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 1.33 by root, Wed Aug 5 22:40:51 2009 UTC vs.
Revision 1.37 by root, Fri Aug 7 16:47:23 2009 UTC

22 snd $port2, ping => $port1; 22 snd $port2, ping => $port1;
23 23
24 # more, smarter, matches (_any_ is exported by this module) 24 # more, smarter, matches (_any_ is exported by this module)
25 rcv $port, [child_died => $pid] => sub { ... 25 rcv $port, [child_died => $pid] => sub { ...
26 rcv $port, [_any_, _any_, 3] => sub { .. $_[2] is 3 26 rcv $port, [_any_, _any_, 3] => sub { .. $_[2] is 3
27
28 # monitoring
29 mon $port, $cb->(@msg) # callback is invoked on death
30 mon $port, $otherport # kill otherport on abnormal death
31 mon $port, $otherport, @msg # send message on death
27 32
28=head1 DESCRIPTION 33=head1 DESCRIPTION
29 34
30This module (-family) implements a simple message passing framework. 35This module (-family) implements a simple message passing framework.
31 36
127 132
128=item $noderef = node_of $port 133=item $noderef = node_of $port
129 134
130Extracts and returns the noderef from a portid or a noderef. 135Extracts and returns the noderef from a portid or a noderef.
131 136
137=item initialise_node $noderef, $seednode, $seednode...
138
139=item initialise_node "slave/", $master, $master...
140
141Before a node can talk to other nodes on the network it has to initialise
142itself - the minimum a node needs to know is it's own name, and optionally
143it should know the noderefs of some other nodes in the network.
144
145This function initialises a node - it must be called exactly once (or
146never) before calling other AnyEvent::MP functions.
147
148All arguments are noderefs, which can be either resolved or unresolved.
149
150There are two types of networked nodes, public nodes and slave nodes:
151
152=over 4
153
154=item public nodes
155
156For public nodes, C<$noderef> must either be a (possibly unresolved)
157noderef, in which case it will be resolved, or C<undef> (or missing), in
158which case the noderef will be guessed.
159
160Afterwards, the node will bind itself on all endpoints and try to connect
161to all additional C<$seednodes> that are specified. Seednodes are optional
162and can be used to quickly bootstrap the node into an existing network.
163
164=item slave nodes
165
166When the C<$noderef> is the special string C<slave/>, then the node will
167become a slave node. Slave nodes cannot be contacted from outside and will
168route most of their traffic to the master node that they attach to.
169
170At least one additional noderef is required: The node will try to connect
171to all of them and will become a slave attached to the first node it can
172successfully connect to.
173
174=back
175
176This function will block until all nodes have been resolved and, for slave
177nodes, until it has successfully established a connection to a master
178server.
179
180Example: become a public node listening on the default node.
181
182 initialise_node;
183
184Example: become a public node, and try to contact some well-known master
185servers to become part of the network.
186
187 initialise_node undef, "master1", "master2";
188
189Example: become a public node listening on port C<4041>.
190
191 initialise_node 4041;
192
193Example: become a public node, only visible on localhost port 4044.
194
195 initialise_node "locahost:4044";
196
197Example: become a slave node to any of the specified master servers.
198
199 initialise_node "slave/", "master1", "192.168.13.17", "mp.example.net";
200
132=item $cv = resolve_node $noderef 201=item $cv = resolve_node $noderef
133 202
134Takes an unresolved node reference that may contain hostnames and 203Takes an unresolved node reference that may contain hostnames and
135abbreviated IDs, resolves all of them and returns a resolved node 204abbreviated IDs, resolves all of them and returns a resolved node
136reference. 205reference.
233 $port 302 $port
234} 303}
235 304
236=item reg $port, $name 305=item reg $port, $name
237 306
238Registers the given port under the name C<$name>. If the name already 307=item reg $name
239exists it is replaced. 308
309Registers the given port (or C<$SELF><<< if missing) under the name
310C<$name>. If the name already exists it is replaced.
240 311
241A port can only be registered under one well known name. 312A port can only be registered under one well known name.
242 313
243A port automatically becomes unregistered when it is killed. 314A port automatically becomes unregistered when it is killed.
244 315
245=cut 316=cut
246 317
247sub reg(@) { 318sub reg(@) {
248 my ($port, $name) = @_; 319 my $port = @_ > 1 ? shift : $SELF || Carp::croak 'reg: called with one argument only, but $SELF not set,';
249 320
250 $REG{$name} = $port; 321 $REG{$_[0]} = $port;
251} 322}
252 323
253=item rcv $port, $callback->(@msg) 324=item rcv $port, $callback->(@msg)
254 325
255Replaces the callback on the specified miniport (after converting it to 326Replaces the callback on the specified miniport (after converting it to
260=item rcv $port, $smartmatch => $callback->(@msg), ... 331=item rcv $port, $smartmatch => $callback->(@msg), ...
261 332
262=item rcv $port, [$smartmatch...] => $callback->(@msg), ... 333=item rcv $port, [$smartmatch...] => $callback->(@msg), ...
263 334
264Register callbacks to be called on matching messages on the given full 335Register callbacks to be called on matching messages on the given full
265port (after converting it to one if required). 336port (after converting it to one if required) and return the port.
266 337
267The callback has to return a true value when its work is done, after 338The callback has to return a true value when its work is done, after
268which is will be removed, or a false value in which case it will stay 339which is will be removed, or a false value in which case it will stay
269registered. 340registered.
270 341
282exported by this module) matches any single element of the message. 353exported by this module) matches any single element of the message.
283 354
284While not required, it is highly recommended that the first matching 355While not required, it is highly recommended that the first matching
285element is a string identifying the message. The one-string-only match is 356element is a string identifying the message. The one-string-only match is
286also the most efficient match (by far). 357also the most efficient match (by far).
358
359Example: create a port and bind receivers on it in one go.
360
361 my $port = rcv port,
362 msg1 => sub { ...; 0 },
363 msg2 => sub { ...; 0 },
364 ;
365
366Example: create a port, bind receivers and send it in a message elsewhere
367in one go:
368
369 snd $otherport, reply =>
370 rcv port,
371 msg1 => sub { ...; 0 },
372 ...
373 ;
287 374
288=cut 375=cut
289 376
290sub rcv($@) { 377sub rcv($@) {
291 my $port = shift; 378 my $port = shift;
398 } 485 }
399} 486}
400 487
401=item $guard = mon $port, $cb->(@reason) 488=item $guard = mon $port, $cb->(@reason)
402 489
403=item $guard = mon $port, $otherport 490=item $guard = mon $port, $rcvport
404 491
492=item $guard = mon $port
493
405=item $guard = mon $port, $otherport, @msg 494=item $guard = mon $port, $rcvport, @msg
406 495
407Monitor the given port and do something when the port is killed. 496Monitor the given port and do something when the port is killed, and
497optionally return a guard that can be used to stop monitoring again.
408 498
409In the first form, the callback is simply called with any number 499In the first form (callback), the callback is simply called with any
410of C<@reason> elements (no @reason means that the port was deleted 500number of C<@reason> elements (no @reason means that the port was deleted
411"normally"). Note also that I<< the callback B<must> never die >>, so use 501"normally"). Note also that I<< the callback B<must> never die >>, so use
412C<eval> if unsure. 502C<eval> if unsure.
413 503
414In the second form, the other port will be C<kil>'ed with C<@reason>, iff 504In the second form (another port given), the other port (C<$rcvport)
415a @reason was specified, i.e. on "normal" kils nothing happens, while 505will be C<kil>'ed with C<@reason>, iff a @reason was specified, i.e. on
416under all other conditions, the other port is killed with the same reason. 506"normal" kils nothing happens, while under all other conditions, the other
507port is killed with the same reason.
417 508
509The third form (kill self) is the same as the second form, except that
510C<$rvport> defaults to C<$SELF>.
511
418In the last form, a message of the form C<@msg, @reason> will be C<snd>. 512In the last form (message), a message of the form C<@msg, @reason> will be
513C<snd>.
514
515As a rule of thumb, monitoring requests should always monitor a port from
516a local port (or callback). The reason is that kill messages might get
517lost, just like any other message. Another less obvious reason is that
518even monitoring requests can get lost (for exmaple, when the connection
519to the other node goes down permanently). When monitoring a port locally
520these problems do not exist.
419 521
420Example: call a given callback when C<$port> is killed. 522Example: call a given callback when C<$port> is killed.
421 523
422 mon $port, sub { warn "port died because of <@_>\n" }; 524 mon $port, sub { warn "port died because of <@_>\n" };
423 525
424Example: kill ourselves when C<$port> is killed abnormally. 526Example: kill ourselves when C<$port> is killed abnormally.
425 527
426 mon $port, $self; 528 mon $port;
427 529
428Example: send us a restart message another C<$port> is killed. 530Example: send us a restart message when another C<$port> is killed.
429 531
430 mon $port, $self => "restart"; 532 mon $port, $self => "restart";
431 533
432=cut 534=cut
433 535
434sub mon { 536sub mon {
435 my ($noderef, $port) = split /#/, shift, 2; 537 my ($noderef, $port) = split /#/, shift, 2;
436 538
437 my $node = $NODE{$noderef} || add_node $noderef; 539 my $node = $NODE{$noderef} || add_node $noderef;
438 540
439 my $cb = shift; 541 my $cb = @_ ? $_[0] : $SELF || Carp::croak 'mon: called with one argument only, but $SELF not set,';
440 542
441 unless (ref $cb) { 543 unless (ref $cb) {
442 if (@_) { 544 if (@_) {
443 # send a kill info message 545 # send a kill info message
444 my (@msg) = ($cb, @_); 546 my (@msg) = @_;
445 $cb = sub { snd @msg, @_ }; 547 $cb = sub { snd @msg, @_ };
446 } else { 548 } else {
447 # simply kill other port 549 # simply kill other port
448 my $port = $cb; 550 my $port = $cb;
449 $cb = sub { kil $port, @_ if @_ }; 551 $cb = sub { kil $port, @_ if @_ };
475=cut 577=cut
476 578
477sub mon_guard { 579sub mon_guard {
478 my ($port, @refs) = @_; 580 my ($port, @refs) = @_;
479 581
582 #TODO: mon-less form?
583
480 mon $port, sub { 0 && @refs } 584 mon $port, sub { 0 && @refs }
481} 585}
482 586
483=item lnk $port1, $port2
484
485Link two ports. This is simply a shorthand for:
486
487 mon $port1, $port2;
488 mon $port2, $port1;
489
490It means that if either one is killed abnormally, the other one gets
491killed as well.
492
493=item kil $port[, @reason] 587=item kil $port[, @reason]
494 588
495Kill the specified port with the given C<@reason>. 589Kill the specified port with the given C<@reason>.
496 590
497If no C<@reason> is specified, then the port is killed "normally" (linked 591If no C<@reason> is specified, then the port is killed "normally" (linked
503Runtime errors while evaluating C<rcv> callbacks or inside C<psub> blocks 597Runtime errors while evaluating C<rcv> callbacks or inside C<psub> blocks
504will be reported as reason C<< die => $@ >>. 598will be reported as reason C<< die => $@ >>.
505 599
506Transport/communication errors are reported as C<< transport_error => 600Transport/communication errors are reported as C<< transport_error =>
507$message >>. 601$message >>.
508
509=back
510
511=head1 FUNCTIONS FOR NODES
512
513=over 4
514
515=item initialise_node $noderef, $seednode, $seednode...
516
517=item initialise_node "slave/", $master, $master...
518
519Initialises a node - must be called exactly once before calling other
520AnyEvent::MP functions when talking to other nodes is required.
521
522All arguments are noderefs, which can be either resolved or unresolved.
523
524There are two types of networked nodes, public nodes and slave nodes:
525
526=over 4
527
528=item public nodes
529
530For public nodes, C<$noderef> must either be a (possibly unresolved)
531noderef, in which case it will be resolved, or C<undef> (or missing), in
532which case the noderef will be guessed.
533
534Afterwards, the node will bind itself on all endpoints and try to connect
535to all additional C<$seednodes> that are specified. Seednodes are optional
536and can be used to quickly bootstrap the node into an existing network.
537
538=item slave nodes
539
540When the C<$noderef> is the special string C<slave/>, then the node will
541become a slave node. Slave nodes cannot be contacted from outside and will
542route most of their traffic to the master node that they attach to.
543
544At least one additional noderef is required: The node will try to connect
545to all of them and will become a slave attached to the first node it can
546successfully connect to.
547
548=back
549
550This function will block until all nodes have been resolved and, for slave
551nodes, until it has successfully established a connection to a master
552server.
553
554Example: become a public node listening on the default node.
555
556 initialise_node;
557
558Example: become a public node, and try to contact some well-known master
559servers to become part of the network.
560
561 initialise_node undef, "master1", "master2";
562
563Example: become a public node listening on port C<4041>.
564
565 initialise_node 4041;
566
567Example: become a public node, only visible on localhost port 4044.
568
569 initialise_node "locahost:4044";
570
571Example: become a slave node to any of the specified master servers.
572
573 initialise_node "slave/", "master1", "192.168.13.17", "mp.example.net";
574
575=cut
576 602
577=back 603=back
578 604
579=head1 NODE MESSAGES 605=head1 NODE MESSAGES
580 606
622 648
623=back 649=back
624 650
625=head1 AnyEvent::MP vs. Distributed Erlang 651=head1 AnyEvent::MP vs. Distributed Erlang
626 652
627AnyEvent::MP got lots of its ideas from distributed erlang (erlang node 653AnyEvent::MP got lots of its ideas from distributed Erlang (Erlang node
628== aemp node, erlang process == aemp port), so many of the documents and 654== aemp node, Erlang process == aemp port), so many of the documents and
629programming techniques employed by erlang apply to AnyEvent::MP. Here is a 655programming techniques employed by Erlang apply to AnyEvent::MP. Here is a
630sample: 656sample:
631 657
632 http://www.erlang.se/doc/programming_rules.shtml 658 http://www.Erlang.se/doc/programming_rules.shtml
633 http://erlang.org/doc/getting_started/part_frame.html # chapters 3 and 4 659 http://Erlang.org/doc/getting_started/part_frame.html # chapters 3 and 4
634 http://erlang.org/download/erlang-book-part1.pdf # chapters 5 and 6 660 http://Erlang.org/download/Erlang-book-part1.pdf # chapters 5 and 6
635 http://erlang.org/download/armstrong_thesis_2003.pdf # chapters 4 and 5 661 http://Erlang.org/download/armstrong_thesis_2003.pdf # chapters 4 and 5
636 662
637Despite the similarities, there are also some important differences: 663Despite the similarities, there are also some important differences:
638 664
639=over 4 665=over 4
640 666
651 677
652Erlang uses processes that selctively receive messages, and therefore 678Erlang uses processes that selctively receive messages, and therefore
653needs a queue. AEMP is event based, queuing messages would serve no useful 679needs a queue. AEMP is event based, queuing messages would serve no useful
654purpose. 680purpose.
655 681
656(But see L<Coro::MP> for a more erlang-like process model on top of AEMP). 682(But see L<Coro::MP> for a more Erlang-like process model on top of AEMP).
657 683
658=item * Erlang sends are synchronous, AEMP sends are asynchronous. 684=item * Erlang sends are synchronous, AEMP sends are asynchronous.
659 685
660Sending messages in erlang is synchronous and blocks the process. AEMP 686Sending messages in Erlang is synchronous and blocks the process. AEMP
661sends are immediate, connection establishment is handled in the 687sends are immediate, connection establishment is handled in the
662background. 688background.
663 689
664=item * Erlang can silently lose messages, AEMP cannot. 690=item * Erlang can silently lose messages, AEMP cannot.
665 691
668and c, and the other side only receives messages a and c). 694and c, and the other side only receives messages a and c).
669 695
670AEMP guarantees correct ordering, and the guarantee that there are no 696AEMP guarantees correct ordering, and the guarantee that there are no
671holes in the message sequence. 697holes in the message sequence.
672 698
673=item * In erlang, processes can be declared dead and later be found to be 699=item * In Erlang, processes can be declared dead and later be found to be
674alive. 700alive.
675 701
676In erlang it can happen that a monitored process is declared dead and 702In Erlang it can happen that a monitored process is declared dead and
677linked processes get killed, but later it turns out that the process is 703linked processes get killed, but later it turns out that the process is
678still alive - and can receive messages. 704still alive - and can receive messages.
679 705
680In AEMP, when port monitoring detects a port as dead, then that port will 706In AEMP, when port monitoring detects a port as dead, then that port will
681eventually be killed - it cannot happen that a node detects a port as dead 707eventually be killed - it cannot happen that a node detects a port as dead
682and then later sends messages to it, finding it is still alive. 708and then later sends messages to it, finding it is still alive.
683 709
684=item * Erlang can send messages to the wrong port, AEMP does not. 710=item * Erlang can send messages to the wrong port, AEMP does not.
685 711
686In erlang it is quite possible that a node that restarts reuses a process 712In Erlang it is quite possible that a node that restarts reuses a process
687ID known to other nodes for a completely different process, causing 713ID known to other nodes for a completely different process, causing
688messages destined for that process to end up in an unrelated process. 714messages destined for that process to end up in an unrelated process.
689 715
690AEMP never reuses port IDs, so old messages or old port IDs floating 716AEMP never reuses port IDs, so old messages or old port IDs floating
691around in the network will not be sent to an unrelated port. 717around in the network will not be sent to an unrelated port.
697securely authenticate nodes. 723securely authenticate nodes.
698 724
699=item * The AEMP protocol is optimised for both text-based and binary 725=item * The AEMP protocol is optimised for both text-based and binary
700communications. 726communications.
701 727
702The AEMP protocol, unlike the erlang protocol, supports both 728The AEMP protocol, unlike the Erlang protocol, supports both
703language-independent text-only protocols (good for debugging) and binary, 729language-independent text-only protocols (good for debugging) and binary,
704language-specific serialisers (e.g. Storable). 730language-specific serialisers (e.g. Storable).
705 731
706It has also been carefully designed to be implementable in other languages 732It has also been carefully designed to be implementable in other languages
707with a minimum of work while gracefully degrading fucntionality to make the 733with a minimum of work while gracefully degrading fucntionality to make the
708protocol simple. 734protocol simple.
709 735
736=item * AEMP has more flexible monitoring options than Erlang.
737
738In Erlang, you can chose to receive I<all> exit signals as messages
739or I<none>, there is no in-between, so monitoring single processes is
740difficult to implement. Monitoring in AEMP is more flexible than in
741Erlang, as one can choose between automatic kill, exit message or callback
742on a per-process basis.
743
744=item * Erlang tries to hide remote/local connections, AEMP does not.
745
746Monitoring in Erlang is not an indicator of process death/crashes,
747as linking is (except linking is unreliable in Erlang).
748
749In AEMP, you don't "look up" registered port names or send to named ports
750that might or might not be persistent. Instead, you normally spawn a port
751on the remote node. The init function monitors the you, and you monitor
752the remote port. Since both monitors are local to the node, they are much
753more reliable.
754
755This also saves round-trips and avoids sending messages to the wrong port
756(hard to do in Erlang).
757
710=back 758=back
711 759
712=head1 SEE ALSO 760=head1 SEE ALSO
713 761
714L<AnyEvent>. 762L<AnyEvent>.

Diff Legend

Removed lines
+ Added lines
< Changed lines
> Changed lines