[ViewVC] Diff of: cvs/AnyEvent/README

Comparing AnyEvent/README (file contents):
Revision 1.40 by root, Tue Jun 23 23:37:32 2009 UTC vs.
Revision 1.59 by root, Tue Jan 5 10:45:25 2010 UTC

…		…
1	NAME	1	NAME
2	AnyEvent - provide framework for multiple event loops	2	AnyEvent - the DBI of event loop programming
3		3
4	EV, Event, Glib, Tk, Perl, Event::Lib, Qt and POE are various supported	4	EV, Event, Glib, Tk, Perl, Event::Lib, Irssi, rxvt-unicode, IO::Async,
5	event loops.	5	Qt and POE are various supported event loops/environments.
6		6
7	SYNOPSIS	7	SYNOPSIS
8	use AnyEvent;	8	use AnyEvent;
9		9
10	# file descriptor readable	10	# file descriptor readable
37		37
38	INTRODUCTION/TUTORIAL	38	INTRODUCTION/TUTORIAL
39	This manpage is mainly a reference manual. If you are interested in a	39	This manpage is mainly a reference manual. If you are interested in a
40	tutorial or some gentle introduction, have a look at the AnyEvent::Intro	40	tutorial or some gentle introduction, have a look at the AnyEvent::Intro
41	manpage.	41	manpage.
		42
		43	SUPPORT
		44	There is a mailinglist for discussing all things AnyEvent, and an IRC
		45	channel, too.
		46
		47	See the AnyEvent project page at the Schmorpforge Ta-Sa Software
		48	Repository, at <http://anyevent.schmorp.de>, for more info.
42		49
43	WHY YOU SHOULD USE THIS MODULE (OR NOT)	50	WHY YOU SHOULD USE THIS MODULE (OR NOT)
44	Glib, POE, IO::Async, Event... CPAN offers event models by the dozen	51	Glib, POE, IO::Async, Event... CPAN offers event models by the dozen
45	nowadays. So what is different about AnyEvent?	52	nowadays. So what is different about AnyEvent?
46		53
…		…
166	Note that "my $w; $w =" combination. This is necessary because in Perl,	173	Note that "my $w; $w =" combination. This is necessary because in Perl,
167	my variables are only visible after the statement in which they are	174	my variables are only visible after the statement in which they are
168	declared.	175	declared.
169		176
170	I/O WATCHERS	177	I/O WATCHERS
		178	$w = AnyEvent->io (
		179	fh => <filehandle_or_fileno>,
		180	poll => <"r" or "w">,
		181	cb => <callback>,
		182	);
		183
171	You can create an I/O watcher by calling the "AnyEvent->io" method with	184	You can create an I/O watcher by calling the "AnyEvent->io" method with
172	the following mandatory key-value pairs as arguments:	185	the following mandatory key-value pairs as arguments:
173		186
174	"fh" is the Perl file handle (not file descriptor) to watch for	187	"fh" is the Perl file handle (or a naked file descriptor) to watch for
175	events (AnyEvent might or might not keep a reference to this file	188	events (AnyEvent might or might not keep a reference to this file
176	handle). Note that only file handles pointing to things for which	189	handle). Note that only file handles pointing to things for which
177	non-blocking operation makes sense are allowed. This includes sockets,	190	non-blocking operation makes sense are allowed. This includes sockets,
178	most character devices, pipes, fifos and so on, but not for example	191	most character devices, pipes, fifos and so on, but not for example
179	files or block devices.	192	files or block devices.
…		…
203	warn "read: $input\n";	216	warn "read: $input\n";
204	undef $w;	217	undef $w;
205	});	218	});
206		219
207	TIME WATCHERS	220	TIME WATCHERS
		221	$w = AnyEvent->timer (after => <seconds>, cb => <callback>);
		222
		223	$w = AnyEvent->timer (
		224	after => <fractional_seconds>,
		225	interval => <fractional_seconds>,
		226	cb => <callback>,
		227	);
		228
208	You can create a time watcher by calling the "AnyEvent->timer" method	229	You can create a time watcher by calling the "AnyEvent->timer" method
209	with the following mandatory arguments:	230	with the following mandatory arguments:
210		231
211	"after" specifies after how many seconds (fractional values are	232	"after" specifies after how many seconds (fractional values are
212	supported) the callback should be invoked. "cb" is the callback to	233	supported) the callback should be invoked. "cb" is the callback to
…		…
333	time, which might affect timers and time-outs.	354	time, which might affect timers and time-outs.
334		355
335	When this is the case, you can call this method, which will update	356	When this is the case, you can call this method, which will update
336	the event loop's idea of "current time".	357	the event loop's idea of "current time".
337		358
		359	A typical example would be a script in a web server (e.g.
		360	"mod_perl") - when mod_perl executes the script, then the event loop
		361	will have the wrong idea about the "current time" (being potentially
		362	far in the past, when the script ran the last time). In that case
		363	you should arrange a call to "AnyEvent->now_update" each time the
		364	web server process wakes up again (e.g. at the start of your script,
		365	or in a handler).
		366
338	Note that updating the time might cause some events to be handled.	367	Note that updating the time might cause some events to be handled.
339		368
340	SIGNAL WATCHERS	369	SIGNAL WATCHERS
		370	$w = AnyEvent->signal (signal => <uppercase_signal_name>, cb => <callback>);
		371
341	You can watch for signals using a signal watcher, "signal" is the signal	372	You can watch for signals using a signal watcher, "signal" is the signal
342	name in uppercase and without any "SIG" prefix, "cb" is the Perl	373	name in uppercase and without any "SIG" prefix, "cb" is the Perl
343	callback to be invoked whenever a signal occurs.	374	callback to be invoked whenever a signal occurs.
344		375
345	Although the callback might get passed parameters, their value and	376	Although the callback might get passed parameters, their value and
…		…
350	invocation, and callback invocation will be synchronous. Synchronous	381	invocation, and callback invocation will be synchronous. Synchronous
351	means that it might take a while until the signal gets handled by the	382	means that it might take a while until the signal gets handled by the
352	process, but it is guaranteed not to interrupt any other callbacks.	383	process, but it is guaranteed not to interrupt any other callbacks.
353		384
354	The main advantage of using these watchers is that you can share a	385	The main advantage of using these watchers is that you can share a
355	signal between multiple watchers.	386	signal between multiple watchers, and AnyEvent will ensure that signals
		387	will not interrupt your program at bad times.
356		388
357	This watcher might use %SIG, so programs overwriting those signals	389	This watcher might use %SIG (depending on the event loop used), so
358	directly will likely not work correctly.	390	programs overwriting those signals directly will likely not work
		391	correctly.
359		392
360	Example: exit on SIGINT	393	Example: exit on SIGINT
361		394
362	my $w = AnyEvent->signal (signal => "INT", cb => sub { exit 1 });	395	my $w = AnyEvent->signal (signal => "INT", cb => sub { exit 1 });
363		396
		397	Restart Behaviour
		398	While restart behaviour is up to the event loop implementation, most
		399	will not restart syscalls (that includes Async::Interrupt and AnyEvent's
		400	pure perl implementation).
		401
		402	Safe/Unsafe Signals
		403	Perl signals can be either "safe" (synchronous to opcode handling) or
		404	"unsafe" (asynchronous) - the former might get delayed indefinitely, the
		405	latter might corrupt your memory.
		406
		407	AnyEvent signal handlers are, in addition, synchronous to the event
		408	loop, i.e. they will not interrupt your running perl program but will
		409	only be called as part of the normal event handling (just like timer,
		410	I/O etc. callbacks, too).
		411
		412	Signal Races, Delays and Workarounds
		413	Many event loops (e.g. Glib, Tk, Qt, IO::Async) do not support attaching
		414	callbacks to signals in a generic way, which is a pity, as you cannot do
		415	race-free signal handling in perl, requiring C libraries for this.
		416	AnyEvent will try to do it's best, which means in some cases, signals
		417	will be delayed. The maximum time a signal might be delayed is specified
		418	in $AnyEvent::MAX_SIGNAL_LATENCY (default: 10 seconds). This variable
		419	can be changed only before the first signal watcher is created, and
		420	should be left alone otherwise. This variable determines how often
		421	AnyEvent polls for signals (in case a wake-up was missed). Higher values
		422	will cause fewer spurious wake-ups, which is better for power and CPU
		423	saving.
		424
		425	All these problems can be avoided by installing the optional
		426	Async::Interrupt module, which works with most event loops. It will not
		427	work with inherently broken event loops such as Event or Event::Lib (and
		428	not with POE currently, as POE does it's own workaround with one-second
		429	latency). For those, you just have to suffer the delays.
		430
364	CHILD PROCESS WATCHERS	431	CHILD PROCESS WATCHERS
		432	$w = AnyEvent->child (pid => <process id>, cb => <callback>);
		433
365	You can also watch on a child process exit and catch its exit status.	434	You can also watch on a child process exit and catch its exit status.
366		435
367	The child process is specified by the "pid" argument (if set to 0, it	436	The child process is specified by the "pid" argument (one some backends,
368	watches for any child process exit). The watcher will triggered only	437	using 0 watches for any child process exit, on others this will croak).
369	when the child process has finished and an exit status is available, not	438	The watcher will be triggered only when the child process has finished
370	on any trace events (stopped/continued).	439	and an exit status is available, not on any trace events
		440	(stopped/continued).
371		441
372	The callback will be called with the pid and exit status (as returned by	442	The callback will be called with the pid and exit status (as returned by
373	waitpid), so unlike other watcher types, you can rely on child watcher	443	waitpid), so unlike other watcher types, you can rely on child watcher
374	callback arguments.	444	callback arguments.
375		445
…		…
380		450
381	There is a slight catch to child watchers, however: you usually start	451	There is a slight catch to child watchers, however: you usually start
382	them after the child process was created, and this means the process	452	them after the child process was created, and this means the process
383	could have exited already (and no SIGCHLD will be sent anymore).	453	could have exited already (and no SIGCHLD will be sent anymore).
384		454
385	Not all event models handle this correctly (POE doesn't), but even for	455	Not all event models handle this correctly (neither POE nor IO::Async
		456	do, see their AnyEvent::Impl manpages for details), but even for event
386	event models that do handle this correctly, they usually need to be	457	models that do handle this correctly, they usually need to be loaded
387	loaded before the process exits (i.e. before you fork in the first	458	before the process exits (i.e. before you fork in the first place).
388	place).	459	AnyEvent's pure perl event loop handles all cases correctly regardless
		460	of when you start the watcher.
389		461
390	This means you cannot create a child watcher as the very first thing in	462	This means you cannot create a child watcher as the very first thing in
391	an AnyEvent program, you have to create at least one watcher before	463	an AnyEvent program, you have to create at least one watcher before
392	you "fork" the child (alternatively, you can call "AnyEvent::detect").	464	you "fork" the child (alternatively, you can call "AnyEvent::detect").
		465
		466	As most event loops do not support waiting for child events, they will
		467	be emulated by AnyEvent in most cases, in which the latency and race
		468	problems mentioned in the description of signal watchers apply.
393		469
394	Example: fork a process and wait for it	470	Example: fork a process and wait for it
395		471
396	my $done = AnyEvent->condvar;	472	my $done = AnyEvent->condvar;
397		473
…		…
408		484
409	# do something else, then wait for process exit	485	# do something else, then wait for process exit
410	$done->recv;	486	$done->recv;
411		487
412	IDLE WATCHERS	488	IDLE WATCHERS
413	Sometimes there is a need to do something, but it is not so important to	489	$w = AnyEvent->idle (cb => <callback>);
414	do it instantly, but only when there is nothing better to do. This
415	"nothing better to do" is usually defined to be "no other events need
416	attention by the event loop".
417		490
418	Idle watchers ideally get invoked when the event loop has nothing better	491	Repeatedly invoke the callback after the process becomes idle, until
419	to do, just before it would block the process to wait for new events.	492	either the watcher is destroyed or new events have been detected.
420	Instead of blocking, the idle watcher is invoked.
421		493
422	Most event loops unfortunately do not really support idle watchers (only	494	Idle watchers are useful when there is a need to do something, but it is
		495	not so important (or wise) to do it instantly. The callback will be
		496	invoked only when there is "nothing better to do", which is usually
		497	defined as "all outstanding events have been handled and no new events
		498	have been detected". That means that idle watchers ideally get invoked
		499	when the event loop has just polled for new events but none have been
		500	detected. Instead of blocking to wait for more events, the idle watchers
		501	will be invoked.
		502
		503	Unfortunately, most event loops do not really support idle watchers
423	EV, Event and Glib do it in a usable fashion) - for the rest, AnyEvent	504	(only EV, Event and Glib do it in a usable fashion) - for the rest,
424	will simply call the callback "from time to time".	505	AnyEvent will simply call the callback "from time to time".
425		506
426	Example: read lines from STDIN, but only process them when the program	507	Example: read lines from STDIN, but only process them when the program
427	is otherwise idle:	508	is otherwise idle:
428		509
429	my @lines; # read data	510	my @lines; # read data
…		…
442	}	523	}
443	});	524	});
444	});	525	});
445		526
446	CONDITION VARIABLES	527	CONDITION VARIABLES
		528	$cv = AnyEvent->condvar;
		529
		530	$cv->send (<list>);
		531	my @res = $cv->recv;
		532
447	If you are familiar with some event loops you will know that all of them	533	If you are familiar with some event loops you will know that all of them
448	require you to run some blocking "loop", "run" or similar function that	534	require you to run some blocking "loop", "run" or similar function that
449	will actively watch for new events and call your callbacks.	535	will actively watch for new events and call your callbacks.
450		536
451	AnyEvent is different, it expects somebody else to run the event loop	537	AnyEvent is slightly different: it expects somebody else to run the
452	and will only block when necessary (usually when told by the user).	538	event loop and will only block when necessary (usually when told by the
		539	user).
453		540
454	The instrument to do that is called a "condition variable", so called	541	The instrument to do that is called a "condition variable", so called
455	because they represent a condition that must become true.	542	because they represent a condition that must become true.
456		543
		544	Now is probably a good time to look at the examples further below.
		545
457	Condition variables can be created by calling the "AnyEvent->condvar"	546	Condition variables can be created by calling the "AnyEvent->condvar"
458	method, usually without arguments. The only argument pair allowed is	547	method, usually without arguments. The only argument pair allowed is
459
460	"cb", which specifies a callback to be called when the condition	548	"cb", which specifies a callback to be called when the condition
461	variable becomes true, with the condition variable as the first argument	549	variable becomes true, with the condition variable as the first argument
462	(but not the results).	550	(but not the results).
463		551
464	After creation, the condition variable is "false" until it becomes	552	After creation, the condition variable is "false" until it becomes
…		…
469	Condition variables are similar to callbacks, except that you can	557	Condition variables are similar to callbacks, except that you can
470	optionally wait for them. They can also be called merge points - points	558	optionally wait for them. They can also be called merge points - points
471	in time where multiple outstanding events have been processed. And yet	559	in time where multiple outstanding events have been processed. And yet
472	another way to call them is transactions - each condition variable can	560	another way to call them is transactions - each condition variable can
473	be used to represent a transaction, which finishes at some point and	561	be used to represent a transaction, which finishes at some point and
474	delivers a result.	562	delivers a result. And yet some people know them as "futures" - a
		563	promise to compute/deliver something that you can wait for.
475		564
476	Condition variables are very useful to signal that something has	565	Condition variables are very useful to signal that something has
477	finished, for example, if you write a module that does asynchronous http	566	finished, for example, if you write a module that does asynchronous http
478	requests, then a condition variable would be the ideal candidate to	567	requests, then a condition variable would be the ideal candidate to
479	signal the availability of results. The user can either act when the	568	signal the availability of results. The user can either act when the
…		…
513	after => 1,	602	after => 1,
514	cb => sub { $result_ready->send },	603	cb => sub { $result_ready->send },
515	);	604	);
516		605
517	# this "blocks" (while handling events) till the callback	606	# this "blocks" (while handling events) till the callback
518	# calls send	607	# calls ->send
519	$result_ready->recv;	608	$result_ready->recv;
520		609
521	Example: wait for a timer, but take advantage of the fact that condition	610	Example: wait for a timer, but take advantage of the fact that condition
522	variables are also code references.	611	variables are also callable directly.
523		612
524	my $done = AnyEvent->condvar;	613	my $done = AnyEvent->condvar;
525	my $delay = AnyEvent->timer (after => 5, cb => $done);	614	my $delay = AnyEvent->timer (after => 5, cb => $done);
526	$done->recv;	615	$done->recv;
527		616
…		…
533		622
534	...	623	...
535		624
536	my @info = $couchdb->info->recv;	625	my @info = $couchdb->info->recv;
537		626
538	And this is how you would just ste a callback to be called whenever the	627	And this is how you would just set a callback to be called whenever the
539	results are available:	628	results are available:
540		629
541	$couchdb->info->cb (sub {	630	$couchdb->info->cb (sub {
542	my @info = $_[0]->recv;	631	my @info = $_[0]->recv;
543	});	632	});
…		…
558		647
559	Any arguments passed to the "send" call will be returned by all	648	Any arguments passed to the "send" call will be returned by all
560	future "->recv" calls.	649	future "->recv" calls.
561		650
562	Condition variables are overloaded so one can call them directly (as	651	Condition variables are overloaded so one can call them directly (as
563	a code reference). Calling them directly is the same as calling	652	if they were a code reference). Calling them directly is the same as
564	"send". Note, however, that many C-based event loops do not handle	653	calling "send".
565	overloading, so as tempting as it may be, passing a condition
566	variable instead of a callback does not work. Both the pure perl and
567	EV loops support overloading, however, as well as all functions that
568	use perl to invoke a callback (as in AnyEvent::Socket and
569	AnyEvent::DNS for example).
570		654
571	$cv->croak ($error)	655	$cv->croak ($error)
572	Similar to send, but causes all call's to "->recv" to invoke	656	Similar to send, but causes all call's to "->recv" to invoke
573	"Carp::croak" with the given error message/object/scalar.	657	"Carp::croak" with the given error message/object/scalar.
574		658
575	This can be used to signal any errors to the condition variable	659	This can be used to signal any errors to the condition variable
576	user/consumer.	660	user/consumer. Doing it this way instead of calling "croak" directly
		661	delays the error detetcion, but has the overwhelmign advantage that
		662	it diagnoses the error at the place where the result is expected,
		663	and not deep in some event clalback without connection to the actual
		664	code causing the problem.
577		665
578	$cv->begin ([group callback])	666	$cv->begin ([group callback])
579	$cv->end	667	$cv->end
580	These two methods are EXPERIMENTAL and MIGHT CHANGE.
581
582	These two methods can be used to combine many transactions/events	668	These two methods can be used to combine many transactions/events
583	into one. For example, a function that pings many hosts in parallel	669	into one. For example, a function that pings many hosts in parallel
584	might want to use a condition variable for the whole process.	670	might want to use a condition variable for the whole process.
585		671
586	Every call to "->begin" will increment a counter, and every call to	672	Every call to "->begin" will increment a counter, and every call to
587	"->end" will decrement it. If the counter reaches 0 in "->end", the	673	"->end" will decrement it. If the counter reaches 0 in "->end", the
588	(last) callback passed to "begin" will be executed. That callback is	674	(last) callback passed to "begin" will be executed, passing the
589	supposed to call "->send", but that is not required. If no	675	condvar as first argument. That callback is supposed to call
		676	"->send", but that is not required. If no group callback was set,
590	callback was set, "send" will be called without any arguments.	677	"send" will be called without any arguments.
591		678
592	Let's clarify this with the ping example:	679	You can think of "$cv->send" giving you an OR condition (one call
		680	sends), while "$cv->begin" and "$cv->end" giving you an AND
		681	condition (all "begin" calls must be "end"'ed before the condvar
		682	sends).
		683
		684	Let's start with a simple example: you have two I/O watchers (for
		685	example, STDOUT and STDERR for a program), and you want to wait for
		686	both streams to close before activating a condvar:
593		687
594	my $cv = AnyEvent->condvar;	688	my $cv = AnyEvent->condvar;
595		689
		690	$cv->begin; # first watcher
		691	my $w1 = AnyEvent->io (fh => $fh1, cb => sub {
		692	defined sysread $fh1, my $buf, 4096
		693	or $cv->end;
		694	});
		695
		696	$cv->begin; # second watcher
		697	my $w2 = AnyEvent->io (fh => $fh2, cb => sub {
		698	defined sysread $fh2, my $buf, 4096
		699	or $cv->end;
		700	});
		701
		702	$cv->recv;
		703
		704	This works because for every event source (EOF on file handle),
		705	there is one call to "begin", so the condvar waits for all calls to
		706	"end" before sending.
		707
		708	The ping example mentioned above is slightly more complicated, as
		709	the there are results to be passwd back, and the number of tasks
		710	that are begung can potentially be zero:
		711
		712	my $cv = AnyEvent->condvar;
		713
596	my %result;	714	my %result;
597	$cv->begin (sub { $cv->send (\%result) });	715	$cv->begin (sub { shift->send (\%result) });
598		716
599	for my $host (@list_of_hosts) {	717	for my $host (@list_of_hosts) {
600	$cv->begin;	718	$cv->begin;
601	ping_host_then_call_callback $host, sub {	719	ping_host_then_call_callback $host, sub {
602	$result{$host} = ...;	720	$result{$host} = ...;
…		…
617	the loop, which serves two important purposes: first, it sets the	735	the loop, which serves two important purposes: first, it sets the
618	callback to be called once the counter reaches 0, and second, it	736	callback to be called once the counter reaches 0, and second, it
619	ensures that "send" is called even when "no" hosts are being pinged	737	ensures that "send" is called even when "no" hosts are being pinged
620	(the loop doesn't execute once).	738	(the loop doesn't execute once).
621		739
622	This is the general pattern when you "fan out" into multiple	740	This is the general pattern when you "fan out" into multiple (but
623	subrequests: use an outer "begin"/"end" pair to set the callback and	741	potentially none) subrequests: use an outer "begin"/"end" pair to
624	ensure "end" is called at least once, and then, for each subrequest	742	set the callback and ensure "end" is called at least once, and then,
625	you start, call "begin" and for each subrequest you finish, call	743	for each subrequest you start, call "begin" and for each subrequest
626	"end".	744	you finish, call "end".
627		745
628	METHODS FOR CONSUMERS	746	METHODS FOR CONSUMERS
629	These methods should only be used by the consuming side, i.e. the code	747	These methods should only be used by the consuming side, i.e. the code
630	awaits the condition.	748	awaits the condition.
631		749
…		…
640	function will call "croak".	758	function will call "croak".
641		759
642	In list context, all parameters passed to "send" will be returned,	760	In list context, all parameters passed to "send" will be returned,
643	in scalar context only the first one will be returned.	761	in scalar context only the first one will be returned.
644		762
		763	Note that doing a blocking wait in a callback is not supported by
		764	any event loop, that is, recursive invocation of a blocking "->recv"
		765	is not allowed, and the "recv" call will "croak" if such a condition
		766	is detected. This condition can be slightly loosened by using
		767	Coro::AnyEvent, which allows you to do a blocking "->recv" from any
		768	thread that doesn't run the event loop itself.
		769
645	Not all event models support a blocking wait - some die in that case	770	Not all event models support a blocking wait - some die in that case
646	(programs might want to do that to stay interactive), so *if you are	771	(programs might want to do that to stay interactive), so *if you are
647	using this from a module, never require a blocking wait*, but let	772	using this from a module, never require a blocking wait*. Instead,
648	the caller decide whether the call will block or not (for example,	773	let the caller decide whether the call will block or not (for
649	by coupling condition variables with some kind of request results	774	example, by coupling condition variables with some kind of request
650	and supporting callbacks so the caller knows that getting the result	775	results and supporting callbacks so the caller knows that getting
651	will not block, while still supporting blocking waits if the caller	776	the result will not block, while still supporting blocking waits if
652	so desires).	777	the caller so desires).
653
654	Another reason never to "->recv" in a module is that you cannot
655	sensibly have two "->recv"'s in parallel, as that would require
656	multiple interpreters or coroutines/threads, none of which
657	"AnyEvent" can supply.
658
659	The Coro module, however, can and does supply coroutines and, in
660	fact, Coro::AnyEvent replaces AnyEvent's condvars by coroutine-safe
661	versions and also integrates coroutines into AnyEvent, making
662	blocking "->recv" calls perfectly safe as long as they are done from
663	another coroutine (one that doesn't run the event loop).
664		778
665	You can ensure that "-recv" never blocks by setting a callback and	779	You can ensure that "-recv" never blocks by setting a callback and
666	only calling "->recv" from within that callback (or at a later	780	only calling "->recv" from within that callback (or at a later
667	time). This will work even when the event loop does not support	781	time). This will work even when the event loop does not support
668	blocking waits otherwise.	782	blocking waits otherwise.
…		…
673		787
674	$cb = $cv->cb ($cb->($cv))	788	$cb = $cv->cb ($cb->($cv))
675	This is a mutator function that returns the callback set and	789	This is a mutator function that returns the callback set and
676	optionally replaces it before doing so.	790	optionally replaces it before doing so.
677		791
678	The callback will be called when the condition becomes "true", i.e.	792	The callback will be called when the condition becomes (or already
679	when "send" or "croak" are called, with the only argument being the	793	was) "true", i.e. when "send" or "croak" are called (or were
680	condition variable itself. Calling "recv" inside the callback or at	794	called), with the only argument being the condition variable itself.
		795	Calling "recv" inside the callback or at any later time is
681	any later time is guaranteed not to block.	796	guaranteed not to block.
		797
		798	SUPPORTED EVENT LOOPS/BACKENDS
		799	The available backend classes are (every class has its own manpage):
		800
		801	Backends that are autoprobed when no other event loop can be found.
		802	EV is the preferred backend when no other event loop seems to be in
		803	use. If EV is not installed, then AnyEvent will fall back to its own
		804	pure-perl implementation, which is available everywhere as it comes
		805	with AnyEvent itself.
		806
		807	AnyEvent::Impl::EV based on EV (interface to libev, best choice).
		808	AnyEvent::Impl::Perl pure-perl implementation, fast and portable.
		809
		810	Backends that are transparently being picked up when they are used.
		811	These will be used when they are currently loaded when the first
		812	watcher is created, in which case it is assumed that the application
		813	is using them. This means that AnyEvent will automatically pick the
		814	right backend when the main program loads an event module before
		815	anything starts to create watchers. Nothing special needs to be done
		816	by the main program.
		817
		818	AnyEvent::Impl::Event based on Event, very stable, few glitches.
		819	AnyEvent::Impl::Glib based on Glib, slow but very stable.
		820	AnyEvent::Impl::Tk based on Tk, very broken.
		821	AnyEvent::Impl::EventLib based on Event::Lib, leaks memory and worse.
		822	AnyEvent::Impl::POE based on POE, very slow, some limitations.
		823	AnyEvent::Impl::Irssi used when running within irssi.
		824
		825	Backends with special needs.
		826	Qt requires the Qt::Application to be instantiated first, but will
		827	otherwise be picked up automatically. As long as the main program
		828	instantiates the application before any AnyEvent watchers are
		829	created, everything should just work.
		830
		831	AnyEvent::Impl::Qt based on Qt.
		832
		833	Support for IO::Async can only be partial, as it is too broken and
		834	architecturally limited to even support the AnyEvent API. It also is
		835	the only event loop that needs the loop to be set explicitly, so it
		836	can only be used by a main program knowing about AnyEvent. See
		837	AnyEvent::Impl::Async for the gory details.
		838
		839	AnyEvent::Impl::IOAsync based on IO::Async, cannot be autoprobed.
		840
		841	Event loops that are indirectly supported via other backends.
		842	Some event loops can be supported via other modules:
		843
		844	There is no direct support for WxWidgets (Wx) or Prima.
		845
		846	WxWidgets has no support for watching file handles. However, you can
		847	use WxWidgets through the POE adaptor, as POE has a Wx backend that
		848	simply polls 20 times per second, which was considered to be too
		849	horrible to even consider for AnyEvent.
		850
		851	Prima is not supported as nobody seems to be using it, but it has a
		852	POE backend, so it can be supported through POE.
		853
		854	AnyEvent knows about both Prima and Wx, however, and will try to
		855	load POE when detecting them, in the hope that POE will pick them
		856	up, in which case everything will be automatic.
682		857
683	GLOBAL VARIABLES AND FUNCTIONS	858	GLOBAL VARIABLES AND FUNCTIONS
		859	These are not normally required to use AnyEvent, but can be useful to
		860	write AnyEvent extension modules.
		861
684	$AnyEvent::MODEL	862	$AnyEvent::MODEL
685	Contains "undef" until the first watcher is being created. Then it	863	Contains "undef" until the first watcher is being created, before
		864	the backend has been autodetected.
		865
686	contains the event model that is being used, which is the name of	866	Afterwards it contains the event model that is being used, which is
687	the Perl class implementing the model. This class is usually one of	867	the name of the Perl class implementing the model. This class is
688	the "AnyEvent::Impl:xxx" modules, but can be any other class in the	868	usually one of the "AnyEvent::Impl:xxx" modules, but can be any
689	case AnyEvent has been extended at runtime (e.g. in rxvt-unicode).	869	other class in the case AnyEvent has been extended at runtime (e.g.
690		870	in rxvt-unicode it will be "urxvt::anyevent").
691	The known classes so far are:
692
693	AnyEvent::Impl::EV based on EV (an interface to libev, best choice).
694	AnyEvent::Impl::Event based on Event, second best choice.
695	AnyEvent::Impl::Perl pure-perl implementation, fast and portable.
696	AnyEvent::Impl::Glib based on Glib, third-best choice.
697	AnyEvent::Impl::Tk based on Tk, very bad choice.
698	AnyEvent::Impl::Qt based on Qt, cannot be autoprobed (see its docs).
699	AnyEvent::Impl::EventLib based on Event::Lib, leaks memory and worse.
700	AnyEvent::Impl::POE based on POE, not generic enough for full support.
701
702	There is no support for WxWidgets, as WxWidgets has no support for
703	watching file handles. However, you can use WxWidgets through the
704	POE Adaptor, as POE has a Wx backend that simply polls 20 times per
705	second, which was considered to be too horrible to even consider for
706	AnyEvent. Likewise, other POE backends can be used by AnyEvent by
707	using it's adaptor.
708
709	AnyEvent knows about Prima and Wx and will try to use POE when
710	autodetecting them.
711		871
712	AnyEvent::detect	872	AnyEvent::detect
713	Returns $AnyEvent::MODEL, forcing autodetection of the event model	873	Returns $AnyEvent::MODEL, forcing autodetection of the event model
714	if necessary. You should only call this function right before you	874	if necessary. You should only call this function right before you
715	would have created an AnyEvent watcher anyway, that is, as late as	875	would have created an AnyEvent watcher anyway, that is, as late as
716	possible at runtime.	876	possible at runtime, and not e.g. while initialising of your module.
		877
		878	If you need to do some initialisation before AnyEvent watchers are
		879	created, use "post_detect".
717		880
718	$guard = AnyEvent::post_detect { BLOCK }	881	$guard = AnyEvent::post_detect { BLOCK }
719	Arranges for the code block to be executed as soon as the event	882	Arranges for the code block to be executed as soon as the event
720	model is autodetected (or immediately if this has already happened).	883	model is autodetected (or immediately if this has already happened).
721		884
		885	The block will be executed after the actual backend has been
		886	detected ($AnyEvent::MODEL is set), but before any watchers have
		887	been created, so it is possible to e.g. patch @AnyEvent::ISA or do
		888	other initialisations - see the sources of AnyEvent::Strict or
		889	AnyEvent::AIO to see how this is used.
		890
		891	The most common usage is to create some global watchers, without
		892	forcing event module detection too early, for example, AnyEvent::AIO
		893	creates and installs the global IO::AIO watcher in a "post_detect"
		894	block to avoid autodetecting the event module at load time.
		895
722	If called in scalar or list context, then it creates and returns an	896	If called in scalar or list context, then it creates and returns an
723	object that automatically removes the callback again when it is	897	object that automatically removes the callback again when it is
		898	destroyed (or "undef" when the hook was immediately executed). See
724	destroyed. See Coro::BDB for a case where this is useful.	899	AnyEvent::AIO for a case where this is useful.
		900
		901	Example: Create a watcher for the IO::AIO module and store it in
		902	$WATCHER. Only do so after the event loop is initialised, though.
		903
		904	our WATCHER;
		905
		906	my $guard = AnyEvent::post_detect {
		907	$WATCHER = AnyEvent->io (fh => IO::AIO::poll_fileno, poll => 'r', cb => \&IO::AIO::poll_cb);
		908	};
		909
		910	# the \|\|= is important in case post_detect immediately runs the block,
		911	# as to not clobber the newly-created watcher. assigning both watcher and
		912	# post_detect guard to the same variable has the advantage of users being
		913	# able to just C<undef $WATCHER> if the watcher causes them grief.
		914
		915	$WATCHER \|\|= $guard;
725		916
726	@AnyEvent::post_detect	917	@AnyEvent::post_detect
727	If there are any code references in this array (you can "push" to it	918	If there are any code references in this array (you can "push" to it
728	before or after loading AnyEvent), then they will called directly	919	before or after loading AnyEvent), then they will called directly
729	after the event loop has been chosen.	920	after the event loop has been chosen.
730		921
731	You should check $AnyEvent::MODEL before adding to this array,	922	You should check $AnyEvent::MODEL before adding to this array,
732	though: if it contains a true value then the event loop has already	923	though: if it is defined then the event loop has already been
733	been detected, and the array will be ignored.	924	detected, and the array will be ignored.
734		925
735	Best use "AnyEvent::post_detect { BLOCK }" instead.	926	Best use "AnyEvent::post_detect { BLOCK }" when your application
		927	allows it, as it takes care of these details.
		928
		929	This variable is mainly useful for modules that can do something
		930	useful when AnyEvent is used and thus want to know when it is
		931	initialised, but do not need to even load it by default. This array
		932	provides the means to hook into AnyEvent passively, without loading
		933	it.
		934
		935	Example: To load Coro::AnyEvent whenever Coro and AnyEvent are used
		936	together, you could put this into Coro (this is the actual code used
		937	by Coro to accomplish this):
		938
		939	if (defined $AnyEvent::MODEL) {
		940	# AnyEvent already initialised, so load Coro::AnyEvent
		941	require Coro::AnyEvent;
		942	} else {
		943	# AnyEvent not yet initialised, so make sure to load Coro::AnyEvent
		944	# as soon as it is
		945	push @AnyEvent::post_detect, sub { require Coro::AnyEvent };
		946	}
736		947
737	WHAT TO DO IN A MODULE	948	WHAT TO DO IN A MODULE
738	As a module author, you should "use AnyEvent" and call AnyEvent methods	949	As a module author, you should "use AnyEvent" and call AnyEvent methods
739	freely, but you should not load a specific event module or rely on it.	950	freely, but you should not load a specific event module or rely on it.
740		951
…		…
791	variable somewhere, waiting for it, and sending it when the program	1002	variable somewhere, waiting for it, and sending it when the program
792	should exit cleanly.	1003	should exit cleanly.
793		1004
794	OTHER MODULES	1005	OTHER MODULES
795	The following is a non-exhaustive list of additional modules that use	1006	The following is a non-exhaustive list of additional modules that use
796	AnyEvent and can therefore be mixed easily with other AnyEvent modules	1007	AnyEvent as a client and can therefore be mixed easily with other
797	in the same program. Some of the modules come with AnyEvent, some are	1008	AnyEvent modules and other event loops in the same program. Some of the
798	available via CPAN.	1009	modules come with AnyEvent, most are available via CPAN.
799		1010
800	AnyEvent::Util	1011	AnyEvent::Util
801	Contains various utility functions that replace often-used but	1012	Contains various utility functions that replace often-used but
802	blocking functions such as "inet_aton" by event-/callback-based	1013	blocking functions such as "inet_aton" by event-/callback-based
803	versions.	1014	versions.
…		…
809	more.	1020	more.
810		1021
811	AnyEvent::Handle	1022	AnyEvent::Handle
812	Provide read and write buffers, manages watchers for reads and	1023	Provide read and write buffers, manages watchers for reads and
813	writes, supports raw and formatted I/O, I/O queued and fully	1024	writes, supports raw and formatted I/O, I/O queued and fully
814	transparent and non-blocking SSL/TLS.	1025	transparent and non-blocking SSL/TLS (via AnyEvent::TLS.
815		1026
816	AnyEvent::DNS	1027	AnyEvent::DNS
817	Provides rich asynchronous DNS resolver capabilities.	1028	Provides rich asynchronous DNS resolver capabilities.
818		1029
819	AnyEvent::HTTP	1030	AnyEvent::HTTP
…		…
840		1051
841	AnyEvent::GPSD	1052	AnyEvent::GPSD
842	A non-blocking interface to gpsd, a daemon delivering GPS	1053	A non-blocking interface to gpsd, a daemon delivering GPS
843	information.	1054	information.
844		1055
		1056	AnyEvent::IRC
		1057	AnyEvent based IRC client module family (replacing the older
		1058	Net::IRC3).
		1059
		1060	AnyEvent::XMPP
		1061	AnyEvent based XMPP (Jabber protocol) module family (replacing the
		1062	older Net::XMPP2>.
		1063
845	AnyEvent::IGS	1064	AnyEvent::IGS
846	A non-blocking interface to the Internet Go Server protocol (used by	1065	A non-blocking interface to the Internet Go Server protocol (used by
847	App::IGS).	1066	App::IGS).
848		1067
849	AnyEvent::IRC
850	AnyEvent based IRC client module family (replacing the older
851	Net::IRC3).
852
853	Net::XMPP2
854	AnyEvent based XMPP (Jabber protocol) module family.
855
856	Net::FCP	1068	Net::FCP
857	AnyEvent-based implementation of the Freenet Client Protocol,	1069	AnyEvent-based implementation of the Freenet Client Protocol,
858	birthplace of AnyEvent.	1070	birthplace of AnyEvent.
859		1071
860	Event::ExecFlow	1072	Event::ExecFlow
861	High level API for event-based execution flow control.	1073	High level API for event-based execution flow control.
862		1074
863	Coro	1075	Coro
864	Has special support for AnyEvent via Coro::AnyEvent.	1076	Has special support for AnyEvent via Coro::AnyEvent.
865		1077
866	IO::Lambda	1078	SIMPLIFIED AE API
867	The lambda approach to I/O - don't ask, look there. Can use	1079	Starting with version 5.0, AnyEvent officially supports a second, much
868	AnyEvent.	1080	simpler, API that is designed to reduce the calling, typing and memory
		1081	overhead.
		1082
		1083	See the AE manpage for details.
869		1084
870	ERROR AND EXCEPTION HANDLING	1085	ERROR AND EXCEPTION HANDLING
871	In general, AnyEvent does not do any error handling - it relies on the	1086	In general, AnyEvent does not do any error handling - it relies on the
872	caller to do that if required. The AnyEvent::Strict module (see also the	1087	caller to do that if required. The AnyEvent::Strict module (see also the
873	"PERL_ANYEVENT_STRICT" environment variable, below) provides strict	1088	"PERL_ANYEVENT_STRICT" environment variable, below) provides strict
…		…
901	by "PERL_ANYEVENT_MODEL".	1116	by "PERL_ANYEVENT_MODEL".
902		1117
903	When set to 2 or higher, cause AnyEvent to report to STDERR which	1118	When set to 2 or higher, cause AnyEvent to report to STDERR which
904	event model it chooses.	1119	event model it chooses.
905		1120
		1121	When set to 8 or higher, then AnyEvent will report extra information
		1122	on which optional modules it loads and how it implements certain
		1123	features.
		1124
906	"PERL_ANYEVENT_STRICT"	1125	"PERL_ANYEVENT_STRICT"
907	AnyEvent does not do much argument checking by default, as thorough	1126	AnyEvent does not do much argument checking by default, as thorough
908	argument checking is very costly. Setting this variable to a true	1127	argument checking is very costly. Setting this variable to a true
909	value will cause AnyEvent to load "AnyEvent::Strict" and then to	1128	value will cause AnyEvent to load "AnyEvent::Strict" and then to
910	thoroughly check the arguments passed to most method calls. If it	1129	thoroughly check the arguments passed to most method calls. If it
911	finds any problems it will croak.	1130	finds any problems, it will croak.
912		1131
913	In other words, enables "strict" mode.	1132	In other words, enables "strict" mode.
914		1133
915	Unlike "use strict", it is definitely recommended ot keep it off in	1134	Unlike "use strict" (or it's modern cousin, "use common::sense", it
916	production. Keeping "PERL_ANYEVENT_STRICT=1" in your environment	1135	is definitely recommended to keep it off in production. Keeping
		1136	"PERL_ANYEVENT_STRICT=1" in your environment while developing
917	while developing programs can be very useful, however.	1137	programs can be very useful, however.
918		1138
919	"PERL_ANYEVENT_MODEL"	1139	"PERL_ANYEVENT_MODEL"
920	This can be used to specify the event model to be used by AnyEvent,	1140	This can be used to specify the event model to be used by AnyEvent,
921	before auto detection and -probing kicks in. It must be a string	1141	before auto detection and -probing kicks in. It must be a string
922	consisting entirely of ASCII letters. The string "AnyEvent::Impl::"	1142	consisting entirely of ASCII letters. The string "AnyEvent::Impl::"
…		…
963	EDNS0 in its DNS requests.	1183	EDNS0 in its DNS requests.
964		1184
965	"PERL_ANYEVENT_MAX_FORKS"	1185	"PERL_ANYEVENT_MAX_FORKS"
966	The maximum number of child processes that	1186	The maximum number of child processes that
967	"AnyEvent::Util::fork_call" will create in parallel.	1187	"AnyEvent::Util::fork_call" will create in parallel.
		1188
		1189	"PERL_ANYEVENT_MAX_OUTSTANDING_DNS"
		1190	The default value for the "max_outstanding" parameter for the
		1191	default DNS resolver - this is the maximum number of parallel DNS
		1192	requests that are sent to the DNS server.
		1193
		1194	"PERL_ANYEVENT_RESOLV_CONF"
		1195	The file to use instead of /etc/resolv.conf (or OS-specific
		1196	configuration) in the default resolver. When set to the empty
		1197	string, no default config will be used.
		1198
		1199	"PERL_ANYEVENT_CA_FILE", "PERL_ANYEVENT_CA_PATH".
		1200	When neither "ca_file" nor "ca_path" was specified during
		1201	AnyEvent::TLS context creation, and either of these environment
		1202	variables exist, they will be used to specify CA certificate
		1203	locations instead of a system-dependent default.
		1204
		1205	"PERL_ANYEVENT_AVOID_GUARD" and "PERL_ANYEVENT_AVOID_ASYNC_INTERRUPT"
		1206	When these are set to 1, then the respective modules are not loaded.
		1207	Mostly good for testing AnyEvent itself.
968		1208
969	SUPPLYING YOUR OWN EVENT MODEL INTERFACE	1209	SUPPLYING YOUR OWN EVENT MODEL INTERFACE
970	This is an advanced topic that you do not normally need to use AnyEvent	1210	This is an advanced topic that you do not normally need to use AnyEvent
971	in a module. This section is only of use to event loop authors who want	1211	in a module. This section is only of use to event loop authors who want
972	to provide AnyEvent compatibility.	1212	to provide AnyEvent compatibility.
…		…
1027	warn "read: $input\n"; # output what has been read	1267	warn "read: $input\n"; # output what has been read
1028	$cv->send if $input =~ /^q/i; # quit program if /^q/i	1268	$cv->send if $input =~ /^q/i; # quit program if /^q/i
1029	},	1269	},
1030	);	1270	);
1031		1271
1032	my $time_watcher; # can only be used once
1033
1034	sub new_timer {
1035	$timer = AnyEvent->timer (after => 1, cb => sub {	1272	my $time_watcher = AnyEvent->timer (after => 1, interval => 1, cb => sub {
1036	warn "timeout\n"; # print 'timeout' about every second	1273	warn "timeout\n"; # print 'timeout' at most every second
1037	&new_timer; # and restart the time
1038	});
1039	}	1274	});
1040
1041	new_timer; # create first timer
1042		1275
1043	$cv->recv; # wait until user enters /^q/i	1276	$cv->recv; # wait until user enters /^q/i
1044		1277
1045	REAL-WORLD EXAMPLE	1278	REAL-WORLD EXAMPLE
1046	Consider the Net::FCP module. It features (among others) the following	1279	Consider the Net::FCP module. It features (among others) the following
…		…
1173	through AnyEvent. The benchmark creates a lot of timers (with a zero	1406	through AnyEvent. The benchmark creates a lot of timers (with a zero
1174	timeout) and I/O watchers (watching STDOUT, a pty, to become writable,	1407	timeout) and I/O watchers (watching STDOUT, a pty, to become writable,
1175	which it is), lets them fire exactly once and destroys them again.	1408	which it is), lets them fire exactly once and destroys them again.
1176		1409
1177	Source code for this benchmark is found as eg/bench in the AnyEvent	1410	Source code for this benchmark is found as eg/bench in the AnyEvent
1178	distribution.	1411	distribution. It uses the AE interface, which makes a real difference
		1412	for the EV and Perl backends only.
1179		1413
1180	Explanation of the columns	1414	Explanation of the columns
1181	watcher is the number of event watchers created/destroyed. Since	1415	watcher is the number of event watchers created/destroyed. Since
1182	different event models feature vastly different performances, each event	1416	different event models feature vastly different performances, each event
1183	loop was given a number of watchers so that overall runtime is	1417	loop was given a number of watchers so that overall runtime is
…		…
1202	destroy is the time, in microseconds, that it takes to destroy a	1436	destroy is the time, in microseconds, that it takes to destroy a
1203	single watcher.	1437	single watcher.
1204		1438
1205	Results	1439	Results
1206	name watchers bytes create invoke destroy comment	1440	name watchers bytes create invoke destroy comment
1207	EV/EV 400000 224 0.47 0.35 0.27 EV native interface	1441	EV/EV 100000 223 0.47 0.43 0.27 EV native interface
1208	EV/Any 100000 224 2.88 0.34 0.27 EV + AnyEvent watchers	1442	EV/Any 100000 223 0.48 0.42 0.26 EV + AnyEvent watchers
1209	CoroEV/Any 100000 224 2.85 0.35 0.28 coroutines + Coro::Signal	1443	Coro::EV/Any 100000 223 0.47 0.42 0.26 coroutines + Coro::Signal
1210	Perl/Any 100000 452 4.13 0.73 0.95 pure perl implementation	1444	Perl/Any 100000 431 2.70 0.74 0.92 pure perl implementation
1211	Event/Event 16000 517 32.20 31.80 0.81 Event native interface	1445	Event/Event 16000 516 31.16 31.84 0.82 Event native interface
1212	Event/Any 16000 590 35.85 31.55 1.06 Event + AnyEvent watchers	1446	Event/Any 16000 1203 42.61 34.79 1.80 Event + AnyEvent watchers
		1447	IOAsync/Any 16000 1911 41.92 27.45 16.81 via IO::Async::Loop::IO_Poll
		1448	IOAsync/Any 16000 1726 40.69 26.37 15.25 via IO::Async::Loop::Epoll
1213	Glib/Any 16000 1357 102.33 12.31 51.00 quadratic behaviour	1449	Glib/Any 16000 1118 89.00 12.57 51.17 quadratic behaviour
1214	Tk/Any 2000 1860 27.20 66.31 14.00 SEGV with >> 2000 watchers	1450	Tk/Any 2000 1346 20.96 10.75 8.00 SEGV with >> 2000 watchers
1215	POE/Event 2000 6328 109.99 751.67 14.02 via POE::Loop::Event	1451	POE/Any 2000 6951 108.97 795.32 14.24 via POE::Loop::Event
1216	POE/Select 2000 6027 94.54 809.13 579.80 via POE::Loop::Select	1452	POE/Any 2000 6648 94.79 774.40 575.51 via POE::Loop::Select
1217		1453
1218	Discussion	1454	Discussion
1219	The benchmark does not measure scalability of the event loop very	1455	The benchmark does not measure scalability of the event loop very
1220	well. For example, a select-based event loop (such as the pure perl one)	1456	well. For example, a select-based event loop (such as the pure perl one)
1221	can never compete with an event loop that uses epoll when the number of	1457	can never compete with an event loop that uses epoll when the number of
…		…
1232	benchmark machine, handling an event takes roughly 1600 CPU cycles with	1468	benchmark machine, handling an event takes roughly 1600 CPU cycles with
1233	EV, 3100 CPU cycles with AnyEvent's pure perl loop and almost 3000000	1469	EV, 3100 CPU cycles with AnyEvent's pure perl loop and almost 3000000
1234	CPU cycles with POE.	1470	CPU cycles with POE.
1235		1471
1236	"EV" is the sole leader regarding speed and memory use, which are both	1472	"EV" is the sole leader regarding speed and memory use, which are both
1237	maximal/minimal, respectively. Even when going through AnyEvent, it uses	1473	maximal/minimal, respectively. When using the AE API there is zero
		1474	overhead (when going through the AnyEvent API create is about 5-6 times
		1475	slower, with other times being equal, so still uses far less memory than
1238	far less memory than any other event loop and is still faster than Event	1476	any other event loop and is still faster than Event natively).
1239	natively.
1240		1477
1241	The pure perl implementation is hit in a few sweet spots (both the	1478	The pure perl implementation is hit in a few sweet spots (both the
1242	constant timeout and the use of a single fd hit optimisations in the	1479	constant timeout and the use of a single fd hit optimisations in the
1243	perl interpreter and the backend itself). Nevertheless this shows that	1480	perl interpreter and the backend itself). Nevertheless this shows that
1244	it adds very little overhead in itself. Like any select-based backend	1481	it adds very little overhead in itself. Like any select-based backend
…		…
1246	few of them active), of course, but this was not subject of this	1483	few of them active), of course, but this was not subject of this
1247	benchmark.	1484	benchmark.
1248		1485
1249	The "Event" module has a relatively high setup and callback invocation	1486	The "Event" module has a relatively high setup and callback invocation
1250	cost, but overall scores in on the third place.	1487	cost, but overall scores in on the third place.
		1488
		1489	"IO::Async" performs admirably well, about on par with "Event", even
		1490	when using its pure perl backend.
1251		1491
1252	"Glib"'s memory usage is quite a bit higher, but it features a faster	1492	"Glib"'s memory usage is quite a bit higher, but it features a faster
1253	callback invocation and overall ends up in the same class as "Event".	1493	callback invocation and overall ends up in the same class as "Event".
1254	However, Glib scales extremely badly, doubling the number of watchers	1494	However, Glib scales extremely badly, doubling the number of watchers
1255	increases the processing time by more than a factor of four, making it	1495	increases the processing time by more than a factor of four, making it
…		…
1311	In this benchmark, we use 10000 socket pairs (20000 sockets), of which	1551	In this benchmark, we use 10000 socket pairs (20000 sockets), of which
1312	100 (1%) are active. This mirrors the activity of large servers with	1552	100 (1%) are active. This mirrors the activity of large servers with
1313	many connections, most of which are idle at any one point in time.	1553	many connections, most of which are idle at any one point in time.
1314		1554
1315	Source code for this benchmark is found as eg/bench2 in the AnyEvent	1555	Source code for this benchmark is found as eg/bench2 in the AnyEvent
1316	distribution.	1556	distribution. It uses the AE interface, which makes a real difference
		1557	for the EV and Perl backends only.
1317		1558
1318	Explanation of the columns	1559	Explanation of the columns
1319	sockets is the number of sockets, and twice the number of "servers"	1560	sockets is the number of sockets, and twice the number of "servers"
1320	(as each server has a read and write socket end).	1561	(as each server has a read and write socket end).
1321		1562
…		…
1326	single "request", that is, reading the token from the pipe and	1567	single "request", that is, reading the token from the pipe and
1327	forwarding it to another server. This includes deleting the old timeout	1568	forwarding it to another server. This includes deleting the old timeout
1328	and creating a new one that moves the timeout into the future.	1569	and creating a new one that moves the timeout into the future.
1329		1570
1330	Results	1571	Results
1331	name sockets create request	1572	name sockets create request
1332	EV 20000 69.01 11.16	1573	EV 20000 62.66 7.99
1333	Perl 20000 73.32 35.87	1574	Perl 20000 68.32 32.64
1334	Event 20000 212.62 257.32	1575	IOAsync 20000 174.06 101.15 epoll
1335	Glib 20000 651.16 1896.30	1576	IOAsync 20000 174.67 610.84 poll
		1577	Event 20000 202.69 242.91
		1578	Glib 20000 557.01 1689.52
1336	POE 20000 349.67 12317.24 uses POE::Loop::Event	1579	POE 20000 341.54 12086.32 uses POE::Loop::Event
1337		1580
1338	Discussion	1581	Discussion
1339	This benchmark does measure scalability and overall performance of the	1582	This benchmark does measure scalability and overall performance of the
1340	particular event loop.	1583	particular event loop.
1341		1584
1342	EV is again fastest. Since it is using epoll on my system, the setup	1585	EV is again fastest. Since it is using epoll on my system, the setup
1343	time is relatively high, though.	1586	time is relatively high, though.
1344		1587
1345	Perl surprisingly comes second. It is much faster than the C-based event	1588	Perl surprisingly comes second. It is much faster than the C-based event
1346	loops Event and Glib.	1589	loops Event and Glib.
		1590
		1591	IO::Async performs very well when using its epoll backend, and still
		1592	quite good compared to Glib when using its pure perl backend.
1347		1593
1348	Event suffers from high setup time as well (look at its code and you	1594	Event suffers from high setup time as well (look at its code and you
1349	will understand why). Callback invocation also has a high overhead	1595	will understand why). Callback invocation also has a high overhead
1350	compared to the "$_->() for .."-style loop that the Perl event loop	1596	compared to the "$_->() for .."-style loop that the Perl event loop
1351	uses. Event uses select or poll in basically all documented	1597	uses. Event uses select or poll in basically all documented
…		…
1407	THE IO::Lambda BENCHMARK	1653	THE IO::Lambda BENCHMARK
1408	Recently I was told about the benchmark in the IO::Lambda manpage, which	1654	Recently I was told about the benchmark in the IO::Lambda manpage, which
1409	could be misinterpreted to make AnyEvent look bad. In fact, the	1655	could be misinterpreted to make AnyEvent look bad. In fact, the
1410	benchmark simply compares IO::Lambda with POE, and IO::Lambda looks	1656	benchmark simply compares IO::Lambda with POE, and IO::Lambda looks
1411	better (which shouldn't come as a surprise to anybody). As such, the	1657	better (which shouldn't come as a surprise to anybody). As such, the
1412	benchmark is fine, and shows that the AnyEvent backend from IO::Lambda	1658	benchmark is fine, and mostly shows that the AnyEvent backend from
1413	isn't very optimal. But how would AnyEvent compare when used without the	1659	IO::Lambda isn't very optimal. But how would AnyEvent compare when used
1414	extra baggage? To explore this, I wrote the equivalent benchmark for	1660	without the extra baggage? To explore this, I wrote the equivalent
1415	AnyEvent.	1661	benchmark for AnyEvent.
1416		1662
1417	The benchmark itself creates an echo-server, and then, for 500 times,	1663	The benchmark itself creates an echo-server, and then, for 500 times,
1418	connects to the echo server, sends a line, waits for the reply, and then	1664	connects to the echo server, sends a line, waits for the reply, and then
1419	creates the next connection. This is a rather bad benchmark, as it	1665	creates the next connection. This is a rather bad benchmark, as it
1420	doesn't test the efficiency of the framework, but it is a benchmark	1666	doesn't test the efficiency of the framework or much non-blocking I/O,
1421	nevertheless.	1667	but it is a benchmark nevertheless.
1422		1668
1423	name runtime	1669	name runtime
1424	Lambda/select 0.330 sec	1670	Lambda/select 0.330 sec
1425	+ optimized 0.122 sec	1671	+ optimized 0.122 sec
1426	Lambda/AnyEvent 0.327 sec	1672	Lambda/AnyEvent 0.327 sec
…		…
1432		1678
1433	AnyEvent/select/nb 0.085 sec	1679	AnyEvent/select/nb 0.085 sec
1434	AnyEvent/EV/nb 0.068 sec	1680	AnyEvent/EV/nb 0.068 sec
1435	+state machine 0.134 sec	1681	+state machine 0.134 sec
1436		1682
1437	The benchmark is also a bit unfair (my fault) - the IO::Lambda	1683	The benchmark is also a bit unfair (my fault): the IO::Lambda/POE
1438	benchmarks actually make blocking connects and use 100% blocking I/O,	1684	benchmarks actually make blocking connects and use 100% blocking I/O,
1439	defeating the purpose of an event-based solution. All of the newly	1685	defeating the purpose of an event-based solution. All of the newly
1440	written AnyEvent benchmarks use 100% non-blocking connects (using	1686	written AnyEvent benchmarks use 100% non-blocking connects (using
1441	AnyEvent::Socket::tcp_connect and the asynchronous pure perl DNS	1687	AnyEvent::Socket::tcp_connect and the asynchronous pure perl DNS
1442	resolver), so AnyEvent is at a disadvantage here as non-blocking	1688	resolver), so AnyEvent is at a disadvantage here, as non-blocking
1443	connects generally require a lot more bookkeeping and event handling	1689	connects generally require a lot more bookkeeping and event handling
1444	than blocking connects (which involve a single syscall only).	1690	than blocking connects (which involve a single syscall only).
1445		1691
1446	The last AnyEvent benchmark additionally uses AnyEvent::Handle, which	1692	The last AnyEvent benchmark additionally uses AnyEvent::Handle, which
1447	offers similar expressive power as POE and IO::Lambda (using	1693	offers similar expressive power as POE and IO::Lambda, using
1448	conventional Perl syntax), which means both the echo server and the	1694	conventional Perl syntax. This means that both the echo server and the
1449	client are 100% non-blocking w.r.t. I/O, further placing it at a	1695	client are 100% non-blocking, further placing it at a disadvantage.
1450	disadvantage.
1451		1696
1452	As you can see, AnyEvent + EV even beats the hand-optimised "raw sockets	1697	As you can see, the AnyEvent + EV combination even beats the
1453	benchmark", while AnyEvent + its pure perl backend easily beats	1698	hand-optimised "raw sockets benchmark", while AnyEvent + its pure perl
1454	IO::Lambda and POE.	1699	backend easily beats IO::Lambda and POE.
1455		1700
1456	And even the 100% non-blocking version written using the high-level (and	1701	And even the 100% non-blocking version written using the high-level (and
1457	slow :) AnyEvent::Handle abstraction beats both POE and IO::Lambda, even	1702	slow :) AnyEvent::Handle abstraction beats both POE and IO::Lambda
		1703	higher level ("unoptimised") abstractions by a large margin, even though
1458	thought it does all of DNS, tcp-connect and socket I/O in a non-blocking	1704	it does all of DNS, tcp-connect and socket I/O in a non-blocking way.
1459	way.
1460		1705
1461	The two AnyEvent benchmarks can be found as eg/ae0.pl and eg/ae2.pl in	1706	The two AnyEvent benchmarks programs can be found as eg/ae0.pl and
1462	the AnyEvent distribution, the remaining benchmarks are part of the	1707	eg/ae2.pl in the AnyEvent distribution, the remaining benchmarks are
1463	IO::lambda distribution and were used without any changes.	1708	part of the IO::Lambda distribution and were used without any changes.
1464		1709
1465	SIGNALS	1710	SIGNALS
1466	AnyEvent currently installs handlers for these signals:	1711	AnyEvent currently installs handlers for these signals:
1467		1712
1468	SIGCHLD	1713	SIGCHLD
1469	A handler for "SIGCHLD" is installed by AnyEvent's child watcher	1714	A handler for "SIGCHLD" is installed by AnyEvent's child watcher
1470	emulation for event loops that do not support them natively. Also,	1715	emulation for event loops that do not support them natively. Also,
1471	some event loops install a similar handler.	1716	some event loops install a similar handler.
		1717
		1718	Additionally, when AnyEvent is loaded and SIGCHLD is set to IGNORE,
		1719	then AnyEvent will reset it to default, to avoid losing child exit
		1720	statuses.
1472		1721
1473	SIGPIPE	1722	SIGPIPE
1474	A no-op handler is installed for "SIGPIPE" when $SIG{PIPE} is	1723	A no-op handler is installed for "SIGPIPE" when $SIG{PIPE} is
1475	"undef" when AnyEvent gets loaded.	1724	"undef" when AnyEvent gets loaded.
1476		1725
…		…
1484	it is that this way, the handler will be restored to defaults on	1733	it is that this way, the handler will be restored to defaults on
1485	exec.	1734	exec.
1486		1735
1487	Feel free to install your own handler, or reset it to defaults.	1736	Feel free to install your own handler, or reset it to defaults.
1488		1737
		1738	RECOMMENDED/OPTIONAL MODULES
		1739	One of AnyEvent's main goals is to be 100% Pure-Perl(tm): only perl (and
		1740	it's built-in modules) are required to use it.
		1741
		1742	That does not mean that AnyEvent won't take advantage of some additional
		1743	modules if they are installed.
		1744
		1745	This section explains which additional modules will be used, and how
		1746	they affect AnyEvent's operation.
		1747
		1748	Async::Interrupt
		1749	This slightly arcane module is used to implement fast signal
		1750	handling: To my knowledge, there is no way to do completely
		1751	race-free and quick signal handling in pure perl. To ensure that
		1752	signals still get delivered, AnyEvent will start an interval timer
		1753	to wake up perl (and catch the signals) with some delay (default is
		1754	10 seconds, look for $AnyEvent::MAX_SIGNAL_LATENCY).
		1755
		1756	If this module is available, then it will be used to implement
		1757	signal catching, which means that signals will not be delayed, and
		1758	the event loop will not be interrupted regularly, which is more
		1759	efficient (and good for battery life on laptops).
		1760
		1761	This affects not just the pure-perl event loop, but also other event
		1762	loops that have no signal handling on their own (e.g. Glib, Tk, Qt).
		1763
		1764	Some event loops (POE, Event, Event::Lib) offer signal watchers
		1765	natively, and either employ their own workarounds (POE) or use
		1766	AnyEvent's workaround (using $AnyEvent::MAX_SIGNAL_LATENCY).
		1767	Installing Async::Interrupt does nothing for those backends.
		1768
		1769	EV This module isn't really "optional", as it is simply one of the
		1770	backend event loops that AnyEvent can use. However, it is simply the
		1771	best event loop available in terms of features, speed and stability:
		1772	It supports the AnyEvent API optimally, implements all the watcher
		1773	types in XS, does automatic timer adjustments even when no monotonic
		1774	clock is available, can take avdantage of advanced kernel interfaces
		1775	such as "epoll" and "kqueue", and is the fastest backend by far.
		1776	You can even embed Glib/Gtk2 in it (or vice versa, see EV::Glib and
		1777	Glib::EV).
		1778
		1779	Guard
		1780	The guard module, when used, will be used to implement
		1781	"AnyEvent::Util::guard". This speeds up guards considerably (and
		1782	uses a lot less memory), but otherwise doesn't affect guard
		1783	operation much. It is purely used for performance.
		1784
		1785	JSON and JSON::XS
		1786	One of these modules is required when you want to read or write JSON
		1787	data via AnyEvent::Handle. It is also written in pure-perl, but can
		1788	take advantage of the ultra-high-speed JSON::XS module when it is
		1789	installed.
		1790
		1791	In fact, AnyEvent::Handle will use JSON::XS by default if it is
		1792	installed.
		1793
		1794	Net::SSLeay
		1795	Implementing TLS/SSL in Perl is certainly interesting, but not very
		1796	worthwhile: If this module is installed, then AnyEvent::Handle (with
		1797	the help of AnyEvent::TLS), gains the ability to do TLS/SSL.
		1798
		1799	Time::HiRes
		1800	This module is part of perl since release 5.008. It will be used
		1801	when the chosen event library does not come with a timing source on
		1802	it's own. The pure-perl event loop (AnyEvent::Impl::Perl) will
		1803	additionally use it to try to use a monotonic clock for timing
		1804	stability.
		1805
1489	FORK	1806	FORK
1490	Most event libraries are not fork-safe. The ones who are usually are	1807	Most event libraries are not fork-safe. The ones who are usually are
1491	because they rely on inefficient but fork-safe "select" or "poll" calls.	1808	because they rely on inefficient but fork-safe "select" or "poll" calls
1492	Only EV is fully fork-aware.	1809	- higher performance APIs such as BSD's kqueue or the dreaded Linux
		1810	epoll are usually badly thought-out hacks that are incompatible with
		1811	fork in one way or another. Only EV is fully fork-aware and ensures that
		1812	you continue event-processing in both parent and child (or both, if you
		1813	know what you are doing).
		1814
		1815	This means that, in general, you cannot fork and do event processing in
		1816	the child if the event library was initialised before the fork (which
		1817	usually happens when the first AnyEvent watcher is created, or the
		1818	library is loaded).
1493		1819
1494	If you have to fork, you must either do so before creating your first	1820	If you have to fork, you must either do so before creating your first
1495	watcher OR you must not use AnyEvent at all in the child.	1821	watcher OR you must not use AnyEvent at all in the child OR you must do
		1822	something completely out of the scope of AnyEvent.
		1823
		1824	The problem of doing event processing in the parent and the child is
		1825	much more complicated: even for backends that are fork-aware or
		1826	fork-safe, their behaviour is not usually what you want: fork clones all
		1827	watchers, that means all timers, I/O watchers etc. are active in both
		1828	parent and child, which is almost never what you want. USing "exec" to
		1829	start worker children from some kind of manage rprocess is usually
		1830	preferred, because it is much easier and cleaner, at the expense of
		1831	having to have another binary.
1496		1832
1497	SECURITY CONSIDERATIONS	1833	SECURITY CONSIDERATIONS
1498	AnyEvent can be forced to load any event model via	1834	AnyEvent can be forced to load any event model via
1499	$ENV{PERL_ANYEVENT_MODEL}. While this cannot (to my knowledge) be used	1835	$ENV{PERL_ANYEVENT_MODEL}. While this cannot (to my knowledge) be used
1500	to execute arbitrary code or directly gain access, it can easily be used	1836	to execute arbitrary code or directly gain access, it can easily be used
…		…
1512	Similar considerations apply to $ENV{PERL_ANYEVENT_VERBOSE}, as that can	1848	Similar considerations apply to $ENV{PERL_ANYEVENT_VERBOSE}, as that can
1513	be used to probe what backend is used and gain other information (which	1849	be used to probe what backend is used and gain other information (which
1514	is probably even less useful to an attacker than PERL_ANYEVENT_MODEL),	1850	is probably even less useful to an attacker than PERL_ANYEVENT_MODEL),
1515	and $ENV{PERL_ANYEVENT_STRICT}.	1851	and $ENV{PERL_ANYEVENT_STRICT}.
1516		1852
		1853	Note that AnyEvent will remove all environment variables starting with
		1854	"PERL_ANYEVENT_" from %ENV when it is loaded while taint mode is
		1855	enabled.
		1856
1517	BUGS	1857	BUGS
1518	Perl 5.8 has numerous memleaks that sometimes hit this module and are	1858	Perl 5.8 has numerous memleaks that sometimes hit this module and are
1519	hard to work around. If you suffer from memleaks, first upgrade to Perl	1859	hard to work around. If you suffer from memleaks, first upgrade to Perl
1520	5.10 and check wether the leaks still show up. (Perl 5.10.0 has other	1860	5.10 and check wether the leaks still show up. (Perl 5.10.0 has other
1521	annoying memleaks, such as leaking on "map" and "grep" but it is usually	1861	annoying memleaks, such as leaking on "map" and "grep" but it is usually
…		…
1527	Event modules: EV, EV::Glib, Glib::EV, Event, Glib::Event, Glib, Tk,	1867	Event modules: EV, EV::Glib, Glib::EV, Event, Glib::Event, Glib, Tk,
1528	Event::Lib, Qt, POE.	1868	Event::Lib, Qt, POE.
1529		1869
1530	Implementations: AnyEvent::Impl::EV, AnyEvent::Impl::Event,	1870	Implementations: AnyEvent::Impl::EV, AnyEvent::Impl::Event,
1531	AnyEvent::Impl::Glib, AnyEvent::Impl::Tk, AnyEvent::Impl::Perl,	1871	AnyEvent::Impl::Glib, AnyEvent::Impl::Tk, AnyEvent::Impl::Perl,
1532	AnyEvent::Impl::EventLib, AnyEvent::Impl::Qt, AnyEvent::Impl::POE.	1872	AnyEvent::Impl::EventLib, AnyEvent::Impl::Qt, AnyEvent::Impl::POE,
		1873	AnyEvent::Impl::IOAsync, Anyevent::Impl::Irssi.
1533		1874
1534	Non-blocking file handles, sockets, TCP clients and servers:	1875	Non-blocking file handles, sockets, TCP clients and servers:
1535	AnyEvent::Handle, AnyEvent::Socket.	1876	AnyEvent::Handle, AnyEvent::Socket, AnyEvent::TLS.
1536		1877
1537	Asynchronous DNS: AnyEvent::DNS.	1878	Asynchronous DNS: AnyEvent::DNS.
1538		1879
1539	Coroutine support: Coro, Coro::AnyEvent, Coro::EV, Coro::Event,	1880	Coroutine support: Coro, Coro::AnyEvent, Coro::EV, Coro::Event,
1540		1881
1541	Nontrivial usage examples: Net::FCP, Net::XMPP2, AnyEvent::DNS.	1882	Nontrivial usage examples: AnyEvent::GPSD, AnyEvent::XMPP,
		1883	AnyEvent::HTTP.
1542		1884
1543	AUTHOR	1885	AUTHOR
1544	Marc Lehmann <schmorp@schmorp.de>	1886	Marc Lehmann <schmorp@schmorp.de>
1545	http://home.schmorp.de/	1887	http://home.schmorp.de/
1546		1888

Diff Legend

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

Comparing AnyEvent/README (file contents): Revision 1.40 by root, Tue Jun 23 23:37:32 2009 UTC vs. Revision 1.59 by root, Tue Jan 5 10:45:25 2010 UTC

Diff Legend

Comparing AnyEvent/README (file contents):
Revision 1.40 by root, Tue Jun 23 23:37:32 2009 UTC vs.
Revision 1.59 by root, Tue Jan 5 10:45:25 2010 UTC