _child_wait {	1839	# used by many Impl's
1067	while (0 < (my $pid = waitpid -1, $WNOHANG)) {	1840	sub _emit_childstatus($$) {
		1841	my (undef, $rpid, $rstatus) = @_;
		1842
		1843	$_->($rpid, $rstatus)
1068	$_->($pid, $?) for (values %{ $PID_CB{$pid} || {} }),	1844	for values %{ $PID_CB{$rpid} || {} },
1069	(values %{ $PID_CB{0} || {} });	1845	values %{ $PID_CB{0} || {} };
1070	}
1071
1072	undef $PID_IDLE;
1073	}
1074
1075	sub _sigchld {
1076	# make sure we deliver these changes "synchronous" with the event loop.
1077	$CHLD_DELAY_W ||= AnyEvent->timer (after => 0, cb => sub {
1078	undef $CHLD_DELAY_W;
1079	&_child_wait;
1080	});
1081	}	1846	}
1082		1847
1083	sub child {	1848	sub child {
		1849	eval q{ # poor man's autoloading {}
		1850	*_sigchld = sub {
		1851	my $pid;
		1852
		1853	AnyEvent->_emit_childstatus ($pid, $?)
		1854	while ($pid = waitpid -1, WNOHANG) > 0;
		1855	};
		1856
		1857	*child = sub {
1084	my (undef, %arg) = @_;	1858	my (undef, %arg) = @_;
1085		1859
1086	defined (my $pid = $arg{pid} + 0)	1860	my $pid = $arg{pid};
1087	or Carp::croak "required option 'pid' is missing";	1861	my $cb = $arg{cb};
1088		1862
1089	$PID_CB{$pid}{$arg{cb}} = $arg{cb};	1863	$PID_CB{$pid}{$cb+0} = $cb;
1090		1864
1091	unless ($WNOHANG) {
1092	$WNOHANG = eval { local $SIG{__DIE__}; require POSIX; &POSIX::WNOHANG } || 1;
1093	}
1094
1095	unless ($CHLD_W) {	1865	unless ($CHLD_W) {
1096	$CHLD_W = AnyEvent->signal (signal => 'CHLD', cb => \&_sigchld);	1866	$CHLD_W = AE::signal CHLD => \&_sigchld;
1097	# child could be a zombie already, so make at least one round	1867	# child could be a zombie already, so make at least one round
1098	&_sigchld;	1868	&_sigchld;
1099	}	1869	}
1100		1870
1101	bless [$pid, $arg{cb}], "AnyEvent::Base::Child"	1871	bless [$pid, $cb+0], "AnyEvent::Base::child"
1102	}	1872	};
1103		1873
1104	sub AnyEvent::Base::Child::DESTROY {	1874	*AnyEvent::Base::child::DESTROY = sub {
1105	my ($pid, $cb) = @{$_[0]};	1875	my ($pid, $icb) = @{$_[0]};
1106		1876
1107	delete $PID_CB{$pid}{$cb};	1877	delete $PID_CB{$pid}{$icb};
1108	delete $PID_CB{$pid} unless keys %{ $PID_CB{$pid} };	1878	delete $PID_CB{$pid} unless keys %{ $PID_CB{$pid} };
1109		1879
1110	undef $CHLD_W unless keys %PID_CB;	1880	undef $CHLD_W unless keys %PID_CB;
		1881	};
		1882	};
		1883	die if $@;
		1884
		1885	&child
		1886	}
		1887
		1888	# idle emulation is done by simply using a timer, regardless
		1889	# of whether the process is idle or not, and not letting
		1890	# the callback use more than 50% of the time.
		1891	sub idle {
		1892	eval q{ # poor man's autoloading {}
		1893	*idle = sub {
		1894	my (undef, %arg) = @_;
		1895
		1896	my ($cb, $w, $rcb) = $arg{cb};
		1897
		1898	$rcb = sub {
		1899	if ($cb) {
		1900	$w = AE::time;
		1901	&$cb;
		1902	$w = AE::time - $w;
		1903
		1904	# never use more then 50% of the time for the idle watcher,
		1905	# within some limits
		1906	$w = 0.0001 if $w < 0.0001;
		1907	$w = 5 if $w > 5;
		1908
		1909	$w = AE::timer $w, 0, $rcb;
		1910	} else {
		1911	# clean up...
		1912	undef $w;
		1913	undef $rcb;
		1914	}
		1915	};
		1916
		1917	$w = AE::timer 0.05, 0, $rcb;
		1918
		1919	bless \\$cb, "AnyEvent::Base::idle"
		1920	};
		1921
		1922	*AnyEvent::Base::idle::DESTROY = sub {
		1923	undef $${$_[0]};
		1924	};
		1925	};
		1926	die if $@;
		1927
		1928	&idle
1111	}	1929	}
1112		1930
1113	package AnyEvent::CondVar;	1931	package AnyEvent::CondVar;
1114		1932
1115	our @ISA = AnyEvent::CondVar::Base::;	1933	our @ISA = AnyEvent::CondVar::Base::;
1116		1934
		1935	# only to be used for subclassing
		1936	sub new {
		1937	my $class = shift;
		1938	bless AnyEvent->condvar (@_), $class
		1939	}
		1940
1117	package AnyEvent::CondVar::Base;	1941	package AnyEvent::CondVar::Base;
1118		1942
1119	use overload	1943	#use overload
1120	'&{}' => sub { my $self = shift; sub { $self->send (@_) } },	1944	# '&{}' => sub { my $self = shift; sub { $self->send (@_) } },
1121	fallback => 1;	1945	# fallback => 1;
		1946
		1947	# save 300+ kilobytes by dirtily hardcoding overloading
		1948	${"AnyEvent::CondVar::Base::OVERLOAD"}{dummy}++; # Register with magic by touching.
		1949	*{'AnyEvent::CondVar::Base::()'} = sub { }; # "Make it findable via fetchmethod."
		1950	*{'AnyEvent::CondVar::Base::(&{}'} = sub { my $self = shift; sub { $self->send (@_) } }; # &{}
		1951	${'AnyEvent::CondVar::Base::()'} = 1; # fallback
		1952
		1953	our $WAITING;
1122		1954
1123	sub _send {	1955	sub _send {
1124	# nop	1956	# nop
		1957	}
		1958
		1959	sub _wait {
		1960	AnyEvent->_poll until $_[0]{_ae_sent};
1125	}	1961	}
1126		1962
1127	sub send {	1963	sub send {
1128	my $cv = shift;	1964	my $cv = shift;
1129	$cv->{_ae_sent} = [@_];	1965	$cv->{_ae_sent} = [@_];
…		…
1138		1974
1139	sub ready {	1975	sub ready {
1140	$_[0]{_ae_sent}	1976	$_[0]{_ae_sent}
1141	}	1977	}
1142		1978
1143	sub _wait {
1144	AnyEvent->one_event while !$_[0]{_ae_sent};
1145	}
1146
1147	sub recv {	1979	sub recv {
		1980	unless ($_[0]{_ae_sent}) {
		1981	$WAITING
		1982	and Carp::croak "AnyEvent::CondVar: recursive blocking wait attempted";
		1983
		1984	local $WAITING = 1;
1148	$_[0]->_wait;	1985	$_[0]->_wait;
		1986	}
1149		1987
1150	Carp::croak $_[0]{_ae_croak} if $_[0]{_ae_croak};	1988	$_[0]{_ae_croak}
1151	wantarray ? @{ $_[0]{_ae_sent} } : $_[0]{_ae_sent}[0]	1989	and Carp::croak $_[0]{_ae_croak};
		1990
		1991	wantarray
		1992	? @{ $_[0]{_ae_sent} }
		1993	: $_[0]{_ae_sent}[0]
1152	}	1994	}
1153		1995
1154	sub cb {	1996	sub cb {
1155	$_[0]{_ae_cb} = $_[1] if @_ > 1;	1997	my $cv = shift;
		1998
		1999	@_
		2000	and $cv->{_ae_cb} = shift
		2001	and $cv->{_ae_sent}
		2002	and (delete $cv->{_ae_cb})->($cv);
		2003
1156	$_[0]{_ae_cb}	2004	$cv->{_ae_cb}
1157	}	2005	}
1158		2006
1159	sub begin {	2007	sub begin {
1160	++$_[0]{_ae_counter};	2008	++$_[0]{_ae_counter};
1161	$_[0]{_ae_end_cb} = $_[1] if @_ > 1;	2009	$_[0]{_ae_end_cb} = $_[1] if @_ > 1;
…		…
1166	&{ $_[0]{_ae_end_cb} || sub { $_[0]->send } };	2014	&{ $_[0]{_ae_end_cb} || sub { $_[0]->send } };
1167	}	2015	}
1168		2016
1169	# undocumented/compatibility with pre-3.4	2017	# undocumented/compatibility with pre-3.4
1170	*broadcast = \&send;	2018	*broadcast = \&send;
1171	*wait = \&_wait;	2019	*wait = \&recv;
		2020
		2021	=head1 ERROR AND EXCEPTION HANDLING
		2022
		2023	In general, AnyEvent does not do any error handling - it relies on the
		2024	caller to do that if required. The L<AnyEvent::Strict> module (see also
		2025	the C<PERL_ANYEVENT_STRICT> environment variable, below) provides strict
		2026	checking of all AnyEvent methods, however, which is highly useful during
		2027	development.
		2028
		2029	As for exception handling (i.e. runtime errors and exceptions thrown while
		2030	executing a callback), this is not only highly event-loop specific, but
		2031	also not in any way wrapped by this module, as this is the job of the main
		2032	program.
		2033
		2034	The pure perl event loop simply re-throws the exception (usually
		2035	within C<< condvar->recv >>), the L<Event> and L<EV> modules call C<<
		2036	$Event/EV::DIED->() >>, L<Glib> uses C<< install_exception_handler >> and
		2037	so on.
		2038
		2039	=head1 ENVIRONMENT VARIABLES
		2040
		2041	AnyEvent supports a number of environment variables that tune the
		2042	runtime behaviour. They are usually evaluated when AnyEvent is
		2043	loaded, initialised, or a submodule that uses them is loaded. Many of
		2044	them also cause AnyEvent to load additional modules - for example,
		2045	C<PERL_ANYEVENT_DEBUG_WRAP> causes the L<AnyEvent::Debug> module to be
		2046	loaded.
		2047
		2048	All the environment variables documented here start with
		2049	C<PERL_ANYEVENT_>, which is what AnyEvent considers its own
		2050	namespace. Other modules are encouraged (but by no means required) to use
		2051	C<PERL_ANYEVENT_SUBMODULE> if they have registered the AnyEvent::Submodule
		2052	namespace on CPAN, for any submodule. For example, L<AnyEvent::HTTP> could
		2053	be expected to use C<PERL_ANYEVENT_HTTP_PROXY> (it should not access env
		2054	variables starting with C<AE_>, see below).
		2055
		2056	All variables can also be set via the C<AE_> prefix, that is, instead
		2057	of setting C<PERL_ANYEVENT_VERBOSE> you can also set C<AE_VERBOSE>. In
		2058	case there is a clash btween anyevent and another program that uses
		2059	C<AE_something> you can set the corresponding C<PERL_ANYEVENT_something>
		2060	variable to the empty string, as those variables take precedence.
		2061
		2062	When AnyEvent is first loaded, it copies all C<AE_xxx> env variables
		2063	to their C<PERL_ANYEVENT_xxx> counterpart unless that variable already
		2064	exists. If taint mode is on, then AnyEvent will remove I<all> environment
		2065	variables starting with C<PERL_ANYEVENT_> from C<%ENV> (or replace them
		2066	with C<undef> or the empty string, if the corresaponding C<AE_> variable
		2067	is set).
		2068
		2069	The exact algorithm is currently:
		2070
		2071	1. if taint mode enabled, delete all PERL_ANYEVENT_xyz variables from %ENV
		2072	2. copy over AE_xyz to PERL_ANYEVENT_xyz unless the latter alraedy exists
		2073	3. if taint mode enabled, set all PERL_ANYEVENT_xyz variables to undef.
		2074
		2075	This ensures that child processes will not see the C<AE_> variables.
		2076
		2077	The following environment variables are currently known to AnyEvent:
		2078
		2079	=over 4
		2080
		2081	=item C<PERL_ANYEVENT_VERBOSE>
		2082
		2083	By default, AnyEvent will log messages with loglevel C<4> (C<error>) or
		2084	higher (see L<AnyEvent::Log>). You can set this environment variable to a
		2085	numerical loglevel to make AnyEvent more (or less) talkative.
		2086
		2087	If you want to do more than just set the global logging level
		2088	you should have a look at C<PERL_ANYEVENT_LOG>, which allows much more
		2089	complex specifications.
		2090
		2091	When set to C<0> (C<off>), then no messages whatsoever will be logged with
		2092	everything else at defaults.
		2093
		2094	When set to C<5> or higher (C<warn>), AnyEvent warns about unexpected
		2095	conditions, such as not being able to load the event model specified by
		2096	C<PERL_ANYEVENT_MODEL>, or a guard callback throwing an exception - this
		2097	is the minimum recommended level for use during development.
		2098
		2099	When set to C<7> or higher (info), AnyEvent reports which event model it
		2100	chooses.
		2101
		2102	When set to C<8> or higher (debug), then AnyEvent will report extra
		2103	information on which optional modules it loads and how it implements
		2104	certain features.
		2105
		2106	=item C<PERL_ANYEVENT_LOG>
		2107
		2108	Accepts rather complex logging specifications. For example, you could log
		2109	all C<debug> messages of some module to stderr, warnings and above to
		2110	stderr, and errors and above to syslog, with:
		2111
		2112	PERL_ANYEVENT_LOG=Some::Module=debug,+log:filter=warn,+%syslog:%syslog=error,syslog
		2113
		2114	For the rather extensive details, see L<AnyEvent::Log>.
		2115
		2116	This variable is evaluated when AnyEvent (or L<AnyEvent::Log>) is loaded,
		2117	so will take effect even before AnyEvent has initialised itself.
		2118
		2119	Note that specifying this environment variable causes the L<AnyEvent::Log>
		2120	module to be loaded, while C<PERL_ANYEVENT_VERBOSE> does not, so only
		2121	using the latter saves a few hundred kB of memory unless a module
		2122	explicitly needs the extra features of AnyEvent::Log.
		2123
		2124	=item C<PERL_ANYEVENT_STRICT>
		2125
		2126	AnyEvent does not do much argument checking by default, as thorough
		2127	argument checking is very costly. Setting this variable to a true value
		2128	will cause AnyEvent to load C<AnyEvent::Strict> and then to thoroughly
		2129	check the arguments passed to most method calls. If it finds any problems,
		2130	it will croak.
		2131
		2132	In other words, enables "strict" mode.
		2133
		2134	Unlike C<use strict> (or its modern cousin, C<< use L<common::sense>
		2135	>>, it is definitely recommended to keep it off in production. Keeping
		2136	C<PERL_ANYEVENT_STRICT=1> in your environment while developing programs
		2137	can be very useful, however.
		2138
		2139	=item C<PERL_ANYEVENT_DEBUG_SHELL>
		2140
		2141	If this env variable is nonempty, then its contents will be interpreted by
		2142	C<AnyEvent::Socket::parse_hostport> and C<AnyEvent::Debug::shell> (after
		2143	replacing every occurance of C<$$> by the process pid). The shell object
		2144	is saved in C<$AnyEvent::Debug::SHELL>.
		2145
		2146	This happens when the first watcher is created.
		2147
		2148	For example, to bind a debug shell on a unix domain socket in
		2149	F<< /tmp/debug<pid>.sock >>, you could use this:
		2150
		2151	PERL_ANYEVENT_DEBUG_SHELL=/tmp/debug\$\$.sock perlprog
		2152	# connect with e.g.: socat readline /tmp/debug123.sock
		2153
		2154	Or to bind to tcp port 4545 on localhost:
		2155
		2156	PERL_ANYEVENT_DEBUG_SHELL=127.0.0.1:4545 perlprog
		2157	# connect with e.g.: telnet localhost 4545
		2158
		2159	Note that creating sockets in F</tmp> or on localhost is very unsafe on
		2160	multiuser systems.
		2161
		2162	=item C<PERL_ANYEVENT_DEBUG_WRAP>
		2163
		2164	Can be set to C<0>, C<1> or C<2> and enables wrapping of all watchers for
		2165	debugging purposes. See C<AnyEvent::Debug::wrap> for details.
		2166
		2167	=item C<PERL_ANYEVENT_MODEL>
		2168
		2169	This can be used to specify the event model to be used by AnyEvent, before
		2170	auto detection and -probing kicks in.
		2171
		2172	It normally is a string consisting entirely of ASCII letters (e.g. C<EV>
		2173	or C<IOAsync>). The string C<AnyEvent::Impl::> gets prepended and the
		2174	resulting module name is loaded and - if the load was successful - used as
		2175	event model backend. If it fails to load then AnyEvent will proceed with
		2176	auto detection and -probing.
		2177
		2178	If the string ends with C<::> instead (e.g. C<AnyEvent::Impl::EV::>) then
		2179	nothing gets prepended and the module name is used as-is (hint: C<::> at
		2180	the end of a string designates a module name and quotes it appropriately).
		2181
		2182	For example, to force the pure perl model (L<AnyEvent::Loop::Perl>) you
		2183	could start your program like this:
		2184
		2185	PERL_ANYEVENT_MODEL=Perl perl ...
		2186
		2187	=item C<PERL_ANYEVENT_IO_MODEL>
		2188
		2189	The current file I/O model - see L<AnyEvent::IO> for more info.
		2190
		2191	At the moment, only C<Perl> (small, pure-perl, synchronous) and
		2192	C<IOAIO> (truly asynchronous) are supported. The default is C<IOAIO> if
		2193	L<AnyEvent::AIO> can be loaded, otherwise it is C<Perl>.
		2194
		2195	=item C<PERL_ANYEVENT_PROTOCOLS>
		2196
		2197	Used by both L<AnyEvent::DNS> and L<AnyEvent::Socket> to determine preferences
		2198	for IPv4 or IPv6. The default is unspecified (and might change, or be the result
		2199	of auto probing).
		2200
		2201	Must be set to a comma-separated list of protocols or address families,
		2202	current supported: C<ipv4> and C<ipv6>. Only protocols mentioned will be
		2203	used, and preference will be given to protocols mentioned earlier in the
		2204	list.
		2205
		2206	This variable can effectively be used for denial-of-service attacks
		2207	against local programs (e.g. when setuid), although the impact is likely
		2208	small, as the program has to handle conenction and other failures anyways.
		2209
		2210	Examples: C<PERL_ANYEVENT_PROTOCOLS=ipv4,ipv6> - prefer IPv4 over IPv6,
		2211	but support both and try to use both. C<PERL_ANYEVENT_PROTOCOLS=ipv4>
		2212	- only support IPv4, never try to resolve or contact IPv6
		2213	addresses. C<PERL_ANYEVENT_PROTOCOLS=ipv6,ipv4> support either IPv4 or
		2214	IPv6, but prefer IPv6 over IPv4.
		2215
		2216	=item C<PERL_ANYEVENT_HOSTS>
		2217
		2218	This variable, if specified, overrides the F</etc/hosts> file used by
		2219	L<AnyEvent::Socket>C<::resolve_sockaddr>, i.e. hosts aliases will be read
		2220	from that file instead.
		2221
		2222	=item C<PERL_ANYEVENT_EDNS0>
		2223
		2224	Used by L<AnyEvent::DNS> to decide whether to use the EDNS0 extension for
		2225	DNS. This extension is generally useful to reduce DNS traffic, especially
		2226	when DNSSEC is involved, but some (broken) firewalls drop such DNS
		2227	packets, which is why it is off by default.
		2228
		2229	Setting this variable to C<1> will cause L<AnyEvent::DNS> to announce
		2230	EDNS0 in its DNS requests.
		2231
		2232	=item C<PERL_ANYEVENT_MAX_FORKS>
		2233
		2234	The maximum number of child processes that C<AnyEvent::Util::fork_call>
		2235	will create in parallel.
		2236
		2237	=item C<PERL_ANYEVENT_MAX_OUTSTANDING_DNS>
		2238
		2239	The default value for the C<max_outstanding> parameter for the default DNS
		2240	resolver - this is the maximum number of parallel DNS requests that are
		2241	sent to the DNS server.
		2242
		2243	=item C<PERL_ANYEVENT_MAX_SIGNAL_LATENCY>
		2244
		2245	Perl has inherently racy signal handling (you can basically choose between
		2246	losing signals and memory corruption) - pure perl event loops (including
		2247	C<AnyEvent::Loop>, when C<Async::Interrupt> isn't available) therefore
		2248	have to poll regularly to avoid losing signals.
		2249
		2250	Some event loops are racy, but don't poll regularly, and some event loops
		2251	are written in C but are still racy. For those event loops, AnyEvent
		2252	installs a timer that regularly wakes up the event loop.
		2253
		2254	By default, the interval for this timer is C<10> seconds, but you can
		2255	override this delay with this environment variable (or by setting
		2256	the C<$AnyEvent::MAX_SIGNAL_LATENCY> variable before creating signal
		2257	watchers).
		2258
		2259	Lower values increase CPU (and energy) usage, higher values can introduce
		2260	long delays when reaping children or waiting for signals.
		2261
		2262	The L<AnyEvent::Async> module, if available, will be used to avoid this
		2263	polling (with most event loops).
		2264
		2265	=item C<PERL_ANYEVENT_RESOLV_CONF>
		2266
		2267	The absolute path to a F<resolv.conf>-style file to use instead of
		2268	F</etc/resolv.conf> (or the OS-specific configuration) in the default
		2269	resolver, or the empty string to select the default configuration.
		2270
		2271	=item C<PERL_ANYEVENT_CA_FILE>, C<PERL_ANYEVENT_CA_PATH>.
		2272
		2273	When neither C<ca_file> nor C<ca_path> was specified during
		2274	L<AnyEvent::TLS> context creation, and either of these environment
		2275	variables are nonempty, they will be used to specify CA certificate
		2276	locations instead of a system-dependent default.
		2277
		2278	=item C<PERL_ANYEVENT_AVOID_GUARD> and C<PERL_ANYEVENT_AVOID_ASYNC_INTERRUPT>
		2279
		2280	When these are set to C<1>, then the respective modules are not
		2281	loaded. Mostly good for testing AnyEvent itself.
		2282
		2283	=back
1172		2284
1173	=head1 SUPPLYING YOUR OWN EVENT MODEL INTERFACE	2285	=head1 SUPPLYING YOUR OWN EVENT MODEL INTERFACE
1174		2286
1175	This is an advanced topic that you do not normally need to use AnyEvent in	2287	This is an advanced topic that you do not normally need to use AnyEvent in
1176	a module. This section is only of use to event loop authors who want to	2288	a module. This section is only of use to event loop authors who want to
…		…
1210		2322
1211	I<rxvt-unicode> also cheats a bit by not providing blocking access to	2323	I<rxvt-unicode> also cheats a bit by not providing blocking access to
1212	condition variables: code blocking while waiting for a condition will	2324	condition variables: code blocking while waiting for a condition will
1213	C<die>. This still works with most modules/usages, and blocking calls must	2325	C<die>. This still works with most modules/usages, and blocking calls must
1214	not be done in an interactive application, so it makes sense.	2326	not be done in an interactive application, so it makes sense.
1215
1216	=head1 ENVIRONMENT VARIABLES
1217
1218	The following environment variables are used by this module:
1219
1220	=over 4
1221
1222	=item C<PERL_ANYEVENT_VERBOSE>
1223
1224	By default, AnyEvent will be completely silent except in fatal
1225	conditions. You can set this environment variable to make AnyEvent more
1226	talkative.
1227
1228	When set to C<1> or higher, causes AnyEvent to warn about unexpected
1229	conditions, such as not being able to load the event model specified by
1230	C<PERL_ANYEVENT_MODEL>.
1231
1232	When set to C<2> or higher, cause AnyEvent to report to STDERR which event
1233	model it chooses.
1234
1235	=item C<PERL_ANYEVENT_STRICT>
1236
1237	AnyEvent does not do much argument checking by default, as thorough
1238	argument checking is very costly. Setting this variable to a true value
1239	will cause AnyEvent to load C<AnyEvent::Strict> and then to thoroughly
1240	check the arguments passed to most method calls. If it finds any problems
1241	it will croak.
1242
1243	In other words, enables "strict" mode.
1244
1245	Unlike C<use strict> it is definitely recommended ot keep it off in
1246	production.
1247
1248	=item C<PERL_ANYEVENT_MODEL>
1249
1250	This can be used to specify the event model to be used by AnyEvent, before
1251	auto detection and -probing kicks in. It must be a string consisting
1252	entirely of ASCII letters. The string C<AnyEvent::Impl::> gets prepended
1253	and the resulting module name is loaded and if the load was successful,
1254	used as event model. If it fails to load AnyEvent will proceed with
1255	auto detection and -probing.
1256
1257	This functionality might change in future versions.
1258
1259	For example, to force the pure perl model (L<AnyEvent::Impl::Perl>) you
1260	could start your program like this:
1261
1262	PERL_ANYEVENT_MODEL=Perl perl ...
1263
1264	=item C<PERL_ANYEVENT_PROTOCOLS>
1265
1266	Used by both L<AnyEvent::DNS> and L<AnyEvent::Socket> to determine preferences
1267	for IPv4 or IPv6. The default is unspecified (and might change, or be the result
1268	of auto probing).
1269
1270	Must be set to a comma-separated list of protocols or address families,
1271	current supported: C<ipv4> and C<ipv6>. Only protocols mentioned will be
1272	used, and preference will be given to protocols mentioned earlier in the
1273	list.
1274
1275	This variable can effectively be used for denial-of-service attacks
1276	against local programs (e.g. when setuid), although the impact is likely
1277	small, as the program has to handle connection errors already-
1278
1279	Examples: C<PERL_ANYEVENT_PROTOCOLS=ipv4,ipv6> - prefer IPv4 over IPv6,
1280	but support both and try to use both. C<PERL_ANYEVENT_PROTOCOLS=ipv4>
1281	- only support IPv4, never try to resolve or contact IPv6
1282	addresses. C<PERL_ANYEVENT_PROTOCOLS=ipv6,ipv4> support either IPv4 or
1283	IPv6, but prefer IPv6 over IPv4.
1284
1285	=item C<PERL_ANYEVENT_EDNS0>
1286
1287	Used by L<AnyEvent::DNS> to decide whether to use the EDNS0 extension
1288	for DNS. This extension is generally useful to reduce DNS traffic, but
1289	some (broken) firewalls drop such DNS packets, which is why it is off by
1290	default.
1291
1292	Setting this variable to C<1> will cause L<AnyEvent::DNS> to announce
1293	EDNS0 in its DNS requests.
1294
1295	=item C<PERL_ANYEVENT_MAX_FORKS>
1296
1297	The maximum number of child processes that C<AnyEvent::Util::fork_call>
1298	will create in parallel.
1299
1300	=back
1301		2327
1302	=head1 EXAMPLE PROGRAM	2328	=head1 EXAMPLE PROGRAM
1303		2329
1304	The following program uses an I/O watcher to read data from STDIN, a timer	2330	The following program uses an I/O watcher to read data from STDIN, a timer
1305	to display a message once per second, and a condition variable to quit the	2331	to display a message once per second, and a condition variable to quit the
…		…
1318	warn "read: $input\n"; # output what has been read	2344	warn "read: $input\n"; # output what has been read
1319	$cv->send if $input =~ /^q/i; # quit program if /^q/i	2345	$cv->send if $input =~ /^q/i; # quit program if /^q/i
1320	},	2346	},
1321	);	2347	);
1322		2348
1323	my $time_watcher; # can only be used once
1324
1325	sub new_timer {
1326	$timer = AnyEvent->timer (after => 1, cb => sub {	2349	my $time_watcher = AnyEvent->timer (after => 1, interval => 1, cb => sub {
1327	warn "timeout\n"; # print 'timeout' about every second	2350	warn "timeout\n"; # print 'timeout' at most every second
1328	&new_timer; # and restart the time
1329	});	2351	});
1330	}
1331
1332	new_timer; # create first timer
1333		2352
1334	$cv->recv; # wait until user enters /^q/i	2353	$cv->recv; # wait until user enters /^q/i
1335		2354
1336	=head1 REAL-WORLD EXAMPLE	2355	=head1 REAL-WORLD EXAMPLE
1337		2356
…		…
1410		2429
1411	The actual code goes further and collects all errors (C<die>s, exceptions)	2430	The actual code goes further and collects all errors (C<die>s, exceptions)
1412	that occurred during request processing. The C<result> method detects	2431	that occurred during request processing. The C<result> method detects
1413	whether an exception as thrown (it is stored inside the $txn object)	2432	whether an exception as thrown (it is stored inside the $txn object)
1414	and just throws the exception, which means connection errors and other	2433	and just throws the exception, which means connection errors and other
1415	problems get reported tot he code that tries to use the result, not in a	2434	problems get reported to the code that tries to use the result, not in a
1416	random callback.	2435	random callback.
1417		2436
1418	All of this enables the following usage styles:	2437	All of this enables the following usage styles:
1419		2438
1420	1. Blocking:	2439	1. Blocking:
…		…
1468	through AnyEvent. The benchmark creates a lot of timers (with a zero	2487	through AnyEvent. The benchmark creates a lot of timers (with a zero
1469	timeout) and I/O watchers (watching STDOUT, a pty, to become writable,	2488	timeout) and I/O watchers (watching STDOUT, a pty, to become writable,
1470	which it is), lets them fire exactly once and destroys them again.	2489	which it is), lets them fire exactly once and destroys them again.
1471		2490
1472	Source code for this benchmark is found as F<eg/bench> in the AnyEvent	2491	Source code for this benchmark is found as F<eg/bench> in the AnyEvent
1473	distribution.	2492	distribution. It uses the L<AE> interface, which makes a real difference
		2493	for the EV and Perl backends only.
1474		2494
1475	=head3 Explanation of the columns	2495	=head3 Explanation of the columns
1476		2496
1477	I<watcher> is the number of event watchers created/destroyed. Since	2497	I<watcher> is the number of event watchers created/destroyed. Since
1478	different event models feature vastly different performances, each event	2498	different event models feature vastly different performances, each event
…		…
1499	watcher.	2519	watcher.
1500		2520
1501	=head3 Results	2521	=head3 Results
1502		2522
1503	name watchers bytes create invoke destroy comment	2523	name watchers bytes create invoke destroy comment
1504	EV/EV 400000 244 0.56 0.46 0.31 EV native interface	2524	EV/EV 100000 223 0.47 0.43 0.27 EV native interface
1505	EV/Any 100000 244 2.50 0.46 0.29 EV + AnyEvent watchers	2525	EV/Any 100000 223 0.48 0.42 0.26 EV + AnyEvent watchers
1506	CoroEV/Any 100000 244 2.49 0.44 0.29 coroutines + Coro::Signal	2526	Coro::EV/Any 100000 223 0.47 0.42 0.26 coroutines + Coro::Signal
1507	Perl/Any 100000 513 4.92 0.87 1.12 pure perl implementation	2527	Perl/Any 100000 431 2.70 0.74 0.92 pure perl implementation
1508	Event/Event 16000 516 31.88 31.30 0.85 Event native interface	2528	Event/Event 16000 516 31.16 31.84 0.82 Event native interface
1509	Event/Any 16000 590 35.75 31.42 1.08 Event + AnyEvent watchers	2529	Event/Any 16000 1203 42.61 34.79 1.80 Event + AnyEvent watchers
		2530	IOAsync/Any 16000 1911 41.92 27.45 16.81 via IO::Async::Loop::IO_Poll
		2531	IOAsync/Any 16000 1726 40.69 26.37 15.25 via IO::Async::Loop::Epoll
1510	Glib/Any 16000 1357 98.22 12.41 54.00 quadratic behaviour	2532	Glib/Any 16000 1118 89.00 12.57 51.17 quadratic behaviour
1511	Tk/Any 2000 1860 26.97 67.98 14.00 SEGV with >> 2000 watchers	2533	Tk/Any 2000 1346 20.96 10.75 8.00 SEGV with >> 2000 watchers
1512	POE/Event 2000 6644 108.64 736.02 14.73 via POE::Loop::Event	2534	POE/Any 2000 6951 108.97 795.32 14.24 via POE::Loop::Event
1513	POE/Select 2000 6343 94.13 809.12 565.96 via POE::Loop::Select	2535	POE/Any 2000 6648 94.79 774.40 575.51 via POE::Loop::Select
1514		2536
1515	=head3 Discussion	2537	=head3 Discussion
1516		2538
1517	The benchmark does I<not> measure scalability of the event loop very	2539	The benchmark does I<not> measure scalability of the event loop very
1518	well. For example, a select-based event loop (such as the pure perl one)	2540	well. For example, a select-based event loop (such as the pure perl one)
…		…
1530	benchmark machine, handling an event takes roughly 1600 CPU cycles with	2552	benchmark machine, handling an event takes roughly 1600 CPU cycles with
1531	EV, 3100 CPU cycles with AnyEvent's pure perl loop and almost 3000000 CPU	2553	EV, 3100 CPU cycles with AnyEvent's pure perl loop and almost 3000000 CPU
1532	cycles with POE.	2554	cycles with POE.
1533		2555
1534	C<EV> is the sole leader regarding speed and memory use, which are both	2556	C<EV> is the sole leader regarding speed and memory use, which are both
1535	maximal/minimal, respectively. Even when going through AnyEvent, it uses	2557	maximal/minimal, respectively. When using the L<AE> API there is zero
		2558	overhead (when going through the AnyEvent API create is about 5-6 times
		2559	slower, with other times being equal, so still uses far less memory than
1536	far less memory than any other event loop and is still faster than Event	2560	any other event loop and is still faster than Event natively).
1537	natively.
1538		2561
1539	The pure perl implementation is hit in a few sweet spots (both the	2562	The pure perl implementation is hit in a few sweet spots (both the
1540	constant timeout and the use of a single fd hit optimisations in the perl	2563	constant timeout and the use of a single fd hit optimisations in the perl
1541	interpreter and the backend itself). Nevertheless this shows that it	2564	interpreter and the backend itself). Nevertheless this shows that it
1542	adds very little overhead in itself. Like any select-based backend its	2565	adds very little overhead in itself. Like any select-based backend its
1543	performance becomes really bad with lots of file descriptors (and few of	2566	performance becomes really bad with lots of file descriptors (and few of
1544	them active), of course, but this was not subject of this benchmark.	2567	them active), of course, but this was not subject of this benchmark.
1545		2568
1546	The C<Event> module has a relatively high setup and callback invocation	2569	The C<Event> module has a relatively high setup and callback invocation
1547	cost, but overall scores in on the third place.	2570	cost, but overall scores in on the third place.
		2571
		2572	C<IO::Async> performs admirably well, about on par with C<Event>, even
		2573	when using its pure perl backend.
1548		2574
1549	C<Glib>'s memory usage is quite a bit higher, but it features a	2575	C<Glib>'s memory usage is quite a bit higher, but it features a
1550	faster callback invocation and overall ends up in the same class as	2576	faster callback invocation and overall ends up in the same class as
1551	C<Event>. However, Glib scales extremely badly, doubling the number of	2577	C<Event>. However, Glib scales extremely badly, doubling the number of
1552	watchers increases the processing time by more than a factor of four,	2578	watchers increases the processing time by more than a factor of four,
…		…
1587	(even when used without AnyEvent), but most event loops have acceptable	2613	(even when used without AnyEvent), but most event loops have acceptable
1588	performance with or without AnyEvent.	2614	performance with or without AnyEvent.
1589		2615
1590	=item * The overhead AnyEvent adds is usually much smaller than the overhead of	2616	=item * The overhead AnyEvent adds is usually much smaller than the overhead of
1591	the actual event loop, only with extremely fast event loops such as EV	2617	the actual event loop, only with extremely fast event loops such as EV
1592	adds AnyEvent significant overhead.	2618	does AnyEvent add significant overhead.
1593		2619
1594	=item * You should avoid POE like the plague if you want performance or	2620	=item * You should avoid POE like the plague if you want performance or
1595	reasonable memory usage.	2621	reasonable memory usage.
1596		2622
1597	=back	2623	=back
…		…
1613	In this benchmark, we use 10000 socket pairs (20000 sockets), of which 100	2639	In this benchmark, we use 10000 socket pairs (20000 sockets), of which 100
1614	(1%) are active. This mirrors the activity of large servers with many	2640	(1%) are active. This mirrors the activity of large servers with many
1615	connections, most of which are idle at any one point in time.	2641	connections, most of which are idle at any one point in time.
1616		2642
1617	Source code for this benchmark is found as F<eg/bench2> in the AnyEvent	2643	Source code for this benchmark is found as F<eg/bench2> in the AnyEvent
1618	distribution.	2644	distribution. It uses the L<AE> interface, which makes a real difference
		2645	for the EV and Perl backends only.
1619		2646
1620	=head3 Explanation of the columns	2647	=head3 Explanation of the columns
1621		2648
1622	I<sockets> is the number of sockets, and twice the number of "servers" (as	2649	I<sockets> is the number of sockets, and twice the number of "servers" (as
1623	each server has a read and write socket end).	2650	each server has a read and write socket end).
…		…
1630	it to another server. This includes deleting the old timeout and creating	2657	it to another server. This includes deleting the old timeout and creating
1631	a new one that moves the timeout into the future.	2658	a new one that moves the timeout into the future.
1632		2659
1633	=head3 Results	2660	=head3 Results
1634		2661
1635	name sockets create request	2662	name sockets create request
1636	EV 20000 69.01 11.16	2663	EV 20000 62.66 7.99
1637	Perl 20000 73.32 35.87	2664	Perl 20000 68.32 32.64
1638	Event 20000 212.62 257.32	2665	IOAsync 20000 174.06 101.15 epoll
1639	Glib 20000 651.16 1896.30	2666	IOAsync 20000 174.67 610.84 poll
		2667	Event 20000 202.69 242.91
		2668	Glib 20000 557.01 1689.52
1640	POE 20000 349.67 12317.24 uses POE::Loop::Event	2669	POE 20000 341.54 12086.32 uses POE::Loop::Event
1641		2670
1642	=head3 Discussion	2671	=head3 Discussion
1643		2672
1644	This benchmark I<does> measure scalability and overall performance of the	2673	This benchmark I<does> measure scalability and overall performance of the
1645	particular event loop.	2674	particular event loop.
…		…
1647	EV is again fastest. Since it is using epoll on my system, the setup time	2676	EV is again fastest. Since it is using epoll on my system, the setup time
1648	is relatively high, though.	2677	is relatively high, though.
1649		2678
1650	Perl surprisingly comes second. It is much faster than the C-based event	2679	Perl surprisingly comes second. It is much faster than the C-based event
1651	loops Event and Glib.	2680	loops Event and Glib.
		2681
		2682	IO::Async performs very well when using its epoll backend, and still quite
		2683	good compared to Glib when using its pure perl backend.
1652		2684
1653	Event suffers from high setup time as well (look at its code and you will	2685	Event suffers from high setup time as well (look at its code and you will
1654	understand why). Callback invocation also has a high overhead compared to	2686	understand why). Callback invocation also has a high overhead compared to
1655	the C<< $_->() for .. >>-style loop that the Perl event loop uses. Event	2687	the C<< $_->() for .. >>-style loop that the Perl event loop uses. Event
1656	uses select or poll in basically all documented configurations.	2688	uses select or poll in basically all documented configurations.
…		…
1719	=item * C-based event loops perform very well with small number of	2751	=item * C-based event loops perform very well with small number of
1720	watchers, as the management overhead dominates.	2752	watchers, as the management overhead dominates.
1721		2753
1722	=back	2754	=back
1723		2755
		2756	=head2 THE IO::Lambda BENCHMARK
		2757
		2758	Recently I was told about the benchmark in the IO::Lambda manpage, which
		2759	could be misinterpreted to make AnyEvent look bad. In fact, the benchmark
		2760	simply compares IO::Lambda with POE, and IO::Lambda looks better (which
		2761	shouldn't come as a surprise to anybody). As such, the benchmark is
		2762	fine, and mostly shows that the AnyEvent backend from IO::Lambda isn't
		2763	very optimal. But how would AnyEvent compare when used without the extra
		2764	baggage? To explore this, I wrote the equivalent benchmark for AnyEvent.
		2765
		2766	The benchmark itself creates an echo-server, and then, for 500 times,
		2767	connects to the echo server, sends a line, waits for the reply, and then
		2768	creates the next connection. This is a rather bad benchmark, as it doesn't
		2769	test the efficiency of the framework or much non-blocking I/O, but it is a
		2770	benchmark nevertheless.
		2771
		2772	name runtime
		2773	Lambda/select 0.330 sec
		2774	+ optimized 0.122 sec
		2775	Lambda/AnyEvent 0.327 sec
		2776	+ optimized 0.138 sec
		2777	Raw sockets/select 0.077 sec
		2778	POE/select, components 0.662 sec
		2779	POE/select, raw sockets 0.226 sec
		2780	POE/select, optimized 0.404 sec
		2781
		2782	AnyEvent/select/nb 0.085 sec
		2783	AnyEvent/EV/nb 0.068 sec
		2784	+state machine 0.134 sec
		2785
		2786	The benchmark is also a bit unfair (my fault): the IO::Lambda/POE
		2787	benchmarks actually make blocking connects and use 100% blocking I/O,
		2788	defeating the purpose of an event-based solution. All of the newly
		2789	written AnyEvent benchmarks use 100% non-blocking connects (using
		2790	AnyEvent::Socket::tcp_connect and the asynchronous pure perl DNS
		2791	resolver), so AnyEvent is at a disadvantage here, as non-blocking connects
		2792	generally require a lot more bookkeeping and event handling than blocking
		2793	connects (which involve a single syscall only).
		2794
		2795	The last AnyEvent benchmark additionally uses L<AnyEvent::Handle>, which
		2796	offers similar expressive power as POE and IO::Lambda, using conventional
		2797	Perl syntax. This means that both the echo server and the client are 100%
		2798	non-blocking, further placing it at a disadvantage.
		2799
		2800	As you can see, the AnyEvent + EV combination even beats the
		2801	hand-optimised "raw sockets benchmark", while AnyEvent + its pure perl
		2802	backend easily beats IO::Lambda and POE.
		2803
		2804	And even the 100% non-blocking version written using the high-level (and
		2805	slow :) L<AnyEvent::Handle> abstraction beats both POE and IO::Lambda
		2806	higher level ("unoptimised") abstractions by a large margin, even though
		2807	it does all of DNS, tcp-connect and socket I/O in a non-blocking way.
		2808
		2809	The two AnyEvent benchmarks programs can be found as F<eg/ae0.pl> and
		2810	F<eg/ae2.pl> in the AnyEvent distribution, the remaining benchmarks are
		2811	part of the IO::Lambda distribution and were used without any changes.
		2812
		2813
		2814	=head1 SIGNALS
		2815
		2816	AnyEvent currently installs handlers for these signals:
		2817
		2818	=over 4
		2819
		2820	=item SIGCHLD
		2821
		2822	A handler for C<SIGCHLD> is installed by AnyEvent's child watcher
		2823	emulation for event loops that do not support them natively. Also, some
		2824	event loops install a similar handler.
		2825
		2826	Additionally, when AnyEvent is loaded and SIGCHLD is set to IGNORE, then
		2827	AnyEvent will reset it to default, to avoid losing child exit statuses.
		2828
		2829	=item SIGPIPE
		2830
		2831	A no-op handler is installed for C<SIGPIPE> when C<$SIG{PIPE}> is C<undef>
		2832	when AnyEvent gets loaded.
		2833
		2834	The rationale for this is that AnyEvent users usually do not really depend
		2835	on SIGPIPE delivery (which is purely an optimisation for shell use, or
		2836	badly-written programs), but C<SIGPIPE> can cause spurious and rare
		2837	program exits as a lot of people do not expect C<SIGPIPE> when writing to
		2838	some random socket.
		2839
		2840	The rationale for installing a no-op handler as opposed to ignoring it is
		2841	that this way, the handler will be restored to defaults on exec.
		2842
		2843	Feel free to install your own handler, or reset it to defaults.
		2844
		2845	=back
		2846
		2847	=cut
		2848
		2849	undef $SIG{CHLD}
		2850	if $SIG{CHLD} eq 'IGNORE';
		2851
		2852	$SIG{PIPE} = sub { }
		2853	unless defined $SIG{PIPE};
		2854
		2855	=head1 RECOMMENDED/OPTIONAL MODULES
		2856
		2857	One of AnyEvent's main goals is to be 100% Pure-Perl(tm): only perl (and
		2858	its built-in modules) are required to use it.
		2859
		2860	That does not mean that AnyEvent won't take advantage of some additional
		2861	modules if they are installed.
		2862
		2863	This section explains which additional modules will be used, and how they
		2864	affect AnyEvent's operation.
		2865
		2866	=over 4
		2867
		2868	=item L<Async::Interrupt>
		2869
		2870	This slightly arcane module is used to implement fast signal handling: To
		2871	my knowledge, there is no way to do completely race-free and quick
		2872	signal handling in pure perl. To ensure that signals still get
		2873	delivered, AnyEvent will start an interval timer to wake up perl (and
		2874	catch the signals) with some delay (default is 10 seconds, look for
		2875	C<$AnyEvent::MAX_SIGNAL_LATENCY>).
		2876
		2877	If this module is available, then it will be used to implement signal
		2878	catching, which means that signals will not be delayed, and the event loop
		2879	will not be interrupted regularly, which is more efficient (and good for
		2880	battery life on laptops).
		2881
		2882	This affects not just the pure-perl event loop, but also other event loops
		2883	that have no signal handling on their own (e.g. Glib, Tk, Qt).
		2884
		2885	Some event loops (POE, Event, Event::Lib) offer signal watchers natively,
		2886	and either employ their own workarounds (POE) or use AnyEvent's workaround
		2887	(using C<$AnyEvent::MAX_SIGNAL_LATENCY>). Installing L<Async::Interrupt>
		2888	does nothing for those backends.
		2889
		2890	=item L<EV>
		2891
		2892	This module isn't really "optional", as it is simply one of the backend
		2893	event loops that AnyEvent can use. However, it is simply the best event
		2894	loop available in terms of features, speed and stability: It supports
		2895	the AnyEvent API optimally, implements all the watcher types in XS, does
		2896	automatic timer adjustments even when no monotonic clock is available,
		2897	can take avdantage of advanced kernel interfaces such as C<epoll> and
		2898	C<kqueue>, and is the fastest backend I<by far>. You can even embed
		2899	L<Glib>/L<Gtk2> in it (or vice versa, see L<EV::Glib> and L<Glib::EV>).
		2900
		2901	If you only use backends that rely on another event loop (e.g. C<Tk>),
		2902	then this module will do nothing for you.
		2903
		2904	=item L<Guard>
		2905
		2906	The guard module, when used, will be used to implement
		2907	C<AnyEvent::Util::guard>. This speeds up guards considerably (and uses a
		2908	lot less memory), but otherwise doesn't affect guard operation much. It is
		2909	purely used for performance.
		2910
		2911	=item L<JSON> and L<JSON::XS>
		2912
		2913	One of these modules is required when you want to read or write JSON data
		2914	via L<AnyEvent::Handle>. L<JSON> is also written in pure-perl, but can take
		2915	advantage of the ultra-high-speed L<JSON::XS> module when it is installed.
		2916
		2917	=item L<Net::SSLeay>
		2918
		2919	Implementing TLS/SSL in Perl is certainly interesting, but not very
		2920	worthwhile: If this module is installed, then L<AnyEvent::Handle> (with
		2921	the help of L<AnyEvent::TLS>), gains the ability to do TLS/SSL.
		2922
		2923	=item L<Time::HiRes>
		2924
		2925	This module is part of perl since release 5.008. It will be used when the
		2926	chosen event library does not come with a timing source of its own. The
		2927	pure-perl event loop (L<AnyEvent::Loop>) will additionally load it to
		2928	try to use a monotonic clock for timing stability.
		2929
		2930	=back
		2931
1724		2932
1725	=head1 FORK	2933	=head1 FORK
1726		2934
1727	Most event libraries are not fork-safe. The ones who are usually are	2935	Most event libraries are not fork-safe. The ones who are usually are
1728	because they rely on inefficient but fork-safe C<select> or C<poll>	2936	because they rely on inefficient but fork-safe C<select> or C<poll> calls
1729	calls. Only L<EV> is fully fork-aware.	2937	- higher performance APIs such as BSD's kqueue or the dreaded Linux epoll
		2938	are usually badly thought-out hacks that are incompatible with fork in
		2939	one way or another. Only L<EV> is fully fork-aware and ensures that you
		2940	continue event-processing in both parent and child (or both, if you know
		2941	what you are doing).
		2942
		2943	This means that, in general, you cannot fork and do event processing in
		2944	the child if the event library was initialised before the fork (which
		2945	usually happens when the first AnyEvent watcher is created, or the library
		2946	is loaded).
1730		2947
1731	If you have to fork, you must either do so I<before> creating your first	2948	If you have to fork, you must either do so I<before> creating your first
1732	watcher OR you must not use AnyEvent at all in the child.	2949	watcher OR you must not use AnyEvent at all in the child OR you must do
		2950	something completely out of the scope of AnyEvent.
		2951
		2952	The problem of doing event processing in the parent I<and> the child
		2953	is much more complicated: even for backends that I<are> fork-aware or
		2954	fork-safe, their behaviour is not usually what you want: fork clones all
		2955	watchers, that means all timers, I/O watchers etc. are active in both
		2956	parent and child, which is almost never what you want. USing C<exec>
		2957	to start worker children from some kind of manage rprocess is usually
		2958	preferred, because it is much easier and cleaner, at the expense of having
		2959	to have another binary.
1733		2960
1734		2961
1735	=head1 SECURITY CONSIDERATIONS	2962	=head1 SECURITY CONSIDERATIONS
1736		2963
1737	AnyEvent can be forced to load any event model via	2964	AnyEvent can be forced to load any event model via
…		…
1749	use AnyEvent;	2976	use AnyEvent;
1750		2977
1751	Similar considerations apply to $ENV{PERL_ANYEVENT_VERBOSE}, as that can	2978	Similar considerations apply to $ENV{PERL_ANYEVENT_VERBOSE}, as that can
1752	be used to probe what backend is used and gain other information (which is	2979	be used to probe what backend is used and gain other information (which is
1753	probably even less useful to an attacker than PERL_ANYEVENT_MODEL), and	2980	probably even less useful to an attacker than PERL_ANYEVENT_MODEL), and
1754	$ENV{PERL_ANYEGENT_STRICT}.	2981	$ENV{PERL_ANYEVENT_STRICT}.
		2982
		2983	Note that AnyEvent will remove I<all> environment variables starting with
		2984	C<PERL_ANYEVENT_> from C<%ENV> when it is loaded while taint mode is
		2985	enabled.
1755		2986
1756		2987
1757	=head1 BUGS	2988	=head1 BUGS
1758		2989
1759	Perl 5.8 has numerous memleaks that sometimes hit this module and are hard	2990	Perl 5.8 has numerous memleaks that sometimes hit this module and are hard
1760	to work around. If you suffer from memleaks, first upgrade to Perl 5.10	2991	to work around. If you suffer from memleaks, first upgrade to Perl 5.10
1761	and check wether the leaks still show up. (Perl 5.10.0 has other annoying	2992	and check wether the leaks still show up. (Perl 5.10.0 has other annoying
1762	mamleaks, such as leaking on C<map> and C<grep> but it is usually not as	2993	memleaks, such as leaking on C<map> and C<grep> but it is usually not as
1763	pronounced).	2994	pronounced).
1764		2995
1765		2996
1766	=head1 SEE ALSO	2997	=head1 SEE ALSO
1767		2998
1768	Utility functions: L<AnyEvent::Util>.	2999	Tutorial/Introduction: L<AnyEvent::Intro>.
1769		3000
1770	Event modules: L<EV>, L<EV::Glib>, L<Glib::EV>, L<Event>, L<Glib::Event>,	3001	FAQ: L<AnyEvent::FAQ>.
1771	L<Glib>, L<Tk>, L<Event::Lib>, L<Qt>, L<POE>.	3002
		3003	Utility functions: L<AnyEvent::Util> (misc. grab-bag), L<AnyEvent::Log>
		3004	(simply logging).
		3005
		3006	Development/Debugging: L<AnyEvent::Strict> (stricter checking),
		3007	L<AnyEvent::Debug> (interactive shell, watcher tracing).
		3008
		3009	Supported event modules: L<AnyEvent::Loop>, L<EV>, L<EV::Glib>,
		3010	L<Glib::EV>, L<Event>, L<Glib::Event>, L<Glib>, L<Tk>, L<Event::Lib>,
		3011	L<Qt>, L<POE>, L<FLTK>.
1772		3012
1773	Implementations: L<AnyEvent::Impl::EV>, L<AnyEvent::Impl::Event>,	3013	Implementations: L<AnyEvent::Impl::EV>, L<AnyEvent::Impl::Event>,
1774	L<AnyEvent::Impl::Glib>, L<AnyEvent::Impl::Tk>, L<AnyEvent::Impl::Perl>,	3014	L<AnyEvent::Impl::Glib>, L<AnyEvent::Impl::Tk>, L<AnyEvent::Impl::Perl>,
1775	L<AnyEvent::Impl::EventLib>, L<AnyEvent::Impl::Qt>,	3015	L<AnyEvent::Impl::EventLib>, L<AnyEvent::Impl::Qt>,
		3016	L<AnyEvent::Impl::POE>, L<AnyEvent::Impl::IOAsync>, L<Anyevent::Impl::Irssi>,
1776	L<AnyEvent::Impl::POE>.	3017	L<AnyEvent::Impl::FLTK>.
1777		3018
1778	Non-blocking file handles, sockets, TCP clients and	3019	Non-blocking handles, pipes, stream sockets, TCP clients and
1779	servers: L<AnyEvent::Handle>, L<AnyEvent::Socket>.	3020	servers: L<AnyEvent::Handle>, L<AnyEvent::Socket>, L<AnyEvent::TLS>.
		3021
		3022	Asynchronous File I/O: L<AnyEvent::IO>.
1780		3023
1781	Asynchronous DNS: L<AnyEvent::DNS>.	3024	Asynchronous DNS: L<AnyEvent::DNS>.
1782		3025
1783	Coroutine support: L<Coro>, L<Coro::AnyEvent>, L<Coro::EV>, L<Coro::Event>,	3026	Thread support: L<Coro>, L<Coro::AnyEvent>, L<Coro::EV>, L<Coro::Event>.
1784		3027
1785	Nontrivial usage examples: L<Net::FCP>, L<Net::XMPP2>, L<AnyEvent::DNS>.	3028	Nontrivial usage examples: L<AnyEvent::GPSD>, L<AnyEvent::IRC>,
		3029	L<AnyEvent::HTTP>.
1786		3030
1787		3031
1788	=head1 AUTHOR	3032	=head1 AUTHOR
1789		3033
1790	Marc Lehmann <schmorp@schmorp.de>	3034	Marc Lehmann <schmorp@schmorp.de>
1791	http://home.schmorp.de/	3035	http://anyevent.schmorp.de
1792		3036
1793	=cut	3037	=cut
1794		3038
1795	1	3039	1
1796		3040

Diff Legend

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