[ViewVC] Diff of: cvs/AnyEvent-MP/MP.pm

Comparing AnyEvent-MP/MP.pm (file contents):
Revision 1.34 by root, Wed Aug 5 23:50:46 2009 UTC vs.
Revision 1.47 by root, Thu Aug 13 01:57:10 2009 UTC

…		…
8		8
9	$NODE # contains this node's noderef	9	$NODE # contains this node's noderef
10	NODE # returns this node's noderef	10	NODE # returns this node's noderef
11	NODE $port # returns the noderef of the port	11	NODE $port # returns the noderef of the port
12		12
		13	$SELF # receiving/own port id in rcv callbacks
		14
		15	# ports are message endpoints
		16
		17	# sending messages
13	snd $port, type => data...;	18	snd $port, type => data...;
		19	snd $port, @msg;
		20	snd @msg_with_first_element_being_a_port;
14		21
15	$SELF # receiving/own port id in rcv callbacks	22	# miniports
		23	my $miniport = port { my @msg = @_; 0 };
16		24
		25	# full ports
		26	my $port = port;
17	rcv $port, smartmatch => $cb->($port, @msg);	27	rcv $port, smartmatch => $cb->(@msg);
18
19	# examples:
20	rcv $port2, ping => sub { snd $_[0], "pong"; 0 };	28	rcv $port, ping => sub { snd $_[0], "pong"; 0 };
21	rcv $port1, pong => sub { warn "pong received\n" };	29	rcv $port, pong => sub { warn "pong received\n"; 0 };
22	snd $port2, ping => $port1;	30
		31	# remote ports
		32	my $port = spawn $node, $initfunc, @initdata;
23		33
24	# more, smarter, matches (_any_ is exported by this module)	34	# more, smarter, matches (_any_ is exported by this module)
25	rcv $port, [child_died => $pid] => sub { ...	35	rcv $port, [child_died => $pid] => sub { ...
26	rcv $port, [_any_, _any_, 3] => sub { .. $_[2] is 3	36	rcv $port, [_any_, _any_, 3] => sub { .. $_[2] is 3
27		37
		38	# monitoring
		39	mon $port, $cb->(@msg) # callback is invoked on death
		40	mon $port, $otherport # kill otherport on abnormal death
		41	mon $port, $otherport, @msg # send message on death
		42
		43	=head1 CURRENT STATUS
		44
		45	AnyEvent::MP - stable API, should work
		46	AnyEvent::MP::Intro - outdated
		47	AnyEvent::MP::Kernel - WIP
		48	AnyEvent::MP::Transport - mostly stable
		49
		50	stay tuned.
		51
28	=head1 DESCRIPTION	52	=head1 DESCRIPTION
29		53
30	This module (-family) implements a simple message passing framework.	54	This module (-family) implements a simple message passing framework.
31		55
32	Despite its simplicity, you can securely message other processes running	56	Despite its simplicity, you can securely message other processes running
…		…
35	For an introduction to this module family, see the L<AnyEvent::MP::Intro>	59	For an introduction to this module family, see the L<AnyEvent::MP::Intro>
36	manual page.	60	manual page.
37		61
38	At the moment, this module family is severly broken and underdocumented,	62	At the moment, this module family is severly broken and underdocumented,
39	so do not use. This was uploaded mainly to reserve the CPAN namespace -	63	so do not use. This was uploaded mainly to reserve the CPAN namespace -
40	stay tuned! The basic API should be finished, however.	64	stay tuned!
41		65
42	=head1 CONCEPTS	66	=head1 CONCEPTS
43		67
44	=over 4	68	=over 4
45		69
…		…
90		114
91	=cut	115	=cut
92		116
93	package AnyEvent::MP;	117	package AnyEvent::MP;
94		118
95	use AnyEvent::MP::Base;	119	use AnyEvent::MP::Kernel;
96		120
97	use common::sense;	121	use common::sense;
98		122
99	use Carp ();	123	use Carp ();
100		124
101	use AE ();	125	use AE ();
102		126
103	use base "Exporter";	127	use base "Exporter";
104		128
105	our $VERSION = '0.1';	129	our $VERSION = $AnyEvent::MP::Kernel::VERSION;
		130
106	our @EXPORT = qw(	131	our @EXPORT = qw(
107	NODE $NODE *SELF node_of _any_	132	NODE $NODE *SELF node_of _any_
108	resolve_node initialise_node	133	resolve_node initialise_node
109	snd rcv mon kil reg psub	134	snd rcv mon kil reg psub spawn
110	port	135	port
111	);	136	);
112		137
113	our $SELF;	138	our $SELF;
114		139
…		…
297	$port	322	$port
298	}	323	}
299		324
300	=item reg $port, $name	325	=item reg $port, $name
301		326
302	Registers the given port under the name C<$name>. If the name already	327	=item reg $name
303	exists it is replaced.	328
		329	Registers the given port (or C<$SELF><<< if missing) under the name
		330	C<$name>. If the name already exists it is replaced.
304		331
305	A port can only be registered under one well known name.	332	A port can only be registered under one well known name.
306		333
307	A port automatically becomes unregistered when it is killed.	334	A port automatically becomes unregistered when it is killed.
308		335
309	=cut	336	=cut
310		337
311	sub reg(@) {	338	sub reg(@) {
312	my ($port, $name) = @_;	339	my $port = @_ > 1 ? shift : $SELF \|\| Carp::croak 'reg: called with one argument only, but $SELF not set,';
313		340
314	$REG{$name} = $port;	341	$REG{$_[0]} = $port;
315	}	342	}
316		343
317	=item rcv $port, $callback->(@msg)	344	=item rcv $port, $callback->(@msg)
318		345
319	Replaces the callback on the specified miniport (after converting it to	346	Replaces the callback on the specified miniport (after converting it to
…		…
324	=item rcv $port, $smartmatch => $callback->(@msg), ...	351	=item rcv $port, $smartmatch => $callback->(@msg), ...
325		352
326	=item rcv $port, [$smartmatch...] => $callback->(@msg), ...	353	=item rcv $port, [$smartmatch...] => $callback->(@msg), ...
327		354
328	Register callbacks to be called on matching messages on the given full	355	Register callbacks to be called on matching messages on the given full
329	port (after converting it to one if required).	356	port (after converting it to one if required) and return the port.
330		357
331	The callback has to return a true value when its work is done, after	358	The callback has to return a true value when its work is done, after
332	which is will be removed, or a false value in which case it will stay	359	which is will be removed, or a false value in which case it will stay
333	registered.	360	registered.
334		361
335	The global C<$SELF> (exported by this module) contains C<$port> while	362	The global C<$SELF> (exported by this module) contains C<$port> while
336	executing the callback.	363	executing the callback.
337		364
338	Runtime errors wdurign callback execution will result in the port being	365	Runtime errors during callback execution will result in the port being
339	C<kil>ed.	366	C<kil>ed.
340		367
341	If the match is an array reference, then it will be matched against the	368	If the match is an array reference, then it will be matched against the
342	first elements of the message, otherwise only the first element is being	369	first elements of the message, otherwise only the first element is being
343	matched.	370	matched.
…		…
346	exported by this module) matches any single element of the message.	373	exported by this module) matches any single element of the message.
347		374
348	While not required, it is highly recommended that the first matching	375	While not required, it is highly recommended that the first matching
349	element is a string identifying the message. The one-string-only match is	376	element is a string identifying the message. The one-string-only match is
350	also the most efficient match (by far).	377	also the most efficient match (by far).
		378
		379	Example: create a port and bind receivers on it in one go.
		380
		381	my $port = rcv port,
		382	msg1 => sub { ...; 0 },
		383	msg2 => sub { ...; 0 },
		384	;
		385
		386	Example: create a port, bind receivers and send it in a message elsewhere
		387	in one go:
		388
		389	snd $otherport, reply =>
		390	rcv port,
		391	msg1 => sub { ...; 0 },
		392	...
		393	;
351		394
352	=cut	395	=cut
353		396
354	sub rcv($@) {	397	sub rcv($@) {
355	my $port = shift;	398	my $port = shift;
…		…
462	}	505	}
463	}	506	}
464		507
465	=item $guard = mon $port, $cb->(@reason)	508	=item $guard = mon $port, $cb->(@reason)
466		509
467	=item $guard = mon $port, $otherport	510	=item $guard = mon $port, $rcvport
468		511
		512	=item $guard = mon $port
		513
469	=item $guard = mon $port, $otherport, @msg	514	=item $guard = mon $port, $rcvport, @msg
470		515
471	Monitor the given port and do something when the port is killed.	516	Monitor the given port and do something when the port is killed or
		517	messages to it were lost, and optionally return a guard that can be used
		518	to stop monitoring again.
472		519
		520	C<mon> effectively guarantees that, in the absence of hardware failures,
		521	that after starting the monitor, either all messages sent to the port
		522	will arrive, or the monitoring action will be invoked after possible
		523	message loss has been detected. No messages will be lost "in between"
		524	(after the first lost message no further messages will be received by the
		525	port). After the monitoring action was invoked, further messages might get
		526	delivered again.
		527
473	In the first form, the callback is simply called with any number	528	In the first form (callback), the callback is simply called with any
474	of C<@reason> elements (no @reason means that the port was deleted	529	number of C<@reason> elements (no @reason means that the port was deleted
475	"normally"). Note also that I<< the callback B<must> never die >>, so use	530	"normally"). Note also that I<< the callback B<must> never die >>, so use
476	C<eval> if unsure.	531	C<eval> if unsure.
477		532
478	In the second form, the other port will be C<kil>'ed with C<@reason>, iff	533	In the second form (another port given), the other port (C<$rcvport>)
479	a @reason was specified, i.e. on "normal" kils nothing happens, while	534	will be C<kil>'ed with C<@reason>, iff a @reason was specified, i.e. on
480	under all other conditions, the other port is killed with the same reason.	535	"normal" kils nothing happens, while under all other conditions, the other
		536	port is killed with the same reason.
481		537
		538	The third form (kill self) is the same as the second form, except that
		539	C<$rvport> defaults to C<$SELF>.
		540
482	In the last form, a message of the form C<@msg, @reason> will be C<snd>.	541	In the last form (message), a message of the form C<@msg, @reason> will be
		542	C<snd>.
		543
		544	As a rule of thumb, monitoring requests should always monitor a port from
		545	a local port (or callback). The reason is that kill messages might get
		546	lost, just like any other message. Another less obvious reason is that
		547	even monitoring requests can get lost (for exmaple, when the connection
		548	to the other node goes down permanently). When monitoring a port locally
		549	these problems do not exist.
483		550
484	Example: call a given callback when C<$port> is killed.	551	Example: call a given callback when C<$port> is killed.
485		552
486	mon $port, sub { warn "port died because of <@_>\n" };	553	mon $port, sub { warn "port died because of <@_>\n" };
487		554
488	Example: kill ourselves when C<$port> is killed abnormally.	555	Example: kill ourselves when C<$port> is killed abnormally.
489		556
490	mon $port, $self;	557	mon $port;
491		558
492	Example: send us a restart message another C<$port> is killed.	559	Example: send us a restart message when another C<$port> is killed.
493		560
494	mon $port, $self => "restart";	561	mon $port, $self => "restart";
495		562
496	=cut	563	=cut
497		564
498	sub mon {	565	sub mon {
499	my ($noderef, $port) = split /#/, shift, 2;	566	my ($noderef, $port) = split /#/, shift, 2;
500		567
501	my $node = $NODE{$noderef} \|\| add_node $noderef;	568	my $node = $NODE{$noderef} \|\| add_node $noderef;
502		569
503	my $cb = shift;	570	my $cb = @_ ? shift : $SELF \|\| Carp::croak 'mon: called with one argument only, but $SELF not set,';
504		571
505	unless (ref $cb) {	572	unless (ref $cb) {
506	if (@_) {	573	if (@_) {
507	# send a kill info message	574	# send a kill info message
508	my (@msg) = ($cb, @_);	575	my (@msg) = ($cb, @_);
…		…
539	=cut	606	=cut
540		607
541	sub mon_guard {	608	sub mon_guard {
542	my ($port, @refs) = @_;	609	my ($port, @refs) = @_;
543		610
		611	#TODO: mon-less form?
		612
544	mon $port, sub { 0 && @refs }	613	mon $port, sub { 0 && @refs }
545	}	614	}
546		615
547	=item lnk $port1, $port2
548
549	Link two ports. This is simply a shorthand for:
550
551	mon $port1, $port2;
552	mon $port2, $port1;
553
554	It means that if either one is killed abnormally, the other one gets
555	killed as well.
556
557	=item kil $port[, @reason]	616	=item kil $port[, @reason]
558		617
559	Kill the specified port with the given C<@reason>.	618	Kill the specified port with the given C<@reason>.
560		619
561	If no C<@reason> is specified, then the port is killed "normally" (linked	620	If no C<@reason> is specified, then the port is killed "normally" (linked
…		…
567	Runtime errors while evaluating C<rcv> callbacks or inside C<psub> blocks	626	Runtime errors while evaluating C<rcv> callbacks or inside C<psub> blocks
568	will be reported as reason C<< die => $@ >>.	627	will be reported as reason C<< die => $@ >>.
569		628
570	Transport/communication errors are reported as C<< transport_error =>	629	Transport/communication errors are reported as C<< transport_error =>
571	$message >>.	630	$message >>.
		631
		632	=cut
		633
		634	=item $port = spawn $node, $initfunc[, @initdata]
		635
		636	Creates a port on the node C<$node> (which can also be a port ID, in which
		637	case it's the node where that port resides).
		638
		639	The port ID of the newly created port is return immediately, and it is
		640	permissible to immediately start sending messages or monitor the port.
		641
		642	After the port has been created, the init function is
		643	called. This function must be a fully-qualified function name
		644	(e.g. C<MyApp::Chat::Server::init>). To specify a function in the main
		645	program, use C<::name>.
		646
		647	If the function doesn't exist, then the node tries to C<require>
		648	the package, then the package above the package and so on (e.g.
		649	C<MyApp::Chat::Server>, C<MyApp::Chat>, C<MyApp>) until the function
		650	exists or it runs out of package names.
		651
		652	The init function is then called with the newly-created port as context
		653	object (C<$SELF>) and the C<@initdata> values as arguments.
		654
		655	A common idiom is to pass your own port, monitor the spawned port, and
		656	in the init function, monitor the original port. This two-way monitoring
		657	ensures that both ports get cleaned up when there is a problem.
		658
		659	Example: spawn a chat server port on C<$othernode>.
		660
		661	# this node, executed from within a port context:
		662	my $server = spawn $othernode, "MyApp::Chat::Server::connect", $SELF;
		663	mon $server;
		664
		665	# init function on C<$othernode>
		666	sub connect {
		667	my ($srcport) = @_;
		668
		669	mon $srcport;
		670
		671	rcv $SELF, sub {
		672	...
		673	};
		674	}
		675
		676	=cut
		677
		678	sub _spawn {
		679	my $port = shift;
		680	my $init = shift;
		681
		682	local $SELF = "$NODE#$port";
		683	eval {
		684	&{ load_func $init }
		685	};
		686	_self_die if $@;
		687	}
		688
		689	sub spawn(@) {
		690	my ($noderef, undef) = split /#/, shift, 2;
		691
		692	my $id = "$RUNIQ." . $ID++;
		693
		694	$_[0] =~ /::/
		695	or Carp::croak "spawn init function must be a fully-qualified name, caught";
		696
		697	($NODE{$noderef} \|\| add_node $noderef)
		698	->send (["", "AnyEvent::MP::_spawn" => $id, @_]);
		699
		700	"$noderef#$id"
		701	}
572		702
573	=back	703	=back
574		704
575	=head1 NODE MESSAGES	705	=head1 NODE MESSAGES
576		706
…		…
618		748
619	=back	749	=back
620		750
621	=head1 AnyEvent::MP vs. Distributed Erlang	751	=head1 AnyEvent::MP vs. Distributed Erlang
622		752
623	AnyEvent::MP got lots of its ideas from distributed erlang (erlang node	753	AnyEvent::MP got lots of its ideas from distributed Erlang (Erlang node
624	== aemp node, erlang process == aemp port), so many of the documents and	754	== aemp node, Erlang process == aemp port), so many of the documents and
625	programming techniques employed by erlang apply to AnyEvent::MP. Here is a	755	programming techniques employed by Erlang apply to AnyEvent::MP. Here is a
626	sample:	756	sample:
627		757
628	http://www.erlang.se/doc/programming_rules.shtml	758	http://www.Erlang.se/doc/programming_rules.shtml
629	http://erlang.org/doc/getting_started/part_frame.html # chapters 3 and 4	759	http://Erlang.org/doc/getting_started/part_frame.html # chapters 3 and 4
630	http://erlang.org/download/erlang-book-part1.pdf # chapters 5 and 6	760	http://Erlang.org/download/Erlang-book-part1.pdf # chapters 5 and 6
631	http://erlang.org/download/armstrong_thesis_2003.pdf # chapters 4 and 5	761	http://Erlang.org/download/armstrong_thesis_2003.pdf # chapters 4 and 5
632		762
633	Despite the similarities, there are also some important differences:	763	Despite the similarities, there are also some important differences:
634		764
635	=over 4	765	=over 4
636		766
…		…
647		777
648	Erlang uses processes that selctively receive messages, and therefore	778	Erlang uses processes that selctively receive messages, and therefore
649	needs a queue. AEMP is event based, queuing messages would serve no useful	779	needs a queue. AEMP is event based, queuing messages would serve no useful
650	purpose.	780	purpose.
651		781
652	(But see L<Coro::MP> for a more erlang-like process model on top of AEMP).	782	(But see L<Coro::MP> for a more Erlang-like process model on top of AEMP).
653		783
654	=item * Erlang sends are synchronous, AEMP sends are asynchronous.	784	=item * Erlang sends are synchronous, AEMP sends are asynchronous.
655		785
656	Sending messages in erlang is synchronous and blocks the process. AEMP	786	Sending messages in Erlang is synchronous and blocks the process. AEMP
657	sends are immediate, connection establishment is handled in the	787	sends are immediate, connection establishment is handled in the
658	background.	788	background.
659		789
660	=item * Erlang can silently lose messages, AEMP cannot.	790	=item * Erlang can silently lose messages, AEMP cannot.
661		791
…		…
664	and c, and the other side only receives messages a and c).	794	and c, and the other side only receives messages a and c).
665		795
666	AEMP guarantees correct ordering, and the guarantee that there are no	796	AEMP guarantees correct ordering, and the guarantee that there are no
667	holes in the message sequence.	797	holes in the message sequence.
668		798
669	=item * In erlang, processes can be declared dead and later be found to be	799	=item * In Erlang, processes can be declared dead and later be found to be
670	alive.	800	alive.
671		801
672	In erlang it can happen that a monitored process is declared dead and	802	In Erlang it can happen that a monitored process is declared dead and
673	linked processes get killed, but later it turns out that the process is	803	linked processes get killed, but later it turns out that the process is
674	still alive - and can receive messages.	804	still alive - and can receive messages.
675		805
676	In AEMP, when port monitoring detects a port as dead, then that port will	806	In AEMP, when port monitoring detects a port as dead, then that port will
677	eventually be killed - it cannot happen that a node detects a port as dead	807	eventually be killed - it cannot happen that a node detects a port as dead
678	and then later sends messages to it, finding it is still alive.	808	and then later sends messages to it, finding it is still alive.
679		809
680	=item * Erlang can send messages to the wrong port, AEMP does not.	810	=item * Erlang can send messages to the wrong port, AEMP does not.
681		811
682	In erlang it is quite possible that a node that restarts reuses a process	812	In Erlang it is quite possible that a node that restarts reuses a process
683	ID known to other nodes for a completely different process, causing	813	ID known to other nodes for a completely different process, causing
684	messages destined for that process to end up in an unrelated process.	814	messages destined for that process to end up in an unrelated process.
685		815
686	AEMP never reuses port IDs, so old messages or old port IDs floating	816	AEMP never reuses port IDs, so old messages or old port IDs floating
687	around in the network will not be sent to an unrelated port.	817	around in the network will not be sent to an unrelated port.
…		…
693	securely authenticate nodes.	823	securely authenticate nodes.
694		824
695	=item * The AEMP protocol is optimised for both text-based and binary	825	=item * The AEMP protocol is optimised for both text-based and binary
696	communications.	826	communications.
697		827
698	The AEMP protocol, unlike the erlang protocol, supports both	828	The AEMP protocol, unlike the Erlang protocol, supports both
699	language-independent text-only protocols (good for debugging) and binary,	829	language-independent text-only protocols (good for debugging) and binary,
700	language-specific serialisers (e.g. Storable).	830	language-specific serialisers (e.g. Storable).
701		831
702	It has also been carefully designed to be implementable in other languages	832	It has also been carefully designed to be implementable in other languages
703	with a minimum of work while gracefully degrading fucntionality to make the	833	with a minimum of work while gracefully degrading fucntionality to make the
704	protocol simple.	834	protocol simple.
705		835
		836	=item * AEMP has more flexible monitoring options than Erlang.
		837
		838	In Erlang, you can chose to receive I<all> exit signals as messages
		839	or I<none>, there is no in-between, so monitoring single processes is
		840	difficult to implement. Monitoring in AEMP is more flexible than in
		841	Erlang, as one can choose between automatic kill, exit message or callback
		842	on a per-process basis.
		843
		844	=item * Erlang tries to hide remote/local connections, AEMP does not.
		845
		846	Monitoring in Erlang is not an indicator of process death/crashes,
		847	as linking is (except linking is unreliable in Erlang).
		848
		849	In AEMP, you don't "look up" registered port names or send to named ports
		850	that might or might not be persistent. Instead, you normally spawn a port
		851	on the remote node. The init function monitors the you, and you monitor
		852	the remote port. Since both monitors are local to the node, they are much
		853	more reliable.
		854
		855	This also saves round-trips and avoids sending messages to the wrong port
		856	(hard to do in Erlang).
		857
		858	=back
		859
		860	=head1 RATIONALE
		861
		862	=over 4
		863
		864	=item Why strings for ports and noderefs, why not objects?
		865
		866	We considered "objects", but found that the actual number of methods
		867	thatc an be called are very low. Since port IDs and noderefs travel over
		868	the network frequently, the serialising/deserialising would add lots of
		869	overhead, as well as having to keep a proxy object.
		870
		871	Strings can easily be printed, easily serialised etc. and need no special
		872	procedures to be "valid".
		873
		874	And a a miniport consists of a single closure stored in a global hash - it
		875	can't become much cheaper.
		876
		877	=item Why favour JSON, why not real serialising format such as Storable?
		878
		879	In fact, any AnyEvent::MP node will happily accept Storable as framing
		880	format, but currently there is no way to make a node use Storable by
		881	default.
		882
		883	The default framing protocol is JSON because a) JSON::XS is many times
		884	faster for small messages and b) most importantly, after years of
		885	experience we found that object serialisation is causing more problems
		886	than it gains: Just like function calls, objects simply do not travel
		887	easily over the network, mostly because they will always be a copy, so you
		888	always have to re-think your design.
		889
		890	Keeping your messages simple, concentrating on data structures rather than
		891	objects, will keep your messages clean, tidy and efficient.
		892
706	=back	893	=back
707		894
708	=head1 SEE ALSO	895	=head1 SEE ALSO
709		896
710	L<AnyEvent>.	897	L<AnyEvent>.

Diff Legend

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

Comparing AnyEvent-MP/MP.pm (file contents): Revision 1.34 by root, Wed Aug 5 23:50:46 2009 UTC vs. Revision 1.47 by root, Thu Aug 13 01:57:10 2009 UTC

Diff Legend

Comparing AnyEvent-MP/MP.pm (file contents):
Revision 1.34 by root, Wed Aug 5 23:50:46 2009 UTC vs.
Revision 1.47 by root, Thu Aug 13 01:57:10 2009 UTC