ViewVC Help
View File | Revision Log | Show Annotations | Download File
/cvs/AnyEvent-MP/MP.pm
(Generate patch)

Comparing AnyEvent-MP/MP.pm (file contents):
Revision 1.29 by root, Tue Aug 4 23:16:57 2009 UTC vs.
Revision 1.84 by root, Tue Sep 8 01:42:14 2009 UTC

4 4
5=head1 SYNOPSIS 5=head1 SYNOPSIS
6 6
7 use AnyEvent::MP; 7 use AnyEvent::MP;
8 8
9 $NODE # contains this node's noderef 9 $NODE # contains this node's node ID
10 NODE # returns this node's noderef 10 NODE # returns this node's node ID
11 NODE $port # returns the noderef of the port
12 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
13 snd $port, type => data...; 20 snd $port, type => data...;
21 snd $port, @msg;
22 snd @msg_with_first_element_being_a_port;
14 23
15 $SELF # receiving/own port id in rcv callbacks 24 # creating/using ports, the simple way
25 my $simple_port = port { my @msg = @_ };
16 26
17 rcv $port, smartmatch => $cb->($port, @msg); 27 # creating/using ports, tagged message matching
18 28 my $port = port;
19 # examples:
20 rcv $port2, ping => sub { snd $_[0], "pong"; 0 }; 29 rcv $port, ping => sub { snd $_[0], "pong" };
21 rcv $port1, pong => sub { warn "pong received\n" }; 30 rcv $port, pong => sub { warn "pong received\n" };
22 snd $port2, ping => $port1;
23 31
24 # more, smarter, matches (_any_ is exported by this module) 32 # create a port on another node
25 rcv $port, [child_died => $pid] => sub { ... 33 my $port = spawn $node, $initfunc, @initdata;
26 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
48stay tuned.
27 49
28=head1 DESCRIPTION 50=head1 DESCRIPTION
29 51
30This module (-family) implements a simple message passing framework. 52This module (-family) implements a simple message passing framework.
31 53
32Despite its simplicity, you can securely message other processes running 54Despite its simplicity, you can securely message other processes running
33on the same or other hosts. 55on the same or other hosts, and you can supervise entities remotely.
34 56
35For an introduction to this module family, see the L<AnyEvent::MP::Intro> 57For an introduction to this module family, see the L<AnyEvent::MP::Intro>
36manual page. 58manual page and the examples under F<eg/>.
37
38At the moment, this module family is severly broken and underdocumented,
39so do not use. This was uploaded mainly to reserve the CPAN namespace -
40stay tuned! The basic API should be finished, however.
41 59
42=head1 CONCEPTS 60=head1 CONCEPTS
43 61
44=over 4 62=over 4
45 63
46=item port 64=item port
47 65
48A port is something you can send messages to (with the C<snd> function). 66Not to be confused with a TCP port, a "port" is something you can send
67messages to (with the C<snd> function).
49 68
50Some ports allow you to register C<rcv> handlers that can match specific 69Ports allow you to register C<rcv> handlers that can match all or just
51messages. All C<rcv> handlers will receive messages they match, messages 70some messages. Messages send to ports will not be queued, regardless of
52will not be queued. 71anything was listening for them or not.
53 72
54=item port id - C<noderef#portname> 73=item port ID - C<nodeid#portname>
55 74
56A port id is normaly the concatenation of a noderef, a hash-mark (C<#>) as 75A port ID is the concatenation of a node ID, a hash-mark (C<#>) as
57separator, and a port name (a printable string of unspecified format). An 76separator, and a port name (a printable string of unspecified format).
58exception is the the node port, whose ID is identical to it's node
59reference.
60 77
61=item node 78=item node
62 79
63A node is a single process containing at least one port - the node 80A node is a single process containing at least one port - the node port,
64port. You can send messages to node ports to find existing ports or to 81which enables nodes to manage each other remotely, and to create new
65create new ports, among other things. 82ports.
66 83
67Nodes are either private (single-process only), slaves (connected to a 84Nodes are either public (have one or more listening ports) or private
68master node only) or public nodes (connectable from unrelated nodes). 85(no listening ports). Private nodes cannot talk to other private nodes
86currently.
69 87
70=item noderef - C<host:port,host:port...>, C<id@noderef>, C<id> 88=item node ID - C<[A-Z_][a-zA-Z0-9_\-.:]*>
71 89
72A node reference is a string that either simply identifies the node (for 90A node ID is a string that uniquely identifies the node within a
73private and slave nodes), or contains a recipe on how to reach a given 91network. Depending on the configuration used, node IDs can look like a
74node (for public nodes). 92hostname, a hostname and a port, or a random string. AnyEvent::MP itself
93doesn't interpret node IDs in any way.
75 94
76This recipe is simply a comma-separated list of C<address:port> pairs (for 95=item binds - C<ip:port>
77TCP/IP, other protocols might look different).
78 96
79Node references come in two flavours: resolved (containing only numerical 97Nodes can only talk to each other by creating some kind of connection to
80addresses) or unresolved (where hostnames are used instead of addresses). 98each other. To do this, nodes should listen on one or more local transport
99endpoints - binds. Currently, only standard C<ip:port> specifications can
100be used, which specify TCP ports to listen on.
81 101
82Before using an unresolved node reference in a message you first have to 102=item seed nodes
83resolve it. 103
104When a node starts, it knows nothing about the network. To teach the node
105about the network it first has to contact some other node within the
106network. This node is called a seed.
107
108Apart from the fact that other nodes know them as seed nodes and they have
109to have fixed listening addresses, seed nodes are perfectly normal nodes -
110any node can function as a seed node for others.
111
112In addition to discovering the network, seed nodes are also used to
113maintain the network and to connect nodes that otherwise would have
114trouble connecting. They form the backbone of the AnyEvent::MP network.
115
116Seed nodes are expected to be long-running, and at least one seed node
117should always be available.
118
119=item seeds - C<host:port>
120
121Seeds are transport endpoint(s) (usually a hostname/IP address and a
122TCP port) of nodes thta should be used as seed nodes.
123
124The nodes listening on those endpoints are expected to be long-running,
125and at least one of those should always be available. When nodes run out
126of connections (e.g. due to a network error), they try to re-establish
127connections to some seednodes again to join the network.
84 128
85=back 129=back
86 130
87=head1 VARIABLES/FUNCTIONS 131=head1 VARIABLES/FUNCTIONS
88 132
90 134
91=cut 135=cut
92 136
93package AnyEvent::MP; 137package AnyEvent::MP;
94 138
95use AnyEvent::MP::Base; 139use AnyEvent::MP::Kernel;
96 140
97use common::sense; 141use common::sense;
98 142
99use Carp (); 143use Carp ();
100 144
101use AE (); 145use AE ();
102 146
103use base "Exporter"; 147use base "Exporter";
104 148
105our $VERSION = '0.1'; 149our $VERSION = $AnyEvent::MP::Kernel::VERSION;
150
106our @EXPORT = qw( 151our @EXPORT = qw(
107 NODE $NODE *SELF node_of _any_ 152 NODE $NODE *SELF node_of after
108 resolve_node 153 configure
109 become_slave become_public
110 snd rcv mon kil reg psub 154 snd rcv mon mon_guard kil reg psub spawn
111 port 155 port
112); 156);
113 157
114our $SELF; 158our $SELF;
115 159
119 kil $SELF, die => $msg; 163 kil $SELF, die => $msg;
120} 164}
121 165
122=item $thisnode = NODE / $NODE 166=item $thisnode = NODE / $NODE
123 167
124The C<NODE> function returns, and the C<$NODE> variable contains 168The C<NODE> function returns, and the C<$NODE> variable contains, the node
125the noderef of the local node. The value is initialised by a call 169ID of the node running in the current process. This value is initialised by
126to C<become_public> or C<become_slave>, after which all local port 170a call to C<configure>.
127identifiers become invalid.
128 171
129=item $noderef = node_of $portid 172=item $nodeid = node_of $port
130 173
131Extracts and returns the noderef from a portid or a noderef. 174Extracts and returns the node ID from a port ID or a node ID.
132 175
133=item $cv = resolve_node $noderef 176=item configure $profile, key => value...
134 177
135Takes an unresolved node reference that may contain hostnames and 178=item configure key => value...
136abbreviated IDs, resolves all of them and returns a resolved node
137reference.
138 179
139In addition to C<address:port> pairs allowed in resolved noderefs, the 180Before a node can talk to other nodes on the network (i.e. enter
140following forms are supported: 181"distributed mode") it has to configure itself - the minimum a node needs
182to know is its own name, and optionally it should know the addresses of
183some other nodes in the network to discover other nodes.
184
185This function configures a node - it must be called exactly once (or
186never) before calling other AnyEvent::MP functions.
141 187
142=over 4 188=over 4
143 189
144=item the empty string 190=item step 1, gathering configuration from profiles
145 191
146An empty-string component gets resolved as if the default port (4040) was 192The function first looks up a profile in the aemp configuration (see the
147specified. 193L<aemp> commandline utility). The profile name can be specified via the
194named C<profile> parameter or can simply be the first parameter). If it is
195missing, then the nodename (F<uname -n>) will be used as profile name.
148 196
149=item naked port numbers (e.g. C<1234>) 197The profile data is then gathered as follows:
150 198
151These are resolved by prepending the local nodename and a colon, to be 199First, all remaining key => value pairs (all of which are conveniently
152further resolved. 200undocumented at the moment) will be interpreted as configuration
201data. Then they will be overwritten by any values specified in the global
202default configuration (see the F<aemp> utility), then the chain of
203profiles chosen by the profile name (and any C<parent> attributes).
153 204
154=item hostnames (e.g. C<localhost:1234>, C<localhost>) 205That means that the values specified in the profile have highest priority
206and the values specified directly via C<configure> have lowest priority,
207and can only be used to specify defaults.
155 208
156These are resolved by using AnyEvent::DNS to resolve them, optionally 209If the profile specifies a node ID, then this will become the node ID of
157looking up SRV records for the C<aemp=4040> port, if no port was 210this process. If not, then the profile name will be used as node ID. The
158specified. 211special node ID of C<anon/> will be replaced by a random node ID.
212
213=item step 2, bind listener sockets
214
215The next step is to look up the binds in the profile, followed by binding
216aemp protocol listeners on all binds specified (it is possible and valid
217to have no binds, meaning that the node cannot be contacted form the
218outside. This means the node cannot talk to other nodes that also have no
219binds, but it can still talk to all "normal" nodes).
220
221If the profile does not specify a binds list, then a default of C<*> is
222used, meaning the node will bind on a dynamically-assigned port on every
223local IP address it finds.
224
225=item step 3, connect to seed nodes
226
227As the last step, the seeds list from the profile is passed to the
228L<AnyEvent::MP::Global> module, which will then use it to keep
229connectivity with at least one node at any point in time.
159 230
160=back 231=back
232
233Example: become a distributed node using the locla node name as profile.
234This should be the most common form of invocation for "daemon"-type nodes.
235
236 configure
237
238Example: become an anonymous node. This form is often used for commandline
239clients.
240
241 configure nodeid => "anon/";
242
243Example: configure a node using a profile called seed, which si suitable
244for a seed node as it binds on all local addresses on a fixed port (4040,
245customary for aemp).
246
247 # use the aemp commandline utility
248 # aemp profile seed nodeid anon/ binds '*:4040'
249
250 # then use it
251 configure profile => "seed";
252
253 # or simply use aemp from the shell again:
254 # aemp run profile seed
255
256 # or provide a nicer-to-remember nodeid
257 # aemp run profile seed nodeid "$(hostname)"
161 258
162=item $SELF 259=item $SELF
163 260
164Contains the current port id while executing C<rcv> callbacks or C<psub> 261Contains the current port id while executing C<rcv> callbacks or C<psub>
165blocks. 262blocks.
166 263
167=item SELF, %SELF, @SELF... 264=item *SELF, SELF, %SELF, @SELF...
168 265
169Due to some quirks in how perl exports variables, it is impossible to 266Due to some quirks in how perl exports variables, it is impossible to
170just export C<$SELF>, all the symbols called C<SELF> are exported by this 267just export C<$SELF>, all the symbols named C<SELF> are exported by this
171module, but only C<$SELF> is currently used. 268module, but only C<$SELF> is currently used.
172 269
173=item snd $portid, type => @data 270=item snd $port, type => @data
174 271
175=item snd $portid, @msg 272=item snd $port, @msg
176 273
177Send the given message to the given port ID, which can identify either 274Send the given message to the given port, which can identify either a
178a local or a remote port, and can be either a string or soemthignt hat 275local or a remote port, and must be a port ID.
179stringifies a sa port ID (such as a port object :).
180 276
181While the message can be about anything, it is highly recommended to use a 277While the message can be almost anything, it is highly recommended to
182string as first element (a portid, or some word that indicates a request 278use a string as first element (a port ID, or some word that indicates a
183type etc.). 279request type etc.) and to consist if only simple perl values (scalars,
280arrays, hashes) - if you think you need to pass an object, think again.
184 281
185The message data effectively becomes read-only after a call to this 282The message data logically becomes read-only after a call to this
186function: modifying any argument is not allowed and can cause many 283function: modifying any argument (or values referenced by them) is
187problems. 284forbidden, as there can be considerable time between the call to C<snd>
285and the time the message is actually being serialised - in fact, it might
286never be copied as within the same process it is simply handed to the
287receiving port.
188 288
189The type of data you can transfer depends on the transport protocol: when 289The type of data you can transfer depends on the transport protocol: when
190JSON is used, then only strings, numbers and arrays and hashes consisting 290JSON is used, then only strings, numbers and arrays and hashes consisting
191of those are allowed (no objects). When Storable is used, then anything 291of those are allowed (no objects). When Storable is used, then anything
192that Storable can serialise and deserialise is allowed, and for the local 292that Storable can serialise and deserialise is allowed, and for the local
193node, anything can be passed. 293node, anything can be passed. Best rely only on the common denominator of
294these.
194 295
195=item kil $portid[, @reason] 296=item $local_port = port
196 297
197Kill the specified port with the given C<@reason>. 298Create a new local port object and returns its port ID. Initially it has
299no callbacks set and will throw an error when it receives messages.
198 300
199If no C<@reason> is specified, then the port is killed "normally" (linked 301=item $local_port = port { my @msg = @_ }
200ports will not be kileld, or even notified).
201 302
202Otherwise, linked ports get killed with the same reason (second form of 303Creates a new local port, and returns its ID. Semantically the same as
203C<mon>, see below). 304creating a port and calling C<rcv $port, $callback> on it.
204 305
205Runtime errors while evaluating C<rcv> callbacks or inside C<psub> blocks 306The block will be called for every message received on the port, with the
206will be reported as reason C<< die => $@ >>. 307global variable C<$SELF> set to the port ID. Runtime errors will cause the
308port to be C<kil>ed. The message will be passed as-is, no extra argument
309(i.e. no port ID) will be passed to the callback.
207 310
208Transport/communication errors are reported as C<< transport_error => 311If you want to stop/destroy the port, simply C<kil> it:
209$message >>.
210 312
211=item $guard = mon $portid, $cb->(@reason) 313 my $port = port {
314 my @msg = @_;
315 ...
316 kil $SELF;
317 };
212 318
213=item $guard = mon $portid, $otherport 319=cut
214 320
215=item $guard = mon $portid, $otherport, @msg 321sub rcv($@);
216 322
323sub _kilme {
324 die "received message on port without callback";
325}
326
327sub port(;&) {
328 my $id = "$UNIQ." . $ID++;
329 my $port = "$NODE#$id";
330
331 rcv $port, shift || \&_kilme;
332
333 $port
334}
335
336=item rcv $local_port, $callback->(@msg)
337
338Replaces the default callback on the specified port. There is no way to
339remove the default callback: use C<sub { }> to disable it, or better
340C<kil> the port when it is no longer needed.
341
342The global C<$SELF> (exported by this module) contains C<$port> while
343executing the callback. Runtime errors during callback execution will
344result in the port being C<kil>ed.
345
346The default callback received all messages not matched by a more specific
347C<tag> match.
348
349=item rcv $local_port, tag => $callback->(@msg_without_tag), ...
350
351Register (or replace) callbacks to be called on messages starting with the
352given tag on the given port (and return the port), or unregister it (when
353C<$callback> is C<$undef> or missing). There can only be one callback
354registered for each tag.
355
356The original message will be passed to the callback, after the first
357element (the tag) has been removed. The callback will use the same
358environment as the default callback (see above).
359
360Example: create a port and bind receivers on it in one go.
361
362 my $port = rcv port,
363 msg1 => sub { ... },
364 msg2 => sub { ... },
365 ;
366
367Example: create a port, bind receivers and send it in a message elsewhere
368in one go:
369
370 snd $otherport, reply =>
371 rcv port,
372 msg1 => sub { ... },
373 ...
374 ;
375
376Example: temporarily register a rcv callback for a tag matching some port
377(e.g. for a rpc reply) and unregister it after a message was received.
378
379 rcv $port, $otherport => sub {
380 my @reply = @_;
381
382 rcv $SELF, $otherport;
383 };
384
385=cut
386
387sub rcv($@) {
388 my $port = shift;
389 my ($nodeid, $portid) = split /#/, $port, 2;
390
391 $NODE{$nodeid} == $NODE{""}
392 or Carp::croak "$port: rcv can only be called on local ports, caught";
393
394 while (@_) {
395 if (ref $_[0]) {
396 if (my $self = $PORT_DATA{$portid}) {
397 "AnyEvent::MP::Port" eq ref $self
398 or Carp::croak "$port: rcv can only be called on message matching ports, caught";
399
400 $self->[2] = shift;
401 } else {
402 my $cb = shift;
403 $PORT{$portid} = sub {
404 local $SELF = $port;
405 eval { &$cb }; _self_die if $@;
406 };
407 }
408 } elsif (defined $_[0]) {
409 my $self = $PORT_DATA{$portid} ||= do {
410 my $self = bless [$PORT{$port} || sub { }, { }, $port], "AnyEvent::MP::Port";
411
412 $PORT{$portid} = sub {
413 local $SELF = $port;
414
415 if (my $cb = $self->[1]{$_[0]}) {
416 shift;
417 eval { &$cb }; _self_die if $@;
418 } else {
419 &{ $self->[0] };
420 }
421 };
422
423 $self
424 };
425
426 "AnyEvent::MP::Port" eq ref $self
427 or Carp::croak "$port: rcv can only be called on message matching ports, caught";
428
429 my ($tag, $cb) = splice @_, 0, 2;
430
431 if (defined $cb) {
432 $self->[1]{$tag} = $cb;
433 } else {
434 delete $self->[1]{$tag};
435 }
436 }
437 }
438
439 $port
440}
441
442=item $closure = psub { BLOCK }
443
444Remembers C<$SELF> and creates a closure out of the BLOCK. When the
445closure is executed, sets up the environment in the same way as in C<rcv>
446callbacks, i.e. runtime errors will cause the port to get C<kil>ed.
447
448This is useful when you register callbacks from C<rcv> callbacks:
449
450 rcv delayed_reply => sub {
451 my ($delay, @reply) = @_;
452 my $timer = AE::timer $delay, 0, psub {
453 snd @reply, $SELF;
454 };
455 };
456
457=cut
458
459sub psub(&) {
460 my $cb = shift;
461
462 my $port = $SELF
463 or Carp::croak "psub can only be called from within rcv or psub callbacks, not";
464
465 sub {
466 local $SELF = $port;
467
468 if (wantarray) {
469 my @res = eval { &$cb };
470 _self_die if $@;
471 @res
472 } else {
473 my $res = eval { &$cb };
474 _self_die if $@;
475 $res
476 }
477 }
478}
479
480=item $guard = mon $port, $cb->(@reason) # call $cb when $port dies
481
482=item $guard = mon $port, $rcvport # kill $rcvport when $port dies
483
484=item $guard = mon $port # kill $SELF when $port dies
485
486=item $guard = mon $port, $rcvport, @msg # send a message when $port dies
487
217Monitor the given port and do something when the port is killed. 488Monitor the given port and do something when the port is killed or
489messages to it were lost, and optionally return a guard that can be used
490to stop monitoring again.
218 491
219In the first form, the callback is simply called with any number 492In the first form (callback), the callback is simply called with any
220of C<@reason> elements (no @reason means that the port was deleted 493number of C<@reason> elements (no @reason means that the port was deleted
221"normally"). Note also that I<< the callback B<must> never die >>, so use 494"normally"). Note also that I<< the callback B<must> never die >>, so use
222C<eval> if unsure. 495C<eval> if unsure.
223 496
224In the second form, the other port will be C<kil>'ed with C<@reason>, iff 497In the second form (another port given), the other port (C<$rcvport>)
225a @reason was specified, i.e. on "normal" kils nothing happens, while 498will be C<kil>'ed with C<@reason>, if a @reason was specified, i.e. on
226under all other conditions, the other port is killed with the same reason. 499"normal" kils nothing happens, while under all other conditions, the other
500port is killed with the same reason.
227 501
502The third form (kill self) is the same as the second form, except that
503C<$rvport> defaults to C<$SELF>.
504
228In the last form, a message of the form C<@msg, @reason> will be C<snd>. 505In the last form (message), a message of the form C<@msg, @reason> will be
506C<snd>.
507
508Monitoring-actions are one-shot: once messages are lost (and a monitoring
509alert was raised), they are removed and will not trigger again.
510
511As a rule of thumb, monitoring requests should always monitor a port from
512a local port (or callback). The reason is that kill messages might get
513lost, just like any other message. Another less obvious reason is that
514even monitoring requests can get lost (for example, when the connection
515to the other node goes down permanently). When monitoring a port locally
516these problems do not exist.
517
518C<mon> effectively guarantees that, in the absence of hardware failures,
519after starting the monitor, either all messages sent to the port will
520arrive, or the monitoring action will be invoked after possible message
521loss has been detected. No messages will be lost "in between" (after
522the first lost message no further messages will be received by the
523port). After the monitoring action was invoked, further messages might get
524delivered again.
525
526Inter-host-connection timeouts and monitoring depend on the transport
527used. The only transport currently implemented is TCP, and AnyEvent::MP
528relies on TCP to detect node-downs (this can take 10-15 minutes on a
529non-idle connection, and usually around two hours for idle conenctions).
530
531This means that monitoring is good for program errors and cleaning up
532stuff eventually, but they are no replacement for a timeout when you need
533to ensure some maximum latency.
229 534
230Example: call a given callback when C<$port> is killed. 535Example: call a given callback when C<$port> is killed.
231 536
232 mon $port, sub { warn "port died because of <@_>\n" }; 537 mon $port, sub { warn "port died because of <@_>\n" };
233 538
234Example: kill ourselves when C<$port> is killed abnormally. 539Example: kill ourselves when C<$port> is killed abnormally.
235 540
236 mon $port, $self; 541 mon $port;
237 542
238Example: send us a restart message another C<$port> is killed. 543Example: send us a restart message when another C<$port> is killed.
239 544
240 mon $port, $self => "restart"; 545 mon $port, $self => "restart";
241 546
242=cut 547=cut
243 548
244sub mon { 549sub mon {
245 my ($noderef, $port, $cb) = ((split /#/, shift, 2), shift); 550 my ($nodeid, $port) = split /#/, shift, 2;
246 551
247 my $node = $NODE{$noderef} || add_node $noderef; 552 my $node = $NODE{$nodeid} || add_node $nodeid;
248 553
249 #TODO: ports must not be references 554 my $cb = @_ ? shift : $SELF || Carp::croak 'mon: called with one argument only, but $SELF not set,';
250 if (!ref $cb or "AnyEvent::MP::Port" eq ref $cb) { 555
556 unless (ref $cb) {
251 if (@_) { 557 if (@_) {
252 # send a kill info message 558 # send a kill info message
253 my (@msg) = ($cb, @_); 559 my (@msg) = ($cb, @_);
254 $cb = sub { snd @msg, @_ }; 560 $cb = sub { snd @msg, @_ };
255 } else { 561 } else {
271is killed, the references will be freed. 577is killed, the references will be freed.
272 578
273Optionally returns a guard that will stop the monitoring. 579Optionally returns a guard that will stop the monitoring.
274 580
275This function is useful when you create e.g. timers or other watchers and 581This function is useful when you create e.g. timers or other watchers and
276want to free them when the port gets killed: 582want to free them when the port gets killed (note the use of C<psub>):
277 583
278 $port->rcv (start => sub { 584 $port->rcv (start => sub {
279 my $timer; $timer = mon_guard $port, AE::timer 1, 1, sub { 585 my $timer; $timer = mon_guard $port, AE::timer 1, 1, psub {
280 undef $timer if 0.9 < rand; 586 undef $timer if 0.9 < rand;
281 }); 587 });
282 }); 588 });
283 589
284=cut 590=cut
285 591
286sub mon_guard { 592sub mon_guard {
287 my ($port, @refs) = @_; 593 my ($port, @refs) = @_;
288 594
595 #TODO: mon-less form?
596
289 mon $port, sub { 0 && @refs } 597 mon $port, sub { 0 && @refs }
290} 598}
291 599
292=item lnk $port1, $port2 600=item kil $port[, @reason]
293 601
294Link two ports. This is simply a shorthand for: 602Kill the specified port with the given C<@reason>.
295 603
296 mon $port1, $port2; 604If no C<@reason> is specified, then the port is killed "normally" (ports
297 mon $port2, $port1; 605monitoring other ports will not necessarily die because a port dies
606"normally").
298 607
299It means that if either one is killed abnormally, the other one gets 608Otherwise, linked ports get killed with the same reason (second form of
300killed as well. 609C<mon>, see above).
301 610
302=item $local_port = port 611Runtime errors while evaluating C<rcv> callbacks or inside C<psub> blocks
612will be reported as reason C<< die => $@ >>.
303 613
304Create a new local port object that supports message matching. 614Transport/communication errors are reported as C<< transport_error =>
615$message >>.
305 616
306=item $portid = port { my @msg = @_; $finished }
307
308Creates a "mini port", that is, a very lightweight port without any
309pattern matching behind it, and returns its ID.
310
311The block will be called for every message received on the port. When the
312callback returns a true value its job is considered "done" and the port
313will be destroyed. Otherwise it will stay alive.
314
315The message will be passed as-is, no extra argument (i.e. no port id) will
316be passed to the callback.
317
318If you need the local port id in the callback, this works nicely:
319
320 my $port; $port = miniport {
321 snd $otherport, reply => $port;
322 };
323
324=cut 617=cut
325 618
326sub port(;&) { 619=item $port = spawn $node, $initfunc[, @initdata]
327 my $id = "$UNIQ." . $ID++;
328 my $port = "$NODE#$id";
329 620
330 if (@_) { 621Creates a port on the node C<$node> (which can also be a port ID, in which
331 my $cb = shift; 622case it's the node where that port resides).
332 $PORT{$id} = sub { 623
333 local $SELF = $port; 624The port ID of the newly created port is returned immediately, and it is
334 eval { 625possible to immediately start sending messages or to monitor the port.
335 &$cb 626
336 and kil $id; 627After the port has been created, the init function is called on the remote
628node, in the same context as a C<rcv> callback. This function must be a
629fully-qualified function name (e.g. C<MyApp::Chat::Server::init>). To
630specify a function in the main program, use C<::name>.
631
632If the function doesn't exist, then the node tries to C<require>
633the package, then the package above the package and so on (e.g.
634C<MyApp::Chat::Server>, C<MyApp::Chat>, C<MyApp>) until the function
635exists or it runs out of package names.
636
637The init function is then called with the newly-created port as context
638object (C<$SELF>) and the C<@initdata> values as arguments. It I<must>
639call one of the C<rcv> functions to set callbacks on C<$SELF>, otherwise
640the port might not get created.
641
642A common idiom is to pass a local port, immediately monitor the spawned
643port, and in the remote init function, immediately monitor the passed
644local port. This two-way monitoring ensures that both ports get cleaned up
645when there is a problem.
646
647C<spawn> guarantees that the C<$initfunc> has no visible effects on the
648caller before C<spawn> returns (by delaying invocation when spawn is
649called for the local node).
650
651Example: spawn a chat server port on C<$othernode>.
652
653 # this node, executed from within a port context:
654 my $server = spawn $othernode, "MyApp::Chat::Server::connect", $SELF;
655 mon $server;
656
657 # init function on C<$othernode>
658 sub connect {
659 my ($srcport) = @_;
660
661 mon $srcport;
662
663 rcv $SELF, sub {
337 }; 664 ...
338 _self_die if $@;
339 };
340 } else {
341 my $self = bless {
342 id => "$NODE#$id",
343 }, "AnyEvent::MP::Port";
344
345 $PORT_DATA{$id} = $self;
346 $PORT{$id} = sub {
347 local $SELF = $port;
348
349 eval {
350 for (@{ $self->{rc0}{$_[0]} }) {
351 $_ && &{$_->[0]}
352 && undef $_;
353 }
354
355 for (@{ $self->{rcv}{$_[0]} }) {
356 $_ && [@_[1 .. @{$_->[1]}]] ~~ $_->[1]
357 && &{$_->[0]}
358 && undef $_;
359 }
360
361 for (@{ $self->{any} }) {
362 $_ && [@_[0 .. $#{$_->[1]}]] ~~ $_->[1]
363 && &{$_->[0]}
364 && undef $_;
365 }
366 };
367 _self_die if $@;
368 }; 665 };
369 } 666 }
370 667
371 $port
372}
373
374=item reg $portid, $name
375
376Registers the given port under the name C<$name>. If the name already
377exists it is replaced.
378
379A port can only be registered under one well known name.
380
381A port automatically becomes unregistered when it is killed.
382
383=cut 668=cut
384 669
385sub reg(@) { 670sub _spawn {
386 my ($portid, $name) = @_; 671 my $port = shift;
672 my $init = shift;
387 673
388 $REG{$name} = $portid; 674 # rcv will create the actual port
389} 675 local $SELF = "$NODE#$port";
390 676 eval {
391=item rcv $portid, tagstring => $callback->(@msg), ... 677 &{ load_func $init }
392
393=item rcv $portid, $smartmatch => $callback->(@msg), ...
394
395=item rcv $portid, [$smartmatch...] => $callback->(@msg), ...
396
397Register callbacks to be called on matching messages on the given port.
398
399The callback has to return a true value when its work is done, after
400which is will be removed, or a false value in which case it will stay
401registered.
402
403The global C<$SELF> (exported by this module) contains C<$portid> while
404executing the callback.
405
406Runtime errors wdurign callback execution will result in the port being
407C<kil>ed.
408
409If the match is an array reference, then it will be matched against the
410first elements of the message, otherwise only the first element is being
411matched.
412
413Any element in the match that is specified as C<_any_> (a function
414exported by this module) matches any single element of the message.
415
416While not required, it is highly recommended that the first matching
417element is a string identifying the message. The one-string-only match is
418also the most efficient match (by far).
419
420=cut
421
422sub rcv($@) {
423 my ($noderef, $port) = split /#/, shift, 2;
424
425 ($NODE{$noderef} || add_node $noderef) == $NODE{""}
426 or Carp::croak "$noderef#$port: rcv can only be called on local ports, caught";
427
428 my $self = $PORT_DATA{$port}
429 or Carp::croak "$noderef#$port: rcv can only be called on message matching ports, caught";
430
431 "AnyEvent::MP::Port" eq ref $self
432 or Carp::croak "$noderef#$port: rcv can only be called on message matching ports, caught";
433
434 while (@_) {
435 my ($match, $cb) = splice @_, 0, 2;
436
437 if (!ref $match) {
438 push @{ $self->{rc0}{$match} }, [$cb];
439 } elsif (("ARRAY" eq ref $match && !ref $match->[0])) {
440 my ($type, @match) = @$match;
441 @match
442 ? push @{ $self->{rcv}{$match->[0]} }, [$cb, \@match]
443 : push @{ $self->{rc0}{$match->[0]} }, [$cb];
444 } else {
445 push @{ $self->{any} }, [$cb, $match];
446 }
447 }
448}
449
450=item $closure = psub { BLOCK }
451
452Remembers C<$SELF> and creates a closure out of the BLOCK. When the
453closure is executed, sets up the environment in the same way as in C<rcv>
454callbacks, i.e. runtime errors will cause the port to get C<kil>ed.
455
456This is useful when you register callbacks from C<rcv> callbacks:
457
458 rcv delayed_reply => sub {
459 my ($delay, @reply) = @_;
460 my $timer = AE::timer $delay, 0, psub {
461 snd @reply, $SELF;
462 };
463 }; 678 };
464
465=cut
466
467sub psub(&) {
468 my $cb = shift;
469
470 my $port = $SELF
471 or Carp::croak "psub can only be called from within rcv or psub callbacks, not";
472
473 sub {
474 local $SELF = $port;
475
476 if (wantarray) {
477 my @res = eval { &$cb };
478 _self_die if $@; 679 _self_die if $@;
479 @res 680}
480 } else { 681
481 my $res = eval { &$cb }; 682sub spawn(@) {
482 _self_die if $@; 683 my ($nodeid, undef) = split /#/, shift, 2;
483 $res 684
484 } 685 my $id = "$RUNIQ." . $ID++;
686
687 $_[0] =~ /::/
688 or Carp::croak "spawn init function must be a fully-qualified name, caught";
689
690 snd_to_func $nodeid, "AnyEvent::MP::_spawn" => $id, @_;
691
692 "$nodeid#$id"
693}
694
695=item after $timeout, @msg
696
697=item after $timeout, $callback
698
699Either sends the given message, or call the given callback, after the
700specified number of seconds.
701
702This is simply a utility function that comes in handy at times - the
703AnyEvent::MP author is not convinced of the wisdom of having it, though,
704so it may go away in the future.
705
706=cut
707
708sub after($@) {
709 my ($timeout, @action) = @_;
710
711 my $t; $t = AE::timer $timeout, 0, sub {
712 undef $t;
713 ref $action[0]
714 ? $action[0]()
715 : snd @action;
485 } 716 };
486} 717}
487 718
488=back 719=back
489 720
490=head1 FUNCTIONS FOR NODES 721=head1 AnyEvent::MP vs. Distributed Erlang
722
723AnyEvent::MP got lots of its ideas from distributed Erlang (Erlang node
724== aemp node, Erlang process == aemp port), so many of the documents and
725programming techniques employed by Erlang apply to AnyEvent::MP. Here is a
726sample:
727
728 http://www.Erlang.se/doc/programming_rules.shtml
729 http://Erlang.org/doc/getting_started/part_frame.html # chapters 3 and 4
730 http://Erlang.org/download/Erlang-book-part1.pdf # chapters 5 and 6
731 http://Erlang.org/download/armstrong_thesis_2003.pdf # chapters 4 and 5
732
733Despite the similarities, there are also some important differences:
491 734
492=over 4 735=over 4
493 736
494=item become_public $noderef 737=item * Node IDs are arbitrary strings in AEMP.
495 738
496Tells the node to become a public node, i.e. reachable from other nodes.
497
498The first argument is the (unresolved) node reference of the local node
499(if missing then the empty string is used).
500
501It is quite common to not specify anything, in which case the local node
502tries to listen on the default port, or to only specify a port number, in
503which case AnyEvent::MP tries to guess the local addresses.
504
505=cut
506
507=back
508
509=head1 NODE MESSAGES
510
511Nodes understand the following messages sent to them. Many of them take
512arguments called C<@reply>, which will simply be used to compose a reply
513message - C<$reply[0]> is the port to reply to, C<$reply[1]> the type and
514the remaining arguments are simply the message data.
515
516While other messages exist, they are not public and subject to change.
517
518=over 4
519
520=cut
521
522=item lookup => $name, @reply
523
524Replies with the port ID of the specified well-known port, or C<undef>.
525
526=item devnull => ...
527
528Generic data sink/CPU heat conversion.
529
530=item relay => $port, @msg
531
532Simply forwards the message to the given port.
533
534=item eval => $string[ @reply]
535
536Evaluates the given string. If C<@reply> is given, then a message of the
537form C<@reply, $@, @evalres> is sent.
538
539Example: crash another node.
540
541 snd $othernode, eval => "exit";
542
543=item time => @reply
544
545Replies the the current node time to C<@reply>.
546
547Example: tell the current node to send the current time to C<$myport> in a
548C<timereply> message.
549
550 snd $NODE, time => $myport, timereply => 1, 2;
551 # => snd $myport, timereply => 1, 2, <time>
552
553=back
554
555=head1 AnyEvent::MP vs. Distributed Erlang
556
557AnyEvent::MP got lots of its ideas from distributed erlang (erlang node
558== aemp node, erlang process == aemp port), so many of the documents and
559programming techniques employed by erlang apply to AnyEvent::MP. Here is a
560sample:
561
562 http://www.erlang.se/doc/programming_rules.shtml
563 http://erlang.org/doc/getting_started/part_frame.html # chapters 3 and 4
564 http://erlang.org/download/erlang-book-part1.pdf # chapters 5 and 6
565 http://erlang.org/download/armstrong_thesis_2003.pdf # chapters 4 and 5
566
567Despite the similarities, there are also some important differences:
568
569=over 4
570
571=item * Node references contain the recipe on how to contact them.
572
573Erlang relies on special naming and DNS to work everywhere in the 739Erlang relies on special naming and DNS to work everywhere in the same
574same way. AEMP relies on each node knowing it's own address(es), with 740way. AEMP relies on each node somehow knowing its own address(es) (e.g. by
575convenience functionality. 741configuration or DNS), but will otherwise discover other odes itself.
576 742
577This means that AEMP requires a less tightly controlled environment at the 743=item * Erlang has a "remote ports are like local ports" philosophy, AEMP
578cost of longer node references and a slightly higher management overhead. 744uses "local ports are like remote ports".
745
746The failure modes for local ports are quite different (runtime errors
747only) then for remote ports - when a local port dies, you I<know> it dies,
748when a connection to another node dies, you know nothing about the other
749port.
750
751Erlang pretends remote ports are as reliable as local ports, even when
752they are not.
753
754AEMP encourages a "treat remote ports differently" philosophy, with local
755ports being the special case/exception, where transport errors cannot
756occur.
579 757
580=item * Erlang uses processes and a mailbox, AEMP does not queue. 758=item * Erlang uses processes and a mailbox, AEMP does not queue.
581 759
582Erlang uses processes that selctively receive messages, and therefore 760Erlang uses processes that selectively receive messages, and therefore
583needs a queue. AEMP is event based, queuing messages would serve no useful 761needs a queue. AEMP is event based, queuing messages would serve no
584purpose. 762useful purpose. For the same reason the pattern-matching abilities of
763AnyEvent::MP are more limited, as there is little need to be able to
764filter messages without dequeuing them.
585 765
586(But see L<Coro::MP> for a more erlang-like process model on top of AEMP). 766(But see L<Coro::MP> for a more Erlang-like process model on top of AEMP).
587 767
588=item * Erlang sends are synchronous, AEMP sends are asynchronous. 768=item * Erlang sends are synchronous, AEMP sends are asynchronous.
589 769
590Sending messages in erlang is synchronous and blocks the process. AEMP 770Sending messages in Erlang is synchronous and blocks the process (and
591sends are immediate, connection establishment is handled in the 771so does not need a queue that can overflow). AEMP sends are immediate,
592background. 772connection establishment is handled in the background.
593 773
594=item * Erlang can silently lose messages, AEMP cannot. 774=item * Erlang suffers from silent message loss, AEMP does not.
595 775
596Erlang makes few guarantees on messages delivery - messages can get lost 776Erlang makes few guarantees on messages delivery - messages can get lost
597without any of the processes realising it (i.e. you send messages a, b, 777without any of the processes realising it (i.e. you send messages a, b,
598and c, and the other side only receives messages a and c). 778and c, and the other side only receives messages a and c).
599 779
600AEMP guarantees correct ordering, and the guarantee that there are no 780AEMP guarantees correct ordering, and the guarantee that after one message
601holes in the message sequence. 781is lost, all following ones sent to the same port are lost as well, until
602 782monitoring raises an error, so there are no silent "holes" in the message
603=item * In erlang, processes can be declared dead and later be found to be 783sequence.
604alive.
605
606In erlang it can happen that a monitored process is declared dead and
607linked processes get killed, but later it turns out that the process is
608still alive - and can receive messages.
609
610In AEMP, when port monitoring detects a port as dead, then that port will
611eventually be killed - it cannot happen that a node detects a port as dead
612and then later sends messages to it, finding it is still alive.
613 784
614=item * Erlang can send messages to the wrong port, AEMP does not. 785=item * Erlang can send messages to the wrong port, AEMP does not.
615 786
616In erlang it is quite possible that a node that restarts reuses a process 787In Erlang it is quite likely that a node that restarts reuses a process ID
617ID known to other nodes for a completely different process, causing 788known to other nodes for a completely different process, causing messages
618messages destined for that process to end up in an unrelated process. 789destined for that process to end up in an unrelated process.
619 790
620AEMP never reuses port IDs, so old messages or old port IDs floating 791AEMP never reuses port IDs, so old messages or old port IDs floating
621around in the network will not be sent to an unrelated port. 792around in the network will not be sent to an unrelated port.
622 793
623=item * Erlang uses unprotected connections, AEMP uses secure 794=item * Erlang uses unprotected connections, AEMP uses secure
624authentication and can use TLS. 795authentication and can use TLS.
625 796
626AEMP can use a proven protocol - SSL/TLS - to protect connections and 797AEMP can use a proven protocol - TLS - to protect connections and
627securely authenticate nodes. 798securely authenticate nodes.
628 799
629=item * The AEMP protocol is optimised for both text-based and binary 800=item * The AEMP protocol is optimised for both text-based and binary
630communications. 801communications.
631 802
632The AEMP protocol, unlike the erlang protocol, supports both 803The AEMP protocol, unlike the Erlang protocol, supports both programming
633language-independent text-only protocols (good for debugging) and binary, 804language independent text-only protocols (good for debugging) and binary,
634language-specific serialisers (e.g. Storable). 805language-specific serialisers (e.g. Storable). By default, unless TLS is
806used, the protocol is actually completely text-based.
635 807
636It has also been carefully designed to be implementable in other languages 808It has also been carefully designed to be implementable in other languages
637with a minimum of work while gracefully degrading fucntionality to make the 809with a minimum of work while gracefully degrading functionality to make the
638protocol simple. 810protocol simple.
639 811
812=item * AEMP has more flexible monitoring options than Erlang.
813
814In Erlang, you can chose to receive I<all> exit signals as messages
815or I<none>, there is no in-between, so monitoring single processes is
816difficult to implement. Monitoring in AEMP is more flexible than in
817Erlang, as one can choose between automatic kill, exit message or callback
818on a per-process basis.
819
820=item * Erlang tries to hide remote/local connections, AEMP does not.
821
822Monitoring in Erlang is not an indicator of process death/crashes, in the
823same way as linking is (except linking is unreliable in Erlang).
824
825In AEMP, you don't "look up" registered port names or send to named ports
826that might or might not be persistent. Instead, you normally spawn a port
827on the remote node. The init function monitors you, and you monitor the
828remote port. Since both monitors are local to the node, they are much more
829reliable (no need for C<spawn_link>).
830
831This also saves round-trips and avoids sending messages to the wrong port
832(hard to do in Erlang).
833
640=back 834=back
641 835
836=head1 RATIONALE
837
838=over 4
839
840=item Why strings for port and node IDs, why not objects?
841
842We considered "objects", but found that the actual number of methods
843that can be called are quite low. Since port and node IDs travel over
844the network frequently, the serialising/deserialising would add lots of
845overhead, as well as having to keep a proxy object everywhere.
846
847Strings can easily be printed, easily serialised etc. and need no special
848procedures to be "valid".
849
850And as a result, a miniport consists of a single closure stored in a
851global hash - it can't become much cheaper.
852
853=item Why favour JSON, why not a real serialising format such as Storable?
854
855In fact, any AnyEvent::MP node will happily accept Storable as framing
856format, but currently there is no way to make a node use Storable by
857default (although all nodes will accept it).
858
859The default framing protocol is JSON because a) JSON::XS is many times
860faster for small messages and b) most importantly, after years of
861experience we found that object serialisation is causing more problems
862than it solves: Just like function calls, objects simply do not travel
863easily over the network, mostly because they will always be a copy, so you
864always have to re-think your design.
865
866Keeping your messages simple, concentrating on data structures rather than
867objects, will keep your messages clean, tidy and efficient.
868
869=back
870
642=head1 SEE ALSO 871=head1 SEE ALSO
872
873L<AnyEvent::MP::Intro> - a gentle introduction.
874
875L<AnyEvent::MP::Kernel> - more, lower-level, stuff.
876
877L<AnyEvent::MP::Global> - network maintainance and port groups, to find
878your applications.
879
880L<AnyEvent::MP::LogCatcher> - simple service to display log messages from
881all nodes.
643 882
644L<AnyEvent>. 883L<AnyEvent>.
645 884
646=head1 AUTHOR 885=head1 AUTHOR
647 886

Diff Legend

Removed lines
+ Added lines
< Changed lines
> Changed lines