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

Comparing AnyEvent-MP/MP.pm (file contents):
Revision 1.22 by root, Tue Aug 4 18:33:30 2009 UTC vs.
Revision 1.32 by root, Wed Aug 5 19:58:46 2009 UTC

…		…
30	This module (-family) implements a simple message passing framework.	30	This module (-family) implements a simple message passing framework.
31		31
32	Despite its simplicity, you can securely message other processes running	32	Despite its simplicity, you can securely message other processes running
33	on the same or other hosts.	33	on the same or other hosts.
34		34
		35	For an introduction to this module family, see the L<AnyEvent::MP::Intro>
		36	manual page.
		37
35	At the moment, this module family is severly brokena nd underdocumented,	38	At the moment, this module family is severly broken and underdocumented,
36	so do not use. This was uploaded mainly to reserve the CPAN namespace -	39	so do not use. This was uploaded mainly to reserve the CPAN namespace -
37	stay tuned!	40	stay tuned! The basic API should be finished, however.
38		41
39	=head1 CONCEPTS	42	=head1 CONCEPTS
40		43
41	=over 4	44	=over 4
42		45
43	=item port	46	=item port
44		47
45	A port is something you can send messages to with the C<snd> function, and	48	A port is something you can send messages to (with the C<snd> function).
46	you can register C<rcv> handlers with. All C<rcv> handlers will receive	49
47	messages they match, messages will not be queued.	50	Some ports allow you to register C<rcv> handlers that can match specific
		51	messages. All C<rcv> handlers will receive messages they match, messages
		52	will not be queued.
48		53
49	=item port id - C<noderef#portname>	54	=item port id - C<noderef#portname>
50		55
51	A port id is always the noderef, a hash-mark (C<#>) as separator, followed	56	A port id is normaly the concatenation of a noderef, a hash-mark (C<#>) as
52	by a port name (a printable string of unspecified format).	57	separator, and a port name (a printable string of unspecified format). An
		58	exception is the the node port, whose ID is identical to its node
		59	reference.
53		60
54	=item node	61	=item node
55		62
56	A node is a single process containing at least one port - the node	63	A node is a single process containing at least one port - the node
57	port. You can send messages to node ports to let them create new ports,	64	port. You can send messages to node ports to find existing ports or to
58	among other things.	65	create new ports, among other things.
59		66
60	Initially, nodes are either private (single-process only) or hidden	67	Nodes are either private (single-process only), slaves (connected to a
61	(connected to a master node only). Only when they epxlicitly "become	68	master node only) or public nodes (connectable from unrelated nodes).
62	public" can you send them messages from unrelated other nodes.
63		69
64	=item noderef - C<host:port,host:port...>, C<id@noderef>, C<id>	70	=item noderef - C<host:port,host:port...>, C<id@noderef>, C<id>
65		71
66	A noderef is a string that either uniquely identifies a given node (for	72	A node reference is a string that either simply identifies the node (for
67	private and hidden nodes), or contains a recipe on how to reach a given	73	private and slave nodes), or contains a recipe on how to reach a given
68	node (for public nodes).	74	node (for public nodes).
69		75
		76	This recipe is simply a comma-separated list of C<address:port> pairs (for
		77	TCP/IP, other protocols might look different).
		78
		79	Node references come in two flavours: resolved (containing only numerical
		80	addresses) or unresolved (where hostnames are used instead of addresses).
		81
		82	Before using an unresolved node reference in a message you first have to
		83	resolve it.
		84
70	=back	85	=back
71		86
72	=head1 VARIABLES/FUNCTIONS	87	=head1 VARIABLES/FUNCTIONS
73		88
74	=over 4	89	=over 4
…		…
85		100
86	use AE ();	101	use AE ();
87		102
88	use base "Exporter";	103	use base "Exporter";
89		104
90	our $VERSION = '0.02';	105	our $VERSION = '0.1';
91	our @EXPORT = qw(	106	our @EXPORT = qw(
92	NODE $NODE *SELF node_of _any_	107	NODE $NODE *SELF node_of _any_
93	become_slave become_public	108	resolve_node initialise_node
94	snd rcv mon kil reg psub	109	snd rcv mon kil reg psub
95	port	110	port
96	);	111	);
97		112
98	our $SELF;	113	our $SELF;
…		…
112		127
113	=item $noderef = node_of $portid	128	=item $noderef = node_of $portid
114		129
115	Extracts and returns the noderef from a portid or a noderef.	130	Extracts and returns the noderef from a portid or a noderef.
116		131
		132	=item $cv = resolve_node $noderef
		133
		134	Takes an unresolved node reference that may contain hostnames and
		135	abbreviated IDs, resolves all of them and returns a resolved node
		136	reference.
		137
		138	In addition to C<address:port> pairs allowed in resolved noderefs, the
		139	following forms are supported:
		140
		141	=over 4
		142
		143	=item the empty string
		144
		145	An empty-string component gets resolved as if the default port (4040) was
		146	specified.
		147
		148	=item naked port numbers (e.g. C<1234>)
		149
		150	These are resolved by prepending the local nodename and a colon, to be
		151	further resolved.
		152
		153	=item hostnames (e.g. C<localhost:1234>, C<localhost>)
		154
		155	These are resolved by using AnyEvent::DNS to resolve them, optionally
		156	looking up SRV records for the C<aemp=4040> port, if no port was
		157	specified.
		158
		159	=back
		160
117	=item $SELF	161	=item $SELF
118		162
119	Contains the current port id while executing C<rcv> callbacks or C<psub>	163	Contains the current port id while executing C<rcv> callbacks or C<psub>
120	blocks.	164	blocks.
121		165
…		…
145	JSON is used, then only strings, numbers and arrays and hashes consisting	189	JSON is used, then only strings, numbers and arrays and hashes consisting
146	of those are allowed (no objects). When Storable is used, then anything	190	of those are allowed (no objects). When Storable is used, then anything
147	that Storable can serialise and deserialise is allowed, and for the local	191	that Storable can serialise and deserialise is allowed, and for the local
148	node, anything can be passed.	192	node, anything can be passed.
149		193
150	=item kil $portid[, @reason]
151
152	Kill the specified port with the given C<@reason>.
153
154	If no C<@reason> is specified, then the port is killed "normally" (linked
155	ports will not be kileld, or even notified).
156
157	Otherwise, linked ports get killed with the same reason (second form of
158	C<mon>, see below).
159
160	Runtime errors while evaluating C<rcv> callbacks or inside C<psub> blocks
161	will be reported as reason C<< die => $@ >>.
162
163	Transport/communication errors are reported as C<< transport_error =>
164	$message >>.
165
166	=item $guard = mon $portid, $cb->(@reason)
167
168	=item $guard = mon $portid, $otherport
169
170	=item $guard = mon $portid, $otherport, @msg
171
172	Monitor the given port and do something when the port is killed.
173
174	In the first form, the callback is simply called with any number
175	of C<@reason> elements (no @reason means that the port was deleted
176	"normally"). Note also that I<< the callback B<must> never die >>, so use
177	C<eval> if unsure.
178
179	In the second form, the other port will be C<kil>'ed with C<@reason>, iff
180	a @reason was specified, i.e. on "normal" kils nothing happens, while
181	under all other conditions, the other port is killed with the same reason.
182
183	In the last form, a message of the form C<@msg, @reason> will be C<snd>.
184
185	Example: call a given callback when C<$port> is killed.
186
187	mon $port, sub { warn "port died because of <@_>\n" };
188
189	Example: kill ourselves when C<$port> is killed abnormally.
190
191	mon $port, $self;
192
193	Example: send us a restart message another C<$port> is killed.
194
195	mon $port, $self => "restart";
196
197	=cut
198
199	sub mon {
200	my ($noderef, $port, $cb) = ((split /#/, shift, 2), shift);
201
202	my $node = $NODE{$noderef} \|\| add_node $noderef;
203
204	#TODO: ports must not be references
205	if (!ref $cb or "AnyEvent::MP::Port" eq ref $cb) {
206	if (@_) {
207	# send a kill info message
208	my (@msg) = ($cb, @_);
209	$cb = sub { snd @msg, @_ };
210	} else {
211	# simply kill other port
212	my $port = $cb;
213	$cb = sub { kil $port, @_ if @_ };
214	}
215	}
216
217	$node->monitor ($port, $cb);
218
219	defined wantarray
220	and AnyEvent::Util::guard { $node->unmonitor ($port, $cb) }
221	}
222
223	=item $guard = mon_guard $port, $ref, $ref...
224
225	Monitors the given C<$port> and keeps the passed references. When the port
226	is killed, the references will be freed.
227
228	Optionally returns a guard that will stop the monitoring.
229
230	This function is useful when you create e.g. timers or other watchers and
231	want to free them when the port gets killed:
232
233	$port->rcv (start => sub {
234	my $timer; $timer = mon_guard $port, AE::timer 1, 1, sub {
235	undef $timer if 0.9 < rand;
236	});
237	});
238
239	=cut
240
241	sub mon_guard {
242	my ($port, @refs) = @_;
243
244	mon $port, sub { 0 && @refs }
245	}
246
247	=item $local_port = port	194	=item $local_port = port
248		195
249	Create a new local port object that supports message matching.	196	Create a new local port object that can be used either as a pattern
		197	matching port ("full port") or a single-callback port ("miniport"),
		198	depending on how C<rcv> callbacks are bound to the object.
250		199
251	=item $portid = port { my @msg = @_; $finished }	200	=item $portid = port { my @msg = @_; $finished }
252		201
253	Creates a "mini port", that is, a very lightweight port without any	202	Creates a "mini port", that is, a very lightweight port without any
254	pattern matching behind it, and returns its ID.	203	pattern matching behind it, and returns its ID.
…		…
260	The message will be passed as-is, no extra argument (i.e. no port id) will	209	The message will be passed as-is, no extra argument (i.e. no port id) will
261	be passed to the callback.	210	be passed to the callback.
262		211
263	If you need the local port id in the callback, this works nicely:	212	If you need the local port id in the callback, this works nicely:
264		213
265	my $port; $port = miniport {	214	my $port; $port = port {
266	snd $otherport, reply => $port;	215	snd $otherport, reply => $port;
267	};	216	};
268		217
269	=cut	218	=cut
270		219
…		…
331	my ($portid, $name) = @_;	280	my ($portid, $name) = @_;
332		281
333	$REG{$name} = $portid;	282	$REG{$name} = $portid;
334	}	283	}
335		284
		285	=item rcv $portid, $callback->(@msg)
		286
		287	Replaces the callback on the specified miniport (or newly created port
		288	object, see C<port>). Full ports are configured with the following calls:
		289
336	=item rcv $portid, tagstring => $callback->(@msg), ...	290	=item rcv $portid, tagstring => $callback->(@msg), ...
337		291
338	=item rcv $portid, $smartmatch => $callback->(@msg), ...	292	=item rcv $portid, $smartmatch => $callback->(@msg), ...
339		293
340	=item rcv $portid, [$smartmatch...] => $callback->(@msg), ...	294	=item rcv $portid, [$smartmatch...] => $callback->(@msg), ...
341		295
342	Register callbacks to be called on matching messages on the given port.	296	Register callbacks to be called on matching messages on the given full
		297	port (or newly created port).
343		298
344	The callback has to return a true value when its work is done, after	299	The callback has to return a true value when its work is done, after
345	which is will be removed, or a false value in which case it will stay	300	which is will be removed, or a false value in which case it will stay
346	registered.	301	registered.
347		302
…		…
363	also the most efficient match (by far).	318	also the most efficient match (by far).
364		319
365	=cut	320	=cut
366		321
367	sub rcv($@) {	322	sub rcv($@) {
		323	my $portid = shift;
368	my ($noderef, $port) = split /#/, shift, 2;	324	my ($noderef, $port) = split /#/, $port, 2;
369		325
370	($NODE{$noderef} \|\| add_node $noderef) == $NODE{""}	326	($NODE{$noderef} \|\| add_node $noderef) == $NODE{""}
371	or Carp::croak "$noderef#$port: rcv can only be called on local ports, caught";	327	or Carp::croak "$noderef#$port: rcv can only be called on local ports, caught";
372		328
373	my $self = $PORT_DATA{$port}	329	my $self = $PORT_DATA{$port}
…		…
388	: push @{ $self->{rc0}{$match->[0]} }, [$cb];	344	: push @{ $self->{rc0}{$match->[0]} }, [$cb];
389	} else {	345	} else {
390	push @{ $self->{any} }, [$cb, $match];	346	push @{ $self->{any} }, [$cb, $match];
391	}	347	}
392	}	348	}
		349
		350	$portid
393	}	351	}
394		352
395	=item $closure = psub { BLOCK }	353	=item $closure = psub { BLOCK }
396		354
397	Remembers C<$SELF> and creates a closure out of the BLOCK. When the	355	Remembers C<$SELF> and creates a closure out of the BLOCK. When the
…		…
428	$res	386	$res
429	}	387	}
430	}	388	}
431	}	389	}
432		390
		391	=item $guard = mon $portid, $cb->(@reason)
		392
		393	=item $guard = mon $portid, $otherport
		394
		395	=item $guard = mon $portid, $otherport, @msg
		396
		397	Monitor the given port and do something when the port is killed.
		398
		399	In the first form, the callback is simply called with any number
		400	of C<@reason> elements (no @reason means that the port was deleted
		401	"normally"). Note also that I<< the callback B<must> never die >>, so use
		402	C<eval> if unsure.
		403
		404	In the second form, the other port will be C<kil>'ed with C<@reason>, iff
		405	a @reason was specified, i.e. on "normal" kils nothing happens, while
		406	under all other conditions, the other port is killed with the same reason.
		407
		408	In the last form, a message of the form C<@msg, @reason> will be C<snd>.
		409
		410	Example: call a given callback when C<$port> is killed.
		411
		412	mon $port, sub { warn "port died because of <@_>\n" };
		413
		414	Example: kill ourselves when C<$port> is killed abnormally.
		415
		416	mon $port, $self;
		417
		418	Example: send us a restart message another C<$port> is killed.
		419
		420	mon $port, $self => "restart";
		421
		422	=cut
		423
		424	sub mon {
		425	my ($noderef, $port) = split /#/, shift, 2;
		426
		427	my $node = $NODE{$noderef} \|\| add_node $noderef;
		428
		429	my $cb = shift;
		430
		431	unless (ref $cb) {
		432	if (@_) {
		433	# send a kill info message
		434	my (@msg) = ($cb, @_);
		435	$cb = sub { snd @msg, @_ };
		436	} else {
		437	# simply kill other port
		438	my $port = $cb;
		439	$cb = sub { kil $port, @_ if @_ };
		440	}
		441	}
		442
		443	$node->monitor ($port, $cb);
		444
		445	defined wantarray
		446	and AnyEvent::Util::guard { $node->unmonitor ($port, $cb) }
		447	}
		448
		449	=item $guard = mon_guard $port, $ref, $ref...
		450
		451	Monitors the given C<$port> and keeps the passed references. When the port
		452	is killed, the references will be freed.
		453
		454	Optionally returns a guard that will stop the monitoring.
		455
		456	This function is useful when you create e.g. timers or other watchers and
		457	want to free them when the port gets killed:
		458
		459	$port->rcv (start => sub {
		460	my $timer; $timer = mon_guard $port, AE::timer 1, 1, sub {
		461	undef $timer if 0.9 < rand;
		462	});
		463	});
		464
		465	=cut
		466
		467	sub mon_guard {
		468	my ($port, @refs) = @_;
		469
		470	mon $port, sub { 0 && @refs }
		471	}
		472
		473	=item lnk $port1, $port2
		474
		475	Link two ports. This is simply a shorthand for:
		476
		477	mon $port1, $port2;
		478	mon $port2, $port1;
		479
		480	It means that if either one is killed abnormally, the other one gets
		481	killed as well.
		482
		483	=item kil $portid[, @reason]
		484
		485	Kill the specified port with the given C<@reason>.
		486
		487	If no C<@reason> is specified, then the port is killed "normally" (linked
		488	ports will not be kileld, or even notified).
		489
		490	Otherwise, linked ports get killed with the same reason (second form of
		491	C<mon>, see below).
		492
		493	Runtime errors while evaluating C<rcv> callbacks or inside C<psub> blocks
		494	will be reported as reason C<< die => $@ >>.
		495
		496	Transport/communication errors are reported as C<< transport_error =>
		497	$message >>.
		498
433	=back	499	=back
434		500
435	=head1 FUNCTIONS FOR NODES	501	=head1 FUNCTIONS FOR NODES
436		502
437	=over 4	503	=over 4
438		504
439	=item become_public endpoint...	505	=item become_public $noderef
440		506
441	Tells the node to become a public node, i.e. reachable from other nodes.	507	Tells the node to become a public node, i.e. reachable from other nodes.
442		508
443	If no arguments are given, or the first argument is C<undef>, then	509	The first argument is the (unresolved) node reference of the local node
444	AnyEvent::MP tries to bind on port C<4040> on all IP addresses that the	510	(if missing then the empty string is used).
445	local nodename resolves to.
446		511
447	Otherwise the first argument must be an array-reference with transport	512	It is quite common to not specify anything, in which case the local node
448	endpoints ("ip:port", "hostname:port") or port numbers (in which case the	513	tries to listen on the default port, or to only specify a port number, in
449	local nodename is used as hostname). The endpoints are all resolved and	514	which case AnyEvent::MP tries to guess the local addresses.
450	will become the node reference.
451		515
452	=cut	516	=cut
453		517
454	=back	518	=back
455		519
…		…
458	Nodes understand the following messages sent to them. Many of them take	522	Nodes understand the following messages sent to them. Many of them take
459	arguments called C<@reply>, which will simply be used to compose a reply	523	arguments called C<@reply>, which will simply be used to compose a reply
460	message - C<$reply[0]> is the port to reply to, C<$reply[1]> the type and	524	message - C<$reply[0]> is the port to reply to, C<$reply[1]> the type and
461	the remaining arguments are simply the message data.	525	the remaining arguments are simply the message data.
462		526
		527	While other messages exist, they are not public and subject to change.
		528
463	=over 4	529	=over 4
464		530
465	=cut	531	=cut
466		532
467	=item lookup => $name, @reply	533	=item lookup => $name, @reply
…		…
495	snd $NODE, time => $myport, timereply => 1, 2;	561	snd $NODE, time => $myport, timereply => 1, 2;
496	# => snd $myport, timereply => 1, 2, <time>	562	# => snd $myport, timereply => 1, 2, <time>
497		563
498	=back	564	=back
499		565
		566	=head1 AnyEvent::MP vs. Distributed Erlang
		567
		568	AnyEvent::MP got lots of its ideas from distributed erlang (erlang node
		569	== aemp node, erlang process == aemp port), so many of the documents and
		570	programming techniques employed by erlang apply to AnyEvent::MP. Here is a
		571	sample:
		572
		573	http://www.erlang.se/doc/programming_rules.shtml
		574	http://erlang.org/doc/getting_started/part_frame.html # chapters 3 and 4
		575	http://erlang.org/download/erlang-book-part1.pdf # chapters 5 and 6
		576	http://erlang.org/download/armstrong_thesis_2003.pdf # chapters 4 and 5
		577
		578	Despite the similarities, there are also some important differences:
		579
		580	=over 4
		581
		582	=item * Node references contain the recipe on how to contact them.
		583
		584	Erlang relies on special naming and DNS to work everywhere in the
		585	same way. AEMP relies on each node knowing it's own address(es), with
		586	convenience functionality.
		587
		588	This means that AEMP requires a less tightly controlled environment at the
		589	cost of longer node references and a slightly higher management overhead.
		590
		591	=item * Erlang uses processes and a mailbox, AEMP does not queue.
		592
		593	Erlang uses processes that selctively receive messages, and therefore
		594	needs a queue. AEMP is event based, queuing messages would serve no useful
		595	purpose.
		596
		597	(But see L<Coro::MP> for a more erlang-like process model on top of AEMP).
		598
		599	=item * Erlang sends are synchronous, AEMP sends are asynchronous.
		600
		601	Sending messages in erlang is synchronous and blocks the process. AEMP
		602	sends are immediate, connection establishment is handled in the
		603	background.
		604
		605	=item * Erlang can silently lose messages, AEMP cannot.
		606
		607	Erlang makes few guarantees on messages delivery - messages can get lost
		608	without any of the processes realising it (i.e. you send messages a, b,
		609	and c, and the other side only receives messages a and c).
		610
		611	AEMP guarantees correct ordering, and the guarantee that there are no
		612	holes in the message sequence.
		613
		614	=item * In erlang, processes can be declared dead and later be found to be
		615	alive.
		616
		617	In erlang it can happen that a monitored process is declared dead and
		618	linked processes get killed, but later it turns out that the process is
		619	still alive - and can receive messages.
		620
		621	In AEMP, when port monitoring detects a port as dead, then that port will
		622	eventually be killed - it cannot happen that a node detects a port as dead
		623	and then later sends messages to it, finding it is still alive.
		624
		625	=item * Erlang can send messages to the wrong port, AEMP does not.
		626
		627	In erlang it is quite possible that a node that restarts reuses a process
		628	ID known to other nodes for a completely different process, causing
		629	messages destined for that process to end up in an unrelated process.
		630
		631	AEMP never reuses port IDs, so old messages or old port IDs floating
		632	around in the network will not be sent to an unrelated port.
		633
		634	=item * Erlang uses unprotected connections, AEMP uses secure
		635	authentication and can use TLS.
		636
		637	AEMP can use a proven protocol - SSL/TLS - to protect connections and
		638	securely authenticate nodes.
		639
		640	=item * The AEMP protocol is optimised for both text-based and binary
		641	communications.
		642
		643	The AEMP protocol, unlike the erlang protocol, supports both
		644	language-independent text-only protocols (good for debugging) and binary,
		645	language-specific serialisers (e.g. Storable).
		646
		647	It has also been carefully designed to be implementable in other languages
		648	with a minimum of work while gracefully degrading fucntionality to make the
		649	protocol simple.
		650
		651	=back
		652
500	=head1 SEE ALSO	653	=head1 SEE ALSO
501		654
502	L<AnyEvent>.	655	L<AnyEvent>.
503		656
504	=head1 AUTHOR	657	=head1 AUTHOR

Diff Legend

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

Comparing AnyEvent-MP/MP.pm (file contents): Revision 1.22 by root, Tue Aug 4 18:33:30 2009 UTC vs. Revision 1.32 by root, Wed Aug 5 19:58:46 2009 UTC

Diff Legend

Comparing AnyEvent-MP/MP.pm (file contents):
Revision 1.22 by root, Tue Aug 4 18:33:30 2009 UTC vs.
Revision 1.32 by root, Wed Aug 5 19:58:46 2009 UTC