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

Comparing AnyEvent-MP/MP.pm (file contents):
Revision 1.5 by root, Sat Aug 1 07:44:02 2009 UTC vs.
Revision 1.79 by root, Fri Sep 4 21:52:09 2009 UTC

…		…
4		4
5	=head1 SYNOPSIS	5	=head1 SYNOPSIS
6		6
7	use AnyEvent::MP;	7	use AnyEvent::MP;
8		8
9	NODE # returns this node identifier
10	$NODE # contains this node identifier	9	$NODE # contains this node's node ID
		10	NODE # returns this node's node ID
11		11
		12	$SELF # receiving/own port id in rcv callbacks
		13
		14	# initialise the node so it can send/receive messages
		15	configure;
		16
		17	# ports are message destinations
		18
		19	# sending messages
12	snd $port, type => data...;	20	snd $port, type => data...;
		21	snd $port, @msg;
		22	snd @msg_with_first_element_being_a_port;
13		23
14	rcv $port, smartmatch => $cb->($port, @msg);	24	# creating/using ports, the simple way
		25	my $simple_port = port { my @msg = @_ };
15		26
16	# examples:	27	# creating/using ports, tagged message matching
		28	my $port = port;
17	rcv $port2, ping => sub { snd $_[0], "pong"; 0 };	29	rcv $port, ping => sub { snd $_[0], "pong" };
18	rcv $port1, pong => sub { warn "pong received\n" };	30	rcv $port, pong => sub { warn "pong received\n" };
19	snd $port2, ping => $port1;
20		31
21	# more, smarter, matches (_any_ is exported by this module)	32	# create a port on another node
22	rcv $port, [child_died => $pid] => sub { ...	33	my $port = spawn $node, $initfunc, @initdata;
23	rcv $port, [_any_, _any_, 3] => sub { .. $_[2] is 3	34
		35	# monitoring
		36	mon $port, $cb->(@msg) # callback is invoked on death
		37	mon $port, $otherport # kill otherport on abnormal death
		38	mon $port, $otherport, @msg # send message on death
		39
		40	=head1 CURRENT STATUS
		41
		42	bin/aemp - stable.
		43	AnyEvent::MP - stable API, should work.
		44	AnyEvent::MP::Intro - explains most concepts.
		45	AnyEvent::MP::Kernel - mostly stable.
		46	AnyEvent::MP::Global - stable but incomplete, protocol not yet final.
		47
		48	stay tuned.
24		49
25	=head1 DESCRIPTION	50	=head1 DESCRIPTION
26		51
27	This module (-family) implements a simple message passing framework.	52	This module (-family) implements a simple message passing framework.
28		53
29	Despite its simplicity, you can securely message other processes running	54	Despite its simplicity, you can securely message other processes running
30	on the same or other hosts.	55	on the same or other hosts, and you can supervise entities remotely.
		56
		57	For an introduction to this module family, see the L<AnyEvent::MP::Intro>
		58	manual page and the examples under F<eg/>.
31		59
32	=head1 CONCEPTS	60	=head1 CONCEPTS
33		61
34	=over 4	62	=over 4
35		63
36	=item port	64	=item port
37		65
38	A port is something you can send messages to with the C<snd> function, and	66	Not to be confused with a TCP port, a "port" is something you can send
39	you can register C<rcv> handlers with. All C<rcv> handlers will receive	67	messages to (with the C<snd> function).
40	messages they match, messages will not be queued.
41		68
		69	Ports allow you to register C<rcv> handlers that can match all or just
		70	some messages. Messages send to ports will not be queued, regardless of
		71	anything was listening for them or not.
		72
42	=item port id - C<noderef#portname>	73	=item port ID - C<nodeid#portname>
43		74
44	A port id is always the noderef, a hash-mark (C<#>) as separator, followed	75	A port ID is the concatenation of a node ID, a hash-mark (C<#>) as
45	by a port name (a printable string of unspecified format).	76	separator, and a port name (a printable string of unspecified format).
46		77
47	=item node	78	=item node
48		79
49	A node is a single process containing at least one port - the node	80	A node is a single process containing at least one port - the node port,
50	port. You can send messages to node ports to let them create new ports,	81	which enables nodes to manage each other remotely, and to create new
51	among other things.	82	ports.
52		83
53	Initially, nodes are either private (single-process only) or hidden	84	Nodes are either public (have one or more listening ports) or private
54	(connected to a master node only). Only when they epxlicitly "become	85	(no listening ports). Private nodes cannot talk to other private nodes
55	public" can you send them messages from unrelated other nodes.	86	currently.
56		87
57	=item noderef - C<host:port,host:port...>, C<id@noderef>, C<id>	88	=item node ID - C<[a-za-Z0-9_\-.:]+>
58		89
59	A noderef is a string that either uniquely identifies a given node (for	90	A node ID is a string that uniquely identifies the node within a
60	private and hidden nodes), or contains a recipe on how to reach a given	91	network. Depending on the configuration used, node IDs can look like a
61	node (for public nodes).	92	hostname, a hostname and a port, or a random string. AnyEvent::MP itself
		93	doesn't interpret node IDs in any way.
		94
		95	=item binds - C<ip:port>
		96
		97	Nodes can only talk to each other by creating some kind of connection to
		98	each other. To do this, nodes should listen on one or more local transport
		99	endpoints - binds. Currently, only standard C<ip:port> specifications can
		100	be used, which specify TCP ports to listen on.
		101
		102	=item seeds - C<host:port>
		103
		104	When a node starts, it knows nothing about the network. To teach the node
		105	about the network it first has to contact some other node within the
		106	network. This node is called a seed.
		107
		108	Seeds are transport endpoint(s) of as many nodes as one wants. Those nodes
		109	are expected to be long-running, and at least one of those should always
		110	be available. When nodes run out of connections (e.g. due to a network
		111	error), they try to re-establish connections to some seednodes again to
		112	join the network.
		113
		114	Apart from being sued for seeding, seednodes are not special in any way -
		115	every public node can be a seednode.
62		116
63	=back	117	=back
64		118
65	=head1 VARIABLES/FUNCTIONS	119	=head1 VARIABLES/FUNCTIONS
66		120
…		…
68		122
69	=cut	123	=cut
70		124
71	package AnyEvent::MP;	125	package AnyEvent::MP;
72		126
73	use AnyEvent::MP::Util ();
74	use AnyEvent::MP::Node;	127	use AnyEvent::MP::Kernel;
75	use AnyEvent::MP::Transport;
76		128
77	use utf8;
78	use common::sense;	129	use common::sense;
79		130
80	use Carp ();	131	use Carp ();
81		132
82	use AE ();	133	use AE ();
83		134
84	use base "Exporter";	135	use base "Exporter";
85		136
86	our $VERSION = '0.0';	137	our $VERSION = $AnyEvent::MP::Kernel::VERSION;
87	our @EXPORT = qw(NODE $NODE $PORT snd rcv _any_);
88		138
89	our $DEFAULT_SECRET;	139	our @EXPORT = qw(
90	our $DEFAULT_PORT = "4040";	140	NODE $NODE *SELF node_of after
		141	configure
		142	snd rcv mon mon_guard kil reg psub spawn
		143	port
		144	);
91		145
92	our $CONNECT_INTERVAL = 5; # new connect every 5s, at least	146	our $SELF;
93	our $CONNECT_TIMEOUT = 30; # includes handshake
94		147
95	sub default_secret {	148	sub _self_die() {
96	unless (defined $DEFAULT_SECRET) {	149	my $msg = $@;
97	if (open my $fh, "<$ENV{HOME}/.aemp-secret") {	150	$msg =~ s/\n+$// unless ref $msg;
98	sysread $fh, $DEFAULT_SECRET, -s $fh;	151	kil $SELF, die => $msg;
99	} else {
100	$DEFAULT_SECRET = AnyEvent::MP::Util::nonce 32;
101	}
102	}
103
104	$DEFAULT_SECRET
105	}	152	}
106		153
107	=item NODE / $NODE	154	=item $thisnode = NODE / $NODE
108		155
109	The C<NODE ()> function and the C<$NODE> variable contain the noderef of	156	The C<NODE> function returns, and the C<$NODE> variable contains, the node
110	the local node. The value is initialised by a call to C<become_public> or	157	ID of the node running in the current process. This value is initialised by
111	C<become_slave>, after which all local port identifiers become invalid.	158	a call to C<configure>.
112		159
113	=cut	160	=item $nodeid = node_of $port
114		161
115	our $UNIQ = sprintf "%x.%x", $$, time; # per-process/node unique cookie	162	Extracts and returns the node ID from a port ID or a node ID.
116	our $PUBLIC = 0;
117	our $NODE;
118	our $PORT;
119		163
120	our %NODE; # node id to transport mapping, or "undef", for local node	164	=item configure $profile, key => value...
121	our %PORT; # local ports
122	our %LISTENER; # local transports
123		165
124	sub NODE() { $NODE }	166	=item configure key => value...
125		167
126	{	168	Before a node can talk to other nodes on the network (i.e. enter
127	use POSIX ();	169	"distributed mode") it has to configure itself - the minimum a node needs
128	my $nodename = (POSIX::uname)[1];	170	to know is its own name, and optionally it should know the addresses of
129	$NODE = "$$\@$nodename";	171	some other nodes in the network to discover other nodes.
130	}
131		172
132	sub _ANY_() { 1 }	173	This function configures a node - it must be called exactly once (or
133	sub _any_() { \&_ANY_ }	174	never) before calling other AnyEvent::MP functions.
134		175
135	sub add_node {	176	=over 4
136	my ($noderef) = @_;
137		177
138	return $NODE{$noderef}	178	=item step 1, gathering configuration from profiles
139	if exists $NODE{$noderef};
140		179
141	for (split /,/, $noderef) {	180	The function first looks up a profile in the aemp configuration (see the
142	return $NODE{$noderef} = $NODE{$_}	181	L<aemp> commandline utility). The profile name can be specified via the
143	if exists $NODE{$_};	182	named C<profile> parameter or can simply be the first parameter). If it is
144	}	183	missing, then the nodename (F<uname -n>) will be used as profile name.
145		184
146	# for indirect sends, use a different class	185	The profile data is then gathered as follows:
147	my $node = new AnyEvent::MP::Node::Direct $noderef;
148		186
149	$NODE{$_} = $node	187	First, all remaining key => value pairs (all of which are conveniently
150	for $noderef, split /,/, $noderef;	188	undocumented at the moment) will be interpreted as configuration
		189	data. Then they will be overwritten by any values specified in the global
		190	default configuration (see the F<aemp> utility), then the chain of
		191	profiles chosen by the profile name (and any C<parent> attributes).
151		192
152	$node	193	That means that the values specified in the profile have highest priority
153	}	194	and the values specified directly via C<configure> have lowest priority,
		195	and can only be used to specify defaults.
154		196
		197	If the profile specifies a node ID, then this will become the node ID of
		198	this process. If not, then the profile name will be used as node ID. The
		199	special node ID of C<anon/> will be replaced by a random node ID.
		200
		201	=item step 2, bind listener sockets
		202
		203	The next step is to look up the binds in the profile, followed by binding
		204	aemp protocol listeners on all binds specified (it is possible and valid
		205	to have no binds, meaning that the node cannot be contacted form the
		206	outside. This means the node cannot talk to other nodes that also have no
		207	binds, but it can still talk to all "normal" nodes).
		208
		209	If the profile does not specify a binds list, then a default of C<*> is
		210	used, meaning the node will bind on a dynamically-assigned port on every
		211	local IP address it finds.
		212
		213	=item step 3, connect to seed nodes
		214
		215	As the last step, the seeds list from the profile is passed to the
		216	L<AnyEvent::MP::Global> module, which will then use it to keep
		217	connectivity with at least one node at any point in time.
		218
		219	=back
		220
		221	Example: become a distributed node using the locla node name as profile.
		222	This should be the most common form of invocation for "daemon"-type nodes.
		223
		224	configure
		225
		226	Example: become an anonymous node. This form is often used for commandline
		227	clients.
		228
		229	configure nodeid => "anon/";
		230
		231	Example: configure a node using a profile called seed, which si suitable
		232	for a seed node as it binds on all local addresses on a fixed port (4040,
		233	customary for aemp).
		234
		235	# use the aemp commandline utility
		236	# aemp profile seed nodeid anon/ binds '*:4040'
		237
		238	# then use it
		239	configure profile => "seed";
		240
		241	# or simply use aemp from the shell again:
		242	# aemp run profile seed
		243
		244	# or provide a nicer-to-remember nodeid
		245	# aemp run profile seed nodeid "$(hostname)"
		246
		247	=item $SELF
		248
		249	Contains the current port id while executing C<rcv> callbacks or C<psub>
		250	blocks.
		251
		252	=item *SELF, SELF, %SELF, @SELF...
		253
		254	Due to some quirks in how perl exports variables, it is impossible to
		255	just export C<$SELF>, all the symbols named C<SELF> are exported by this
		256	module, but only C<$SELF> is currently used.
		257
155	=item snd $portid, type => @data	258	=item snd $port, type => @data
156		259
157	=item snd $portid, @msg	260	=item snd $port, @msg
158		261
159	Send the given message to the given port ID, which can identify either a	262	Send the given message to the given port, which can identify either a
160	local or a remote port.	263	local or a remote port, and must be a port ID.
161		264
162	While the message can be about anything, it is highly recommended to use	265	While the message can be almost anything, it is highly recommended to
163	a constant string as first element.	266	use a string as first element (a port ID, or some word that indicates a
		267	request type etc.) and to consist if only simple perl values (scalars,
		268	arrays, hashes) - if you think you need to pass an object, think again.
164		269
165	The message data effectively becomes read-only after a call to this	270	The message data logically becomes read-only after a call to this
166	function: modifying any argument is not allowed and can cause many	271	function: modifying any argument (or values referenced by them) is
167	problems.	272	forbidden, as there can be considerable time between the call to C<snd>
		273	and the time the message is actually being serialised - in fact, it might
		274	never be copied as within the same process it is simply handed to the
		275	receiving port.
168		276
169	The type of data you can transfer depends on the transport protocol: when	277	The type of data you can transfer depends on the transport protocol: when
170	JSON is used, then only strings, numbers and arrays and hashes consisting	278	JSON is used, then only strings, numbers and arrays and hashes consisting
171	of those are allowed (no objects). When Storable is used, then anything	279	of those are allowed (no objects). When Storable is used, then anything
172	that Storable can serialise and deserialise is allowed, and for the local	280	that Storable can serialise and deserialise is allowed, and for the local
173	node, anything can be passed.	281	node, anything can be passed. Best rely only on the common denominator of
		282	these.
174		283
175	=cut	284	=item $local_port = port
176		285
177	sub snd(@) {	286	Create a new local port object and returns its port ID. Initially it has
178	my ($noderef, $port) = split /#/, shift, 2;	287	no callbacks set and will throw an error when it receives messages.
179		288
180	add_node $noderef	289	=item $local_port = port { my @msg = @_ }
181	unless exists $NODE{$noderef};
182		290
183	$NODE{$noderef}->send (["$port", [@_]]);	291	Creates a new local port, and returns its ID. Semantically the same as
184	}	292	creating a port and calling C<rcv $port, $callback> on it.
185		293
		294	The block will be called for every message received on the port, with the
		295	global variable C<$SELF> set to the port ID. Runtime errors will cause the
		296	port to be C<kil>ed. The message will be passed as-is, no extra argument
		297	(i.e. no port ID) will be passed to the callback.
		298
		299	If you want to stop/destroy the port, simply C<kil> it:
		300
		301	my $port = port {
		302	my @msg = @_;
		303	...
		304	kil $SELF;
		305	};
		306
		307	=cut
		308
		309	sub rcv($@);
		310
		311	sub _kilme {
		312	die "received message on port without callback";
		313	}
		314
		315	sub port(;&) {
		316	my $id = "$UNIQ." . $ID++;
		317	my $port = "$NODE#$id";
		318
		319	rcv $port, shift \|\| \&_kilme;
		320
		321	$port
		322	}
		323
186	=item rcv $portid, type => $callback->(@msg)	324	=item rcv $local_port, $callback->(@msg)
187		325
188	=item rcv $portid, $smartmatch => $callback->(@msg)	326	Replaces the default callback on the specified port. There is no way to
		327	remove the default callback: use C<sub { }> to disable it, or better
		328	C<kil> the port when it is no longer needed.
189		329
190	=item rcv $portid, [$smartmatch...] => $callback->(@msg)	330	The global C<$SELF> (exported by this module) contains C<$port> while
		331	executing the callback. Runtime errors during callback execution will
		332	result in the port being C<kil>ed.
191		333
192	Register a callback on the port identified by C<$portid>, which I<must> be	334	The default callback received all messages not matched by a more specific
193	a local port.	335	C<tag> match.
194		336
195	The callback has to return a true value when its work is done, after	337	=item rcv $local_port, tag => $callback->(@msg_without_tag), ...
196	which is will be removed, or a false value in which case it will stay
197	registered.
198		338
199	If the match is an array reference, then it will be matched against the	339	Register (or replace) callbacks to be called on messages starting with the
200	first elements of the message, otherwise only the first element is being	340	given tag on the given port (and return the port), or unregister it (when
201	matched.	341	C<$callback> is C<$undef> or missing). There can only be one callback
		342	registered for each tag.
202		343
203	Any element in the match that is specified as C<_any_> (a function	344	The original message will be passed to the callback, after the first
204	exported by this module) matches any single element of the message.	345	element (the tag) has been removed. The callback will use the same
		346	environment as the default callback (see above).
205		347
206	While not required, it is highly recommended that the first matching	348	Example: create a port and bind receivers on it in one go.
207	element is a string identifying the message. The one-string-only match is	349
208	also the most efficient match (by far).	350	my $port = rcv port,
		351	msg1 => sub { ... },
		352	msg2 => sub { ... },
		353	;
		354
		355	Example: create a port, bind receivers and send it in a message elsewhere
		356	in one go:
		357
		358	snd $otherport, reply =>
		359	rcv port,
		360	msg1 => sub { ... },
		361	...
		362	;
		363
		364	Example: temporarily register a rcv callback for a tag matching some port
		365	(e.g. for a rpc reply) and unregister it after a message was received.
		366
		367	rcv $port, $otherport => sub {
		368	my @reply = @_;
		369
		370	rcv $SELF, $otherport;
		371	};
209		372
210	=cut	373	=cut
211		374
212	sub rcv($@) {	375	sub rcv($@) {
213	my ($port, $match, $cb) = @_;	376	my $port = shift;
214
215	my $port = $PORT{$port}
216	or do {
217	my ($noderef, $lport) = split /#/, $port;	377	my ($nodeid, $portid) = split /#/, $port, 2;
218	"AnyEvent::MP::Node::Self" eq ref $NODE{$noderef}	378
		379	$NODE{$nodeid} == $NODE{""}
219	or Carp::croak "$port: can only rcv on local ports";	380	or Carp::croak "$port: rcv can only be called on local ports, caught";
220		381
221	$PORT{$lport}	382	while (@_) {
222	or Carp::croak "$port: port does not exist";	383	if (ref $_[0]) {
223		384	if (my $self = $PORT_DATA{$portid}) {
224	$PORT{$port} = $PORT{$lport} # also return	385	"AnyEvent::MP::Port" eq ref $self
225	};	386	or Carp::croak "$port: rcv can only be called on message matching ports, caught";
226		387
227	if (!ref $match) {	388	$self->[2] = shift;
228	push @{ $port->{rc0}{$match} }, [$cb];
229	} elsif (("ARRAY" eq ref $match && !ref $match->[0])) {
230	my ($type, @match) = @$match;
231	@match
232	? push @{ $port->{rcv}{$match->[0]} }, [$cb, \@match]
233	: push @{ $port->{rc0}{$match->[0]} }, [$cb];
234	} else {	389	} else {
235	push @{ $port->{any} }, [$cb, $match];	390	my $cb = shift;
236	}	391	$PORT{$portid} = sub {
237	}	392	local $SELF = $port;
238		393	eval { &$cb }; _self_die if $@;
239	sub _inject {
240	my ($port, $msg) = @{+shift};
241
242	$port = $PORT{$port}
243	or return;
244
245	@_ = @$msg;
246
247	for (@{ $port->{rc0}{$msg->[0]} }) {
248	$_ && &{$_->[0]}
249	&& undef $_;
250	}
251
252	for (@{ $port->{rcv}{$msg->[0]} }) {
253	$_ && [@_[1..$#{$_->[1]}]] ~~ $_->[1]
254	&& &{$_->[0]}
255	&& undef $_;
256	}
257
258	for (@{ $port->{any} }) {
259	$_ && [@_[0..$#{$_->[1]}]] ~~ $_->[1]
260	&& &{$_->[0]}
261	&& undef $_;
262	}
263	}
264
265	sub normalise_noderef($) {
266	my ($noderef) = @_;
267
268	my $cv = AE::cv;
269	my @res;
270
271	$cv->begin (sub {
272	my %seen;
273	my @refs;
274	for (sort { $a->[0] <=> $b->[0] } @res) {
275	push @refs, $_->[1] unless $seen{$_->[1]}++
276	}
277	shift->send (join ",", @refs);
278	});
279
280	$noderef = $DEFAULT_PORT unless length $noderef;
281
282	my $idx;
283	for my $t (split /,/, $noderef) {
284	my $pri = ++$idx;
285
286	#TODO: this should be outside normalise_noderef and in become_public
287	if ($t =~ /^\d*$/) {
288	my $nodename = (POSIX::uname)[1];
289
290	$cv->begin;
291	AnyEvent::Socket::resolve_sockaddr $nodename, $t \|\| "aemp=$DEFAULT_PORT", "tcp", 0, undef, sub {
292	for (@_) {
293	my ($service, $host) = AnyEvent::Socket::unpack_sockaddr $_->[3];
294	push @res, [
295	$pri += 1e-5,
296	AnyEvent::Socket::format_hostport AnyEvent::Socket::format_address $host, $service
297	];
298	}	394	};
299	$cv->end;	395	}
		396	} elsif (defined $_[0]) {
		397	my $self = $PORT_DATA{$portid} \|\|= do {
		398	my $self = bless [$PORT{$port} \|\| sub { }, { }, $port], "AnyEvent::MP::Port";
		399
		400	$PORT{$portid} = sub {
		401	local $SELF = $port;
		402
		403	if (my $cb = $self->[1]{$_[0]}) {
		404	shift;
		405	eval { &$cb }; _self_die if $@;
		406	} else {
		407	&{ $self->[0] };
		408	}
		409	};
		410
		411	$self
300	};	412	};
301		413
302	# my (undef, undef, undef, undef, @ipv4) = gethostbyname $nodename;	414	"AnyEvent::MP::Port" eq ref $self
303	#	415	or Carp::croak "$port: rcv can only be called on message matching ports, caught";
304	# for (@ipv4) {	416
305	# push @res, [	417	my ($tag, $cb) = splice @_, 0, 2;
306	# $pri,	418
307	# AnyEvent::Socket::format_hostport AnyEvent::Socket::format_address $_, $t \|\| $DEFAULT_PORT,	419	if (defined $cb) {
308	# ];	420	$self->[1]{$tag} = $cb;
309	# }
310	} else {	421	} else {
311	my ($host, $port) = AnyEvent::Socket::parse_hostport $t, "aemp=$DEFAULT_PORT"	422	delete $self->[1]{$tag};
312	or Carp::croak "$t: unparsable transport descriptor";
313
314	$cv->begin;
315	AnyEvent::Socket::resolve_sockaddr $host, $port, "tcp", 0, undef, sub {
316	for (@_) {
317	my ($service, $host) = AnyEvent::Socket::unpack_sockaddr $_->[3];
318	push @res, [
319	$pri += 1e-5,
320	AnyEvent::Socket::format_hostport AnyEvent::Socket::format_address $host, $service
321	];
322	}
323	$cv->end;
324	}	423	}
325	}	424	}
326	}	425	}
327		426
328	$cv->end;	427	$port
329
330	$cv
331	}	428	}
332		429
333	sub become_public {	430	=item $closure = psub { BLOCK }
334	return if $PUBLIC;
335		431
336	my $noderef = join ",", ref $_[0] ? @{+shift} : shift;	432	Remembers C<$SELF> and creates a closure out of the BLOCK. When the
337	my @args = @_;	433	closure is executed, sets up the environment in the same way as in C<rcv>
		434	callbacks, i.e. runtime errors will cause the port to get C<kil>ed.
338		435
339	$NODE = (normalise_noderef $noderef)->recv;	436	This is useful when you register callbacks from C<rcv> callbacks:
340		437
341	for my $t (split /,/, $NODE) {	438	rcv delayed_reply => sub {
342	$NODE{$t} = $NODE{""};	439	my ($delay, @reply) = @_;
343		440	my $timer = AE::timer $delay, 0, psub {
344	my ($host, $port) = AnyEvent::Socket::parse_hostport $t;	441	snd @reply, $SELF;
345
346	$LISTENER{$t} = AnyEvent::MP::Transport::mp_server $host, $port,
347	@args,
348	on_error => sub {
349	die "on_error<@_>\n";#d#
350	},
351	on_connect => sub {
352	my ($tp) = @_;
353
354	$NODE{$tp->{remote_id}} = $_[0];
355	},
356	sub {
357	my ($tp) = @_;
358
359	$NODE{"$tp->{peerhost}:$tp->{peerport}"} = $tp;
360	},
361	;	442	};
		443	};
		444
		445	=cut
		446
		447	sub psub(&) {
		448	my $cb = shift;
		449
		450	my $port = $SELF
		451	or Carp::croak "psub can only be called from within rcv or psub callbacks, not";
		452
		453	sub {
		454	local $SELF = $port;
		455
		456	if (wantarray) {
		457	my @res = eval { &$cb };
		458	_self_die if $@;
		459	@res
		460	} else {
		461	my $res = eval { &$cb };
		462	_self_die if $@;
		463	$res
		464	}
362	}	465	}
		466	}
363		467
364	$PUBLIC = 1;	468	=item $guard = mon $port, $cb->(@reason) # call $cb when $port dies
		469
		470	=item $guard = mon $port, $rcvport # kill $rcvport when $port dies
		471
		472	=item $guard = mon $port # kill $SELF when $port dies
		473
		474	=item $guard = mon $port, $rcvport, @msg # send a message when $port dies
		475
		476	Monitor the given port and do something when the port is killed or
		477	messages to it were lost, and optionally return a guard that can be used
		478	to stop monitoring again.
		479
		480	In the first form (callback), the callback is simply called with any
		481	number of C<@reason> elements (no @reason means that the port was deleted
		482	"normally"). Note also that I<< the callback B<must> never die >>, so use
		483	C<eval> if unsure.
		484
		485	In the second form (another port given), the other port (C<$rcvport>)
		486	will be C<kil>'ed with C<@reason>, if a @reason was specified, i.e. on
		487	"normal" kils nothing happens, while under all other conditions, the other
		488	port is killed with the same reason.
		489
		490	The third form (kill self) is the same as the second form, except that
		491	C<$rvport> defaults to C<$SELF>.
		492
		493	In the last form (message), a message of the form C<@msg, @reason> will be
		494	C<snd>.
		495
		496	Monitoring-actions are one-shot: once messages are lost (and a monitoring
		497	alert was raised), they are removed and will not trigger again.
		498
		499	As a rule of thumb, monitoring requests should always monitor a port from
		500	a local port (or callback). The reason is that kill messages might get
		501	lost, just like any other message. Another less obvious reason is that
		502	even monitoring requests can get lost (for example, when the connection
		503	to the other node goes down permanently). When monitoring a port locally
		504	these problems do not exist.
		505
		506	C<mon> effectively guarantees that, in the absence of hardware failures,
		507	after starting the monitor, either all messages sent to the port will
		508	arrive, or the monitoring action will be invoked after possible message
		509	loss has been detected. No messages will be lost "in between" (after
		510	the first lost message no further messages will be received by the
		511	port). After the monitoring action was invoked, further messages might get
		512	delivered again.
		513
		514	Inter-host-connection timeouts and monitoring depend on the transport
		515	used. The only transport currently implemented is TCP, and AnyEvent::MP
		516	relies on TCP to detect node-downs (this can take 10-15 minutes on a
		517	non-idle connection, and usually around two hours for idle conenctions).
		518
		519	This means that monitoring is good for program errors and cleaning up
		520	stuff eventually, but they are no replacement for a timeout when you need
		521	to ensure some maximum latency.
		522
		523	Example: call a given callback when C<$port> is killed.
		524
		525	mon $port, sub { warn "port died because of <@_>\n" };
		526
		527	Example: kill ourselves when C<$port> is killed abnormally.
		528
		529	mon $port;
		530
		531	Example: send us a restart message when another C<$port> is killed.
		532
		533	mon $port, $self => "restart";
		534
		535	=cut
		536
		537	sub mon {
		538	my ($nodeid, $port) = split /#/, shift, 2;
		539
		540	my $node = $NODE{$nodeid} \|\| add_node $nodeid;
		541
		542	my $cb = @_ ? shift : $SELF \|\| Carp::croak 'mon: called with one argument only, but $SELF not set,';
		543
		544	unless (ref $cb) {
		545	if (@_) {
		546	# send a kill info message
		547	my (@msg) = ($cb, @_);
		548	$cb = sub { snd @msg, @_ };
		549	} else {
		550	# simply kill other port
		551	my $port = $cb;
		552	$cb = sub { kil $port, @_ if @_ };
		553	}
		554	}
		555
		556	$node->monitor ($port, $cb);
		557
		558	defined wantarray
		559	and AnyEvent::Util::guard { $node->unmonitor ($port, $cb) }
		560	}
		561
		562	=item $guard = mon_guard $port, $ref, $ref...
		563
		564	Monitors the given C<$port> and keeps the passed references. When the port
		565	is killed, the references will be freed.
		566
		567	Optionally returns a guard that will stop the monitoring.
		568
		569	This function is useful when you create e.g. timers or other watchers and
		570	want to free them when the port gets killed (note the use of C<psub>):
		571
		572	$port->rcv (start => sub {
		573	my $timer; $timer = mon_guard $port, AE::timer 1, 1, psub {
		574	undef $timer if 0.9 < rand;
		575	});
		576	});
		577
		578	=cut
		579
		580	sub mon_guard {
		581	my ($port, @refs) = @_;
		582
		583	#TODO: mon-less form?
		584
		585	mon $port, sub { 0 && @refs }
		586	}
		587
		588	=item kil $port[, @reason]
		589
		590	Kill the specified port with the given C<@reason>.
		591
		592	If no C<@reason> is specified, then the port is killed "normally" (ports
		593	monitoring other ports will not necessarily die because a port dies
		594	"normally").
		595
		596	Otherwise, linked ports get killed with the same reason (second form of
		597	C<mon>, see above).
		598
		599	Runtime errors while evaluating C<rcv> callbacks or inside C<psub> blocks
		600	will be reported as reason C<< die => $@ >>.
		601
		602	Transport/communication errors are reported as C<< transport_error =>
		603	$message >>.
		604
		605	=cut
		606
		607	=item $port = spawn $node, $initfunc[, @initdata]
		608
		609	Creates a port on the node C<$node> (which can also be a port ID, in which
		610	case it's the node where that port resides).
		611
		612	The port ID of the newly created port is returned immediately, and it is
		613	possible to immediately start sending messages or to monitor the port.
		614
		615	After the port has been created, the init function is called on the remote
		616	node, in the same context as a C<rcv> callback. This function must be a
		617	fully-qualified function name (e.g. C<MyApp::Chat::Server::init>). To
		618	specify a function in the main program, use C<::name>.
		619
		620	If the function doesn't exist, then the node tries to C<require>
		621	the package, then the package above the package and so on (e.g.
		622	C<MyApp::Chat::Server>, C<MyApp::Chat>, C<MyApp>) until the function
		623	exists or it runs out of package names.
		624
		625	The init function is then called with the newly-created port as context
		626	object (C<$SELF>) and the C<@initdata> values as arguments.
		627
		628	A common idiom is to pass a local port, immediately monitor the spawned
		629	port, and in the remote init function, immediately monitor the passed
		630	local port. This two-way monitoring ensures that both ports get cleaned up
		631	when there is a problem.
		632
		633	Example: spawn a chat server port on C<$othernode>.
		634
		635	# this node, executed from within a port context:
		636	my $server = spawn $othernode, "MyApp::Chat::Server::connect", $SELF;
		637	mon $server;
		638
		639	# init function on C<$othernode>
		640	sub connect {
		641	my ($srcport) = @_;
		642
		643	mon $srcport;
		644
		645	rcv $SELF, sub {
		646	...
		647	};
		648	}
		649
		650	=cut
		651
		652	sub _spawn {
		653	my $port = shift;
		654	my $init = shift;
		655
		656	local $SELF = "$NODE#$port";
		657	eval {
		658	&{ load_func $init }
		659	};
		660	_self_die if $@;
		661	}
		662
		663	sub spawn(@) {
		664	my ($nodeid, undef) = split /#/, shift, 2;
		665
		666	my $id = "$RUNIQ." . $ID++;
		667
		668	$_[0] =~ /::/
		669	or Carp::croak "spawn init function must be a fully-qualified name, caught";
		670
		671	snd_to_func $nodeid, "AnyEvent::MP::_spawn" => $id, @_;
		672
		673	"$nodeid#$id"
		674	}
		675
		676	=item after $timeout, @msg
		677
		678	=item after $timeout, $callback
		679
		680	Either sends the given message, or call the given callback, after the
		681	specified number of seconds.
		682
		683	This is simply a utility function that comes in handy at times - the
		684	AnyEvent::MP author is not convinced of the wisdom of having it, though,
		685	so it may go away in the future.
		686
		687	=cut
		688
		689	sub after($@) {
		690	my ($timeout, @action) = @_;
		691
		692	my $t; $t = AE::timer $timeout, 0, sub {
		693	undef $t;
		694	ref $action[0]
		695	? $action[0]()
		696	: snd @action;
		697	};
365	}	698	}
366		699
367	=back	700	=back
368		701
369	=head1 NODE MESSAGES	702	=head1 AnyEvent::MP vs. Distributed Erlang
370		703
371	Nodes understand the following messages sent to them. Many of them take	704	AnyEvent::MP got lots of its ideas from distributed Erlang (Erlang node
372	arguments called C<@reply>, which will simply be used to compose a reply	705	== aemp node, Erlang process == aemp port), so many of the documents and
373	message - C<$reply[0]> is the port to reply to, C<$reply[1]> the type and	706	programming techniques employed by Erlang apply to AnyEvent::MP. Here is a
374	the remaining arguments are simply the message data.	707	sample:
		708
		709	http://www.Erlang.se/doc/programming_rules.shtml
		710	http://Erlang.org/doc/getting_started/part_frame.html # chapters 3 and 4
		711	http://Erlang.org/download/Erlang-book-part1.pdf # chapters 5 and 6
		712	http://Erlang.org/download/armstrong_thesis_2003.pdf # chapters 4 and 5
		713
		714	Despite the similarities, there are also some important differences:
375		715
376	=over 4	716	=over 4
377		717
378	=cut	718	=item * Node IDs are arbitrary strings in AEMP.
379		719
380	#############################################################################	720	Erlang relies on special naming and DNS to work everywhere in the same
381	# self node code	721	way. AEMP relies on each node somehow knowing its own address(es) (e.g. by
		722	configuration or DNS), but will otherwise discover other odes itself.
382		723
383	sub _new_port($) {	724	=item * Erlang has a "remote ports are like local ports" philosophy, AEMP
384	my ($name) = @_;	725	uses "local ports are like remote ports".
385		726
386	my ($noderef, $portname) = split /#/, $name;	727	The failure modes for local ports are quite different (runtime errors
		728	only) then for remote ports - when a local port dies, you I<know> it dies,
		729	when a connection to another node dies, you know nothing about the other
		730	port.
387		731
388	$PORT{$name} =	732	Erlang pretends remote ports are as reliable as local ports, even when
389	$PORT{$portname} = {	733	they are not.
390	names => [$name, $portname],
391	};
392	}
393		734
394	$NODE{""} = new AnyEvent::MP::Node::Self noderef => $NODE;	735	AEMP encourages a "treat remote ports differently" philosophy, with local
395	_new_port "";	736	ports being the special case/exception, where transport errors cannot
		737	occur.
396		738
397	=item relay => $port, @msg	739	=item * Erlang uses processes and a mailbox, AEMP does not queue.
398		740
399	Simply forwards the message to the given port.	741	Erlang uses processes that selectively receive messages, and therefore
		742	needs a queue. AEMP is event based, queuing messages would serve no
		743	useful purpose. For the same reason the pattern-matching abilities of
		744	AnyEvent::MP are more limited, as there is little need to be able to
		745	filter messages without dequeuing them.
400		746
401	=cut	747	(But see L<Coro::MP> for a more Erlang-like process model on top of AEMP).
402		748
403	rcv "", relay => \&snd;	749	=item * Erlang sends are synchronous, AEMP sends are asynchronous.
404		750
405	=item eval => $string[ @reply]	751	Sending messages in Erlang is synchronous and blocks the process (and
		752	so does not need a queue that can overflow). AEMP sends are immediate,
		753	connection establishment is handled in the background.
406		754
407	Evaluates the given string. If C<@reply> is given, then a message of the	755	=item * Erlang suffers from silent message loss, AEMP does not.
408	form C<@reply, $@, @evalres> is sent.
409		756
410	Example: crash another node.	757	Erlang makes few guarantees on messages delivery - messages can get lost
		758	without any of the processes realising it (i.e. you send messages a, b,
		759	and c, and the other side only receives messages a and c).
411		760
412	snd $othernode, eval => "exit";	761	AEMP guarantees correct ordering, and the guarantee that after one message
		762	is lost, all following ones sent to the same port are lost as well, until
		763	monitoring raises an error, so there are no silent "holes" in the message
		764	sequence.
413		765
414	=cut	766	=item * Erlang can send messages to the wrong port, AEMP does not.
415		767
416	rcv "", eval => sub {	768	In Erlang it is quite likely that a node that restarts reuses a process ID
417	my (undef, $string, @reply) = @_;	769	known to other nodes for a completely different process, causing messages
418	my @res = eval $string;	770	destined for that process to end up in an unrelated process.
419	snd @reply, "$@", @res if @reply;
420	};
421		771
422	=item time => @reply	772	AEMP never reuses port IDs, so old messages or old port IDs floating
		773	around in the network will not be sent to an unrelated port.
423		774
424	Replies the the current node time to C<@reply>.	775	=item * Erlang uses unprotected connections, AEMP uses secure
		776	authentication and can use TLS.
425		777
426	Example: tell the current node to send the current time to C<$myport> in a	778	AEMP can use a proven protocol - TLS - to protect connections and
427	C<timereply> message.	779	securely authenticate nodes.
428		780
429	snd $NODE, time => $myport, timereply => 1, 2;	781	=item * The AEMP protocol is optimised for both text-based and binary
430	# => snd $myport, timereply => 1, 2, <time>	782	communications.
431		783
432	=cut	784	The AEMP protocol, unlike the Erlang protocol, supports both programming
		785	language independent text-only protocols (good for debugging) and binary,
		786	language-specific serialisers (e.g. Storable). By default, unless TLS is
		787	used, the protocol is actually completely text-based.
433		788
434	rcv "", time => sub { shift; snd @_, AE::time };	789	It has also been carefully designed to be implementable in other languages
		790	with a minimum of work while gracefully degrading functionality to make the
		791	protocol simple.
		792
		793	=item * AEMP has more flexible monitoring options than Erlang.
		794
		795	In Erlang, you can chose to receive I<all> exit signals as messages
		796	or I<none>, there is no in-between, so monitoring single processes is
		797	difficult to implement. Monitoring in AEMP is more flexible than in
		798	Erlang, as one can choose between automatic kill, exit message or callback
		799	on a per-process basis.
		800
		801	=item * Erlang tries to hide remote/local connections, AEMP does not.
		802
		803	Monitoring in Erlang is not an indicator of process death/crashes, in the
		804	same way as linking is (except linking is unreliable in Erlang).
		805
		806	In AEMP, you don't "look up" registered port names or send to named ports
		807	that might or might not be persistent. Instead, you normally spawn a port
		808	on the remote node. The init function monitors you, and you monitor the
		809	remote port. Since both monitors are local to the node, they are much more
		810	reliable (no need for C<spawn_link>).
		811
		812	This also saves round-trips and avoids sending messages to the wrong port
		813	(hard to do in Erlang).
435		814
436	=back	815	=back
437		816
		817	=head1 RATIONALE
		818
		819	=over 4
		820
		821	=item Why strings for port and node IDs, why not objects?
		822
		823	We considered "objects", but found that the actual number of methods
		824	that can be called are quite low. Since port and node IDs travel over
		825	the network frequently, the serialising/deserialising would add lots of
		826	overhead, as well as having to keep a proxy object everywhere.
		827
		828	Strings can easily be printed, easily serialised etc. and need no special
		829	procedures to be "valid".
		830
		831	And as a result, a miniport consists of a single closure stored in a
		832	global hash - it can't become much cheaper.
		833
		834	=item Why favour JSON, why not a real serialising format such as Storable?
		835
		836	In fact, any AnyEvent::MP node will happily accept Storable as framing
		837	format, but currently there is no way to make a node use Storable by
		838	default (although all nodes will accept it).
		839
		840	The default framing protocol is JSON because a) JSON::XS is many times
		841	faster for small messages and b) most importantly, after years of
		842	experience we found that object serialisation is causing more problems
		843	than it solves: Just like function calls, objects simply do not travel
		844	easily over the network, mostly because they will always be a copy, so you
		845	always have to re-think your design.
		846
		847	Keeping your messages simple, concentrating on data structures rather than
		848	objects, will keep your messages clean, tidy and efficient.
		849
		850	=back
		851
438	=head1 SEE ALSO	852	=head1 SEE ALSO
		853
		854	L<AnyEvent::MP::Intro> - a gentle introduction.
		855
		856	L<AnyEvent::MP::Kernel> - more, lower-level, stuff.
		857
		858	L<AnyEvent::MP::Global> - network maintainance and port groups, to find
		859	your applications.
439		860
440	L<AnyEvent>.	861	L<AnyEvent>.
441		862
442	=head1 AUTHOR	863	=head1 AUTHOR
443		864

Diff Legend

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

Comparing AnyEvent-MP/MP.pm (file contents): Revision 1.5 by root, Sat Aug 1 07:44:02 2009 UTC vs. Revision 1.79 by root, Fri Sep 4 21:52:09 2009 UTC

Diff Legend

Comparing AnyEvent-MP/MP.pm (file contents):
Revision 1.5 by root, Sat Aug 1 07:44:02 2009 UTC vs.
Revision 1.79 by root, Fri Sep 4 21:52:09 2009 UTC