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

Comparing cvsroot/AnyEvent-MP/MP.pm (file contents):
Revision 1.3 by root, Sat Aug 1 07:11:45 2009 UTC vs.
Revision 1.70 by root, Sun Aug 30 19:49:47 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 noderef
		10	NODE # returns this node's noderef
		11	NODE $port # returns the noderef of the port
11		12
		13	$SELF # receiving/own port id in rcv callbacks
		14
		15	# initialise the node so it can send/receive messages
		16	initialise_node;
		17
		18	# ports are message endpoints
		19
		20	# sending messages
12	snd $port, type => data...;	21	snd $port, type => data...;
		22	snd $port, @msg;
		23	snd @msg_with_first_element_being_a_port;
13		24
14	rcv $port, smartmatch => $cb->($port, @msg);	25	# creating/using ports, the simple way
		26	my $simple_port = port { my @msg = @_; 0 };
15		27
16	# examples:	28	# creating/using ports, tagged message matching
		29	my $port = port;
17	rcv $port2, ping => sub { snd $_[0], "pong"; 0 };	30	rcv $port, ping => sub { snd $_[0], "pong"; 0 };
18	rcv $port1, pong => sub { warn "pong received\n" };	31	rcv $port, pong => sub { warn "pong received\n"; 0 };
19	snd $port2, ping => $port1;
20		32
21	# more, smarter, matches (_any_ is exported by this module)	33	# create a port on another node
22	rcv $port, [child_died => $pid] => sub { ...	34	my $port = spawn $node, $initfunc, @initdata;
23	rcv $port, [_any_, _any_, 3] => sub { .. $_[2] is 3	35
		36	# monitoring
		37	mon $port, $cb->(@msg) # callback is invoked on death
		38	mon $port, $otherport # kill otherport on abnormal death
		39	mon $port, $otherport, @msg # send message on death
		40
		41	=head1 CURRENT STATUS
		42
		43	AnyEvent::MP - stable API, should work
		44	AnyEvent::MP::Intro - outdated
		45	AnyEvent::MP::Kernel - mostly stable
		46	AnyEvent::MP::Global - mostly stable
		47	AnyEvent::MP::Node - mostly stable, but internal anyways
		48	AnyEvent::MP::Transport - mostly stable, but internal anyways
		49
		50	stay tuned.
24		51
25	=head1 DESCRIPTION	52	=head1 DESCRIPTION
26		53
27	This module (-family) implements a simple message passing framework.	54	This module (-family) implements a simple message passing framework.
28		55
29	Despite its simplicity, you can securely message other processes running	56	Despite its simplicity, you can securely message other processes running
30	on the same or other hosts.	57	on the same or other hosts, and you can supervise entities remotely.
		58
		59	For an introduction to this module family, see the L<AnyEvent::MP::Intro>
		60	manual page and the examples under F<eg/>.
		61
		62	At the moment, this module family is a bit underdocumented.
31		63
32	=head1 CONCEPTS	64	=head1 CONCEPTS
33		65
34	=over 4	66	=over 4
35		67
36	=item port	68	=item port
37		69
38	A port is something you can send messages to with the C<snd> function, and	70	A port is something you can send messages to (with the C<snd> function).
39	you can register C<rcv> handlers with. All C<rcv> handlers will receive
40	messages they match, messages will not be queued.
41		71
		72	Ports allow you to register C<rcv> handlers that can match all or just
		73	some messages. Messages send to ports will not be queued, regardless of
		74	anything was listening for them or not.
		75
42	=item port id - C<noderef#portname>	76	=item port ID - C<nodeid#portname>
43		77
44	A port id is always the noderef, a hash-mark (C<#>) as separator, followed	78	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).	79	separator, and a port name (a printable string of unspecified format).
46		80
47	=item node	81	=item node
48		82
49	A node is a single process containing at least one port - the node	83	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,	84	which enables nodes to manage each other remotely, and to create new
51	among other things.	85	ports.
52		86
53	Initially, nodes are either private (single-process only) or hidden	87	Nodes are either public (have one or more listening ports) or private
54	(connected to a master node only). Only when they epxlicitly "become	88	(no listening ports). Private nodes cannot talk to other private nodes
55	public" can you send them messages from unrelated other nodes.	89	currently.
56		90
57	=item noderef - C<host:port,host:port...>, C<id@noderef, C<id>	91	=item node ID - C<[a-za-Z0-9_\-.:]+>
58		92
59	A noderef is a string that either uniquely identifies a given node (for	93	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	94	network. Depending on the configuration used, node IDs can look like a
61	node (for public nodes).	95	hostname, a hostname and a port, or a random string. AnyEvent::MP itself
		96	doesn't interpret node IDs in any way.
		97
		98	=item binds - C<ip:port>
		99
		100	Nodes can only talk to each other by creating some kind of connection to
		101	each other. To do this, nodes should listen on one or more local transport
		102	endpoints - binds. Currently, only standard C<ip:port> specifications can
		103	be used, which specify TCP ports to listen on.
		104
		105	=item seeds - C<host:port>
		106
		107	When a node starts, it knows nothing about the network. To teach the node
		108	about the network it first has to contact some other node within the
		109	network. This node is called a seed.
		110
		111	Seeds are transport endpoint(s) of as many nodes as one wants. Those nodes
		112	are expected to be long-running, and at least one of those should always
		113	be available. When nodes run out of connections (e.g. due to a network
		114	error), they try to re-establish connections to some seednodes again to
		115	join the network.
		116
		117	Apart from being sued for seeding, seednodes are not special in any way -
		118	every public node can be a seednode.
62		119
63	=back	120	=back
64		121
65	=head1 VARIABLES/FUNCTIONS	122	=head1 VARIABLES/FUNCTIONS
66		123
…		…
68		125
69	=cut	126	=cut
70		127
71	package AnyEvent::MP;	128	package AnyEvent::MP;
72		129
73	use AnyEvent::MP::Util ();
74	use AnyEvent::MP::Node;	130	use AnyEvent::MP::Kernel;
75	use AnyEvent::MP::Transport;
76		131
77	use utf8;
78	use common::sense;	132	use common::sense;
79		133
80	use Carp ();	134	use Carp ();
81		135
82	use AE ();	136	use AE ();
83		137
84	use base "Exporter";	138	use base "Exporter";
85		139
86	our $VERSION = '0.0';	140	our $VERSION = $AnyEvent::MP::Kernel::VERSION;
87	our @EXPORT = qw(NODE $NODE $PORT snd rcv _any_);
88		141
89	our $DEFAULT_SECRET;	142	our @EXPORT = qw(
90	our $DEFAULT_PORT = "4040";	143	NODE $NODE *SELF node_of after
		144	initialise_node
		145	snd rcv mon mon_guard kil reg psub spawn
		146	port
		147	);
91		148
92	our $CONNECT_INTERVAL = 5; # new connect every 5s, at least	149	our $SELF;
93	our $CONNECT_TIMEOUT = 30; # includes handshake
94		150
95	sub default_secret {	151	sub _self_die() {
96	unless (defined $DEFAULT_SECRET) {	152	my $msg = $@;
97	if (open my $fh, "<$ENV{HOME}/.aemp-secret") {	153	$msg =~ s/\n+$// unless ref $msg;
98	sysread $fh, $DEFAULT_SECRET, -s $fh;	154	kil $SELF, die => $msg;
99	} else {
100	$DEFAULT_SECRET = AnyEvent::MP::Util::nonce 32;
101	}
102	}
103
104	$DEFAULT_SECRET
105	}	155	}
106		156
107	=item NODE / $NODE	157	=item $thisnode = NODE / $NODE
108		158
109	The C<NODE ()> function and the C<$NODE> variable contain the noderef of	159	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	160	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.	161	a call to C<initialise_node>.
112		162
113	=cut	163	=item $nodeid = node_of $port
114		164
115	our $UNIQ = sprintf "%x.%x", $$, time; # per-process/node unique cookie	165	Extracts and returns the node ID from a port ID or a node ID.
116	our $PUBLIC = 0;
117	our $NODE;
118	our $PORT;
119		166
120	our %NODE; # node id to transport mapping, or "undef", for local node	167	=item initialise_node $profile_name, key => value...
121	our %PORT; # local ports
122	our %LISTENER; # local transports
123		168
124	sub NODE() { $NODE }	169	Before a node can talk to other nodes on the network (i.e. enter
		170	"distributed mode") it has to initialise itself - the minimum a node needs
		171	to know is its own name, and optionally it should know the addresses of
		172	some other nodes in the network to discover other nodes.
125		173
126	{	174	This function initialises a node - it must be called exactly once (or
127	use POSIX ();	175	never) before calling other AnyEvent::MP functions.
128	my $nodename = (POSIX::uname)[1];
129	$NODE = "$$\@$nodename";
130	}
131		176
132	sub _ANY_() { 1 }	177	The first argument is a profile name. If it is C<undef> or missing, then
133	sub _any_() { \&_ANY_ }	178	the current nodename will be used instead (i.e. F<uname -n>).
134		179
135	sub add_node {	180	The function first looks up the profile in the aemp configuration (see the
136	my ($noderef) = @_;	181	L<aemp> commandline utility). the profile is calculated as follows:
137		182
138	return $NODE{$noderef}	183	First, all remaining key => value pairs (all of which are conviniently
139	if exists $NODE{$noderef};	184	undocumented at the moment) will be used. Then they will be overwritten by
		185	any values specified in the global default configuration (see the F<aemp>
		186	utility), then the chain of profiles selected, if any. That means that
		187	the values specified in the profile have highest priority and the values
		188	specified via C<initialise_node> have lowest priority.
140		189
141	for (split /,/, $noderef) {	190	If the profile specifies a node ID, then this will become the node ID of
142	return $NODE{$noderef} = $NODE{$_}	191	this process. If not, then the profile name will be used as node ID. The
143	if exists $NODE{$_};	192	special node ID of C<anon/> will be replaced by a random node ID.
144	}
145		193
146	# for indirect sends, use a different class	194	The next step is to look up the binds in the profile, followed by binding
147	my $node = new AnyEvent::MP::Node::Direct $noderef;	195	aemp protocol listeners on all binds specified (it is possible and valid
		196	to have no binds, meaning that the node cannot be contacted form the
		197	outside. This means the node cannot talk to other nodes that also have no
		198	binds, but it can still talk to all "normal" nodes).
148		199
149	$NODE{$_} = $node	200	If the profile does not specify a binds list, then a default of C<*> is
150	for $noderef, split /,/, $noderef;	201	used.
151		202
152	$node	203	Lastly, the seeds list from the profile is passed to the
153	}	204	L<AnyEvent::MP::Global> module, which will then use it to keep
		205	connectivity with at least on of those seed nodes at any point in time.
154		206
		207	Example: become a distributed node listening on the guessed noderef, or
		208	the one specified via C<aemp> for the current node. This should be the
		209	most common form of invocation for "daemon"-type nodes.
		210
		211	initialise_node;
		212
		213	Example: become an anonymous node. This form is often used for commandline
		214	clients.
		215
		216	initialise_node "anon/";
		217
		218	Example: become a distributed node. If there is no profile of the given
		219	name, or no binds list was specified, resolve C<localhost:4044> and bind
		220	on the resulting addresses.
		221
		222	initialise_node "localhost:4044";
		223
		224	=item $SELF
		225
		226	Contains the current port id while executing C<rcv> callbacks or C<psub>
		227	blocks.
		228
		229	=item *SELF, SELF, %SELF, @SELF...
		230
		231	Due to some quirks in how perl exports variables, it is impossible to
		232	just export C<$SELF>, all the symbols named C<SELF> are exported by this
		233	module, but only C<$SELF> is currently used.
		234
155	=item snd $portid, type => @data	235	=item snd $port, type => @data
156		236
157	=item snd $portid, @msg	237	=item snd $port, @msg
158		238
159	Send the given message to the given port ID, which can identify either a	239	Send the given message to the given port, which can identify either a
160	local or a remote port.	240	local or a remote port, and must be a port ID.
161		241
162	While the message can be about anything, it is highly recommended to use	242	While the message can be almost anything, it is highly recommended to
163	a constant string as first element.	243	use a string as first element (a port ID, or some word that indicates a
		244	request type etc.) and to consist if only simple perl values (scalars,
		245	arrays, hashes) - if you think you need to pass an object, think again.
164		246
165	The message data effectively becomes read-only after a call to this	247	The message data logically becomes read-only after a call to this
166	function: modifying any argument is not allowed and can cause many	248	function: modifying any argument (or values referenced by them) is
167	problems.	249	forbidden, as there can be considerable time between the call to C<snd>
		250	and the time the message is actually being serialised - in fact, it might
		251	never be copied as within the same process it is simply handed to the
		252	receiving port.
168		253
169	The type of data you can transfer depends on the transport protocol: when	254	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	255	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	256	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	257	that Storable can serialise and deserialise is allowed, and for the local
173	node, anything can be passed.	258	node, anything can be passed. Best rely only on the common denominator of
		259	these.
174		260
175	=cut	261	=item $local_port = port
176		262
177	sub snd($@) {	263	Create a new local port object and returns its port ID. Initially it has
178	my ($noderef, $port) = split /#/, shift, 2;	264	no callbacks set and will throw an error when it receives messages.
179		265
180	add_node $noderef	266	=item $local_port = port { my @msg = @_ }
181	unless exists $NODE{$noderef};
182		267
183	$NODE{$noderef}->send (["$port", [@_]]);	268	Creates a new local port, and returns its ID. Semantically the same as
184	}	269	creating a port and calling C<rcv $port, $callback> on it.
185		270
		271	The block will be called for every message received on the port, with the
		272	global variable C<$SELF> set to the port ID. Runtime errors will cause the
		273	port to be C<kil>ed. The message will be passed as-is, no extra argument
		274	(i.e. no port ID) will be passed to the callback.
		275
		276	If you want to stop/destroy the port, simply C<kil> it:
		277
		278	my $port = port {
		279	my @msg = @_;
		280	...
		281	kil $SELF;
		282	};
		283
		284	=cut
		285
		286	sub rcv($@);
		287
		288	sub _kilme {
		289	die "received message on port without callback";
		290	}
		291
		292	sub port(;&) {
		293	my $id = "$UNIQ." . $ID++;
		294	my $port = "$NODE#$id";
		295
		296	rcv $port, shift \|\| \&_kilme;
		297
		298	$port
		299	}
		300
186	=item rcv $portid, type => $callback->(@msg)	301	=item rcv $local_port, $callback->(@msg)
187		302
188	=item rcv $portid, $smartmatch => $callback->(@msg)	303	Replaces the default callback on the specified port. There is no way to
		304	remove the default callback: use C<sub { }> to disable it, or better
		305	C<kil> the port when it is no longer needed.
189		306
190	=item rcv $portid, [$smartmatch...] => $callback->(@msg)	307	The global C<$SELF> (exported by this module) contains C<$port> while
		308	executing the callback. Runtime errors during callback execution will
		309	result in the port being C<kil>ed.
191		310
192	Register a callback on the port identified by C<$portid>, which I<must> be	311	The default callback received all messages not matched by a more specific
193	a local port.	312	C<tag> match.
194		313
195	The callback has to return a true value when its work is done, after	314	=item rcv $local_port, tag => $callback->(@msg_without_tag), ...
196	which is will be removed, or a false value in which case it will stay
197	registered.
198		315
199	If the match is an array reference, then it will be matched against the	316	Register (or replace) callbacks to be called on messages starting with the
200	first elements of the message, otherwise only the first element is being	317	given tag on the given port (and return the port), or unregister it (when
201	matched.	318	C<$callback> is C<$undef> or missing). There can only be one callback
		319	registered for each tag.
202		320
203	Any element in the match that is specified as C<_any_> (a function	321	The original message will be passed to the callback, after the first
204	exported by this module) matches any single element of the message.	322	element (the tag) has been removed. The callback will use the same
		323	environment as the default callback (see above).
205		324
206	While not required, it is highly recommended that the first matching	325	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	326
208	also the most efficient match (by far).	327	my $port = rcv port,
		328	msg1 => sub { ... },
		329	msg2 => sub { ... },
		330	;
		331
		332	Example: create a port, bind receivers and send it in a message elsewhere
		333	in one go:
		334
		335	snd $otherport, reply =>
		336	rcv port,
		337	msg1 => sub { ... },
		338	...
		339	;
		340
		341	Example: temporarily register a rcv callback for a tag matching some port
		342	(e.g. for a rpc reply) and unregister it after a message was received.
		343
		344	rcv $port, $otherport => sub {
		345	my @reply = @_;
		346
		347	rcv $SELF, $otherport;
		348	};
209		349
210	=cut	350	=cut
211		351
212	sub rcv($@) {	352	sub rcv($@) {
213	my ($port, $match, $cb) = @_;	353	my $port = shift;
214
215	my $port = $PORT{$port}
216	or do {
217	my ($noderef, $lport) = split /#/, $port;	354	my ($noderef, $portid) = split /#/, $port, 2;
218	"AnyEvent::MP::Node::Self" eq ref $NODE{$noderef}	355
		356	$NODE{$noderef} == $NODE{""}
219	or Carp::croak "$port: can only rcv on local ports";	357	or Carp::croak "$port: rcv can only be called on local ports, caught";
220		358
221	$PORT{$lport}	359	while (@_) {
222	or Carp::croak "$port: port does not exist";	360	if (ref $_[0]) {
223		361	if (my $self = $PORT_DATA{$portid}) {
224	$PORT{$port} = $PORT{$lport} # also return	362	"AnyEvent::MP::Port" eq ref $self
225	};	363	or Carp::croak "$port: rcv can only be called on message matching ports, caught";
226		364
227	if (!ref $match) {	365	$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 {	366	} else {
235	push @{ $port->{any} }, [$cb, $match];	367	my $cb = shift;
236	}	368	$PORT{$portid} = sub {
237	}	369	local $SELF = $port;
238		370	eval { &$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	}	371	};
299	$cv->end;	372	}
		373	} elsif (defined $_[0]) {
		374	my $self = $PORT_DATA{$portid} \|\|= do {
		375	my $self = bless [$PORT{$port} \|\| sub { }, { }, $port], "AnyEvent::MP::Port";
		376
		377	$PORT{$portid} = sub {
		378	local $SELF = $port;
		379
		380	if (my $cb = $self->[1]{$_[0]}) {
		381	shift;
		382	eval { &$cb }; _self_die if $@;
		383	} else {
		384	&{ $self->[0] };
		385	}
		386	};
		387
		388	$self
300	};	389	};
301		390
302	# my (undef, undef, undef, undef, @ipv4) = gethostbyname $nodename;	391	"AnyEvent::MP::Port" eq ref $self
303	#	392	or Carp::croak "$port: rcv can only be called on message matching ports, caught";
304	# for (@ipv4) {	393
305	# push @res, [	394	my ($tag, $cb) = splice @_, 0, 2;
306	# $pri,	395
307	# AnyEvent::Socket::format_hostport AnyEvent::Socket::format_address $_, $t \|\| $DEFAULT_PORT,	396	if (defined $cb) {
308	# ];	397	$self->[1]{$tag} = $cb;
309	# }
310	} else {	398	} else {
311	my ($host, $port) = AnyEvent::Socket::parse_hostport $t, "aemp=$DEFAULT_PORT"	399	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	}	400	}
325	}	401	}
326	}	402	}
327		403
328	$cv->end;	404	$port
329
330	$cv
331	}	405	}
332		406
333	sub become_public {	407	=item $closure = psub { BLOCK }
334	return if $PUBLIC;
335		408
336	my $noderef = join ",", ref $_[0] ? @{+shift} : shift;	409	Remembers C<$SELF> and creates a closure out of the BLOCK. When the
337	my @args = @_;	410	closure is executed, sets up the environment in the same way as in C<rcv>
		411	callbacks, i.e. runtime errors will cause the port to get C<kil>ed.
338		412
339	$NODE = (normalise_noderef $noderef)->recv;	413	This is useful when you register callbacks from C<rcv> callbacks:
340		414
341	for my $t (split /,/, $NODE) {	415	rcv delayed_reply => sub {
342	$NODE{$t} = $NODE{""};	416	my ($delay, @reply) = @_;
343		417	my $timer = AE::timer $delay, 0, psub {
344	my ($host, $port) = AnyEvent::Socket::parse_hostport $t;	418	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	;	419	};
		420	};
		421
		422	=cut
		423
		424	sub psub(&) {
		425	my $cb = shift;
		426
		427	my $port = $SELF
		428	or Carp::croak "psub can only be called from within rcv or psub callbacks, not";
		429
		430	sub {
		431	local $SELF = $port;
		432
		433	if (wantarray) {
		434	my @res = eval { &$cb };
		435	_self_die if $@;
		436	@res
		437	} else {
		438	my $res = eval { &$cb };
		439	_self_die if $@;
		440	$res
		441	}
362	}	442	}
363
364	$PUBLIC = 1;
365	}	443	}
366		444
367	#############################################################################	445	=item $guard = mon $port, $cb->(@reason) # call $cb when $port dies
368	# self node code
369		446
370	sub _new_port($) {	447	=item $guard = mon $port, $rcvport # kill $rcvport when $port dies
371	my ($name) = @_;
372		448
		449	=item $guard = mon $port # kill $SELF when $port dies
		450
		451	=item $guard = mon $port, $rcvport, @msg # send a message when $port dies
		452
		453	Monitor the given port and do something when the port is killed or
		454	messages to it were lost, and optionally return a guard that can be used
		455	to stop monitoring again.
		456
		457	C<mon> effectively guarantees that, in the absence of hardware failures,
		458	after starting the monitor, either all messages sent to the port will
		459	arrive, or the monitoring action will be invoked after possible message
		460	loss has been detected. No messages will be lost "in between" (after
		461	the first lost message no further messages will be received by the
		462	port). After the monitoring action was invoked, further messages might get
		463	delivered again.
		464
		465	Note that monitoring-actions are one-shot: once messages are lost (and a
		466	monitoring alert was raised), they are removed and will not trigger again.
		467
		468	In the first form (callback), the callback is simply called with any
		469	number of C<@reason> elements (no @reason means that the port was deleted
		470	"normally"). Note also that I<< the callback B<must> never die >>, so use
		471	C<eval> if unsure.
		472
		473	In the second form (another port given), the other port (C<$rcvport>)
		474	will be C<kil>'ed with C<@reason>, iff a @reason was specified, i.e. on
		475	"normal" kils nothing happens, while under all other conditions, the other
		476	port is killed with the same reason.
		477
		478	The third form (kill self) is the same as the second form, except that
		479	C<$rvport> defaults to C<$SELF>.
		480
		481	In the last form (message), a message of the form C<@msg, @reason> will be
		482	C<snd>.
		483
		484	As a rule of thumb, monitoring requests should always monitor a port from
		485	a local port (or callback). The reason is that kill messages might get
		486	lost, just like any other message. Another less obvious reason is that
		487	even monitoring requests can get lost (for exmaple, when the connection
		488	to the other node goes down permanently). When monitoring a port locally
		489	these problems do not exist.
		490
		491	Example: call a given callback when C<$port> is killed.
		492
		493	mon $port, sub { warn "port died because of <@_>\n" };
		494
		495	Example: kill ourselves when C<$port> is killed abnormally.
		496
		497	mon $port;
		498
		499	Example: send us a restart message when another C<$port> is killed.
		500
		501	mon $port, $self => "restart";
		502
		503	=cut
		504
		505	sub mon {
373	my ($noderef, $portname) = split /#/, $name;	506	my ($noderef, $port) = split /#/, shift, 2;
374		507
375	$PORT{$name} =	508	my $node = $NODE{$noderef} \|\| add_node $noderef;
376	$PORT{$portname} = {	509
377	names => [$name, $portname],	510	my $cb = @_ ? shift : $SELF \|\| Carp::croak 'mon: called with one argument only, but $SELF not set,';
		511
		512	unless (ref $cb) {
		513	if (@_) {
		514	# send a kill info message
		515	my (@msg) = ($cb, @_);
		516	$cb = sub { snd @msg, @_ };
		517	} else {
		518	# simply kill other port
		519	my $port = $cb;
		520	$cb = sub { kil $port, @_ if @_ };
		521	}
		522	}
		523
		524	$node->monitor ($port, $cb);
		525
		526	defined wantarray
		527	and AnyEvent::Util::guard { $node->unmonitor ($port, $cb) }
		528	}
		529
		530	=item $guard = mon_guard $port, $ref, $ref...
		531
		532	Monitors the given C<$port> and keeps the passed references. When the port
		533	is killed, the references will be freed.
		534
		535	Optionally returns a guard that will stop the monitoring.
		536
		537	This function is useful when you create e.g. timers or other watchers and
		538	want to free them when the port gets killed (note the use of C<psub>):
		539
		540	$port->rcv (start => sub {
		541	my $timer; $timer = mon_guard $port, AE::timer 1, 1, psub {
		542	undef $timer if 0.9 < rand;
		543	});
		544	});
		545
		546	=cut
		547
		548	sub mon_guard {
		549	my ($port, @refs) = @_;
		550
		551	#TODO: mon-less form?
		552
		553	mon $port, sub { 0 && @refs }
		554	}
		555
		556	=item kil $port[, @reason]
		557
		558	Kill the specified port with the given C<@reason>.
		559
		560	If no C<@reason> is specified, then the port is killed "normally" (ports
		561	monitoring other ports will not necessarily die because a port dies
		562	"normally").
		563
		564	Otherwise, linked ports get killed with the same reason (second form of
		565	C<mon>, see above).
		566
		567	Runtime errors while evaluating C<rcv> callbacks or inside C<psub> blocks
		568	will be reported as reason C<< die => $@ >>.
		569
		570	Transport/communication errors are reported as C<< transport_error =>
		571	$message >>.
		572
		573	=cut
		574
		575	=item $port = spawn $node, $initfunc[, @initdata]
		576
		577	Creates a port on the node C<$node> (which can also be a port ID, in which
		578	case it's the node where that port resides).
		579
		580	The port ID of the newly created port is returned immediately, and it is
		581	possible to immediately start sending messages or to monitor the port.
		582
		583	After the port has been created, the init function is called on the remote
		584	node, in the same context as a C<rcv> callback. This function must be a
		585	fully-qualified function name (e.g. C<MyApp::Chat::Server::init>). To
		586	specify a function in the main program, use C<::name>.
		587
		588	If the function doesn't exist, then the node tries to C<require>
		589	the package, then the package above the package and so on (e.g.
		590	C<MyApp::Chat::Server>, C<MyApp::Chat>, C<MyApp>) until the function
		591	exists or it runs out of package names.
		592
		593	The init function is then called with the newly-created port as context
		594	object (C<$SELF>) and the C<@initdata> values as arguments.
		595
		596	A common idiom is to pass a local port, immediately monitor the spawned
		597	port, and in the remote init function, immediately monitor the passed
		598	local port. This two-way monitoring ensures that both ports get cleaned up
		599	when there is a problem.
		600
		601	Example: spawn a chat server port on C<$othernode>.
		602
		603	# this node, executed from within a port context:
		604	my $server = spawn $othernode, "MyApp::Chat::Server::connect", $SELF;
		605	mon $server;
		606
		607	# init function on C<$othernode>
		608	sub connect {
		609	my ($srcport) = @_;
		610
		611	mon $srcport;
		612
		613	rcv $SELF, sub {
		614	...
		615	};
		616	}
		617
		618	=cut
		619
		620	sub _spawn {
		621	my $port = shift;
		622	my $init = shift;
		623
		624	local $SELF = "$NODE#$port";
		625	eval {
		626	&{ load_func $init }
378	};	627	};
		628	_self_die if $@;
379	}	629	}
380		630
381	$NODE{""} = new AnyEvent::MP::Node::Self noderef => $NODE;	631	sub spawn(@) {
382	_new_port "";	632	my ($noderef, undef) = split /#/, shift, 2;
383		633
384	rcv "", relay => \&snd;	634	my $id = "$RUNIQ." . $ID++;
		635
		636	$_[0] =~ /::/
		637	or Carp::croak "spawn init function must be a fully-qualified name, caught";
		638
		639	snd_to_func $noderef, "AnyEvent::MP::_spawn" => $id, @_;
		640
		641	"$noderef#$id"
		642	}
		643
		644	=item after $timeout, @msg
		645
		646	=item after $timeout, $callback
		647
		648	Either sends the given message, or call the given callback, after the
		649	specified number of seconds.
		650
		651	This is simply a utility function that comes in handy at times - the
		652	AnyEvent::MP author is not convinced of the wisdom of having it, though,
		653	so it may go away in the future.
		654
		655	=cut
		656
		657	sub after($@) {
		658	my ($timeout, @action) = @_;
		659
		660	my $t; $t = AE::timer $timeout, 0, sub {
		661	undef $t;
		662	ref $action[0]
		663	? $action[0]()
		664	: snd @action;
		665	};
		666	}
385		667
386	=back	668	=back
387		669
		670	=head1 AnyEvent::MP vs. Distributed Erlang
		671
		672	AnyEvent::MP got lots of its ideas from distributed Erlang (Erlang node
		673	== aemp node, Erlang process == aemp port), so many of the documents and
		674	programming techniques employed by Erlang apply to AnyEvent::MP. Here is a
		675	sample:
		676
		677	http://www.Erlang.se/doc/programming_rules.shtml
		678	http://Erlang.org/doc/getting_started/part_frame.html # chapters 3 and 4
		679	http://Erlang.org/download/Erlang-book-part1.pdf # chapters 5 and 6
		680	http://Erlang.org/download/armstrong_thesis_2003.pdf # chapters 4 and 5
		681
		682	Despite the similarities, there are also some important differences:
		683
		684	=over 4
		685
		686	=item * Node IDs are arbitrary strings in AEMP.
		687
		688	Erlang relies on special naming and DNS to work everywhere in the same
		689	way. AEMP relies on each node somehow knowing its own address(es) (e.g. by
		690	configuraiton or DNS), but will otherwise discover other odes itself.
		691
		692	=item * Erlang has a "remote ports are like local ports" philosophy, AEMP
		693	uses "local ports are like remote ports".
		694
		695	The failure modes for local ports are quite different (runtime errors
		696	only) then for remote ports - when a local port dies, you I<know> it dies,
		697	when a connection to another node dies, you know nothing about the other
		698	port.
		699
		700	Erlang pretends remote ports are as reliable as local ports, even when
		701	they are not.
		702
		703	AEMP encourages a "treat remote ports differently" philosophy, with local
		704	ports being the special case/exception, where transport errors cannot
		705	occur.
		706
		707	=item * Erlang uses processes and a mailbox, AEMP does not queue.
		708
		709	Erlang uses processes that selectively receive messages, and therefore
		710	needs a queue. AEMP is event based, queuing messages would serve no
		711	useful purpose. For the same reason the pattern-matching abilities of
		712	AnyEvent::MP are more limited, as there is little need to be able to
		713	filter messages without dequeing them.
		714
		715	(But see L<Coro::MP> for a more Erlang-like process model on top of AEMP).
		716
		717	=item * Erlang sends are synchronous, AEMP sends are asynchronous.
		718
		719	Sending messages in Erlang is synchronous and blocks the process (and
		720	so does not need a queue that can overflow). AEMP sends are immediate,
		721	connection establishment is handled in the background.
		722
		723	=item * Erlang suffers from silent message loss, AEMP does not.
		724
		725	Erlang makes few guarantees on messages delivery - messages can get lost
		726	without any of the processes realising it (i.e. you send messages a, b,
		727	and c, and the other side only receives messages a and c).
		728
		729	AEMP guarantees correct ordering, and the guarantee that after one message
		730	is lost, all following ones sent to the same port are lost as well, until
		731	monitoring raises an error, so there are no silent "holes" in the message
		732	sequence.
		733
		734	=item * Erlang can send messages to the wrong port, AEMP does not.
		735
		736	In Erlang it is quite likely that a node that restarts reuses a process ID
		737	known to other nodes for a completely different process, causing messages
		738	destined for that process to end up in an unrelated process.
		739
		740	AEMP never reuses port IDs, so old messages or old port IDs floating
		741	around in the network will not be sent to an unrelated port.
		742
		743	=item * Erlang uses unprotected connections, AEMP uses secure
		744	authentication and can use TLS.
		745
		746	AEMP can use a proven protocol - TLS - to protect connections and
		747	securely authenticate nodes.
		748
		749	=item * The AEMP protocol is optimised for both text-based and binary
		750	communications.
		751
		752	The AEMP protocol, unlike the Erlang protocol, supports both programming
		753	language independent text-only protocols (good for debugging) and binary,
		754	language-specific serialisers (e.g. Storable). By default, unless TLS is
		755	used, the protocol is actually completely text-based.
		756
		757	It has also been carefully designed to be implementable in other languages
		758	with a minimum of work while gracefully degrading functionality to make the
		759	protocol simple.
		760
		761	=item * AEMP has more flexible monitoring options than Erlang.
		762
		763	In Erlang, you can chose to receive I<all> exit signals as messages
		764	or I<none>, there is no in-between, so monitoring single processes is
		765	difficult to implement. Monitoring in AEMP is more flexible than in
		766	Erlang, as one can choose between automatic kill, exit message or callback
		767	on a per-process basis.
		768
		769	=item * Erlang tries to hide remote/local connections, AEMP does not.
		770
		771	Monitoring in Erlang is not an indicator of process death/crashes, in the
		772	same way as linking is (except linking is unreliable in Erlang).
		773
		774	In AEMP, you don't "look up" registered port names or send to named ports
		775	that might or might not be persistent. Instead, you normally spawn a port
		776	on the remote node. The init function monitors you, and you monitor the
		777	remote port. Since both monitors are local to the node, they are much more
		778	reliable (no need for C<spawn_link>).
		779
		780	This also saves round-trips and avoids sending messages to the wrong port
		781	(hard to do in Erlang).
		782
		783	=back
		784
		785	=head1 RATIONALE
		786
		787	=over 4
		788
		789	=item Why strings for port and node IDs, why not objects?
		790
		791	We considered "objects", but found that the actual number of methods
		792	that can be called are quite low. Since port and node IDs travel over
		793	the network frequently, the serialising/deserialising would add lots of
		794	overhead, as well as having to keep a proxy object everywhere.
		795
		796	Strings can easily be printed, easily serialised etc. and need no special
		797	procedures to be "valid".
		798
		799	And as a result, a miniport consists of a single closure stored in a
		800	global hash - it can't become much cheaper.
		801
		802	=item Why favour JSON, why not a real serialising format such as Storable?
		803
		804	In fact, any AnyEvent::MP node will happily accept Storable as framing
		805	format, but currently there is no way to make a node use Storable by
		806	default (although all nodes will accept it).
		807
		808	The default framing protocol is JSON because a) JSON::XS is many times
		809	faster for small messages and b) most importantly, after years of
		810	experience we found that object serialisation is causing more problems
		811	than it solves: Just like function calls, objects simply do not travel
		812	easily over the network, mostly because they will always be a copy, so you
		813	always have to re-think your design.
		814
		815	Keeping your messages simple, concentrating on data structures rather than
		816	objects, will keep your messages clean, tidy and efficient.
		817
		818	=back
		819
388	=head1 SEE ALSO	820	=head1 SEE ALSO
		821
		822	L<AnyEvent::MP::Intro> - a gentle introduction.
		823
		824	L<AnyEvent::MP::Kernel> - more, lower-level, stuff.
		825
		826	L<AnyEvent::MP::Global> - network maintainance and port groups, to find
		827	your applications.
389		828
390	L<AnyEvent>.	829	L<AnyEvent>.
391		830
392	=head1 AUTHOR	831	=head1 AUTHOR
393		832

Diff Legend

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

Comparing cvsroot/AnyEvent-MP/MP.pm (file contents): Revision 1.3 by root, Sat Aug 1 07:11:45 2009 UTC vs. Revision 1.70 by root, Sun Aug 30 19:49:47 2009 UTC

Diff Legend

Comparing cvsroot/AnyEvent-MP/MP.pm (file contents):
Revision 1.3 by root, Sat Aug 1 07:11:45 2009 UTC vs.
Revision 1.70 by root, Sun Aug 30 19:49:47 2009 UTC