Merge pull request #1887 from krasserm/wip-3785-change-default-processor-id-krasserm

!per #3785 Change default implementation of processorId

akka-docs/rst/java/persistence.rst

@@ -143,20 +143,17 @@ Identifiers
 -----------
 
 A processor must have an identifier that doesn't change across different actor incarnations. It defaults to the
-``String`` representation of processor's path and can be obtained via the ``processorId`` method.
+``String`` representation of processor's path without the address part and can be obtained via the ``processorId``
+method.
 
 .. includecode:: code/docs/persistence/PersistenceDocTest.java#processor-id
 
 Applications can customize a processor's id by specifying an actor name during processor creation as shown in
-section :ref:`processors-java`. This works well when using local actor references but may cause problems with remote
+section :ref:`processors-java`. This changes that processor's name in its actor hierarchy and hence influences only
-actor references because their paths also contain deployment information such as host and port (and actor deployments
+part of the processor id. To fully customize a processor's id, the ``processorId`` method should be overridden.
-are likely to change during the lifetime of an application). In this case, ``UntypedProcessor`` implementation classes
-should override ``processorId``.
 
 .. includecode:: code/docs/persistence/PersistenceDocTest.java#processor-id-override
 
-Later versions of Akka persistence will likely offer a possibility to migrate processor ids.
-
 .. _channels-java:
 
 Channels
@@ -299,11 +296,10 @@ Default is ``Resolve.off()`` which means no resolution. Find out more in the ``D
 Identifiers
 -----------
 
-In the same way as :ref:`processors-java`, channels also have an identifier that defaults to a channel's path. A channel
+In the same way as :ref:`processors`, channels also have an identifier that defaults to a channel's path. A channel
-identifier can therefore be customized by using a custom actor name at channel creation. As already mentioned, this
+identifier can therefore be customized by using a custom actor name at channel creation. This changes that channel's
-works well when using local actor references but may cause problems with remote actor references. In this case, an
+name in its actor hierarchy and hence influences only part of the channel identifier. To fully customize a channel
-application-defined channel id should be provided as argument to ``Channel.props(String)`` or
+identifier, it should be provided as argument ``Channel.props(String)`` or ``PersistentChannel.props(String)``.
-``PersistentChannel.props(String)``.
 
 .. includecode:: code/docs/persistence/PersistenceDocTest.java#channel-id-override
 

akka-docs/rst/scala/persistence.rst

@@ -138,20 +138,17 @@ Identifiers
 -----------
 
 A processor must have an identifier that doesn't change across different actor incarnations. It defaults to the
-``String`` representation of processor's path and can be obtained via the ``processorId`` method.
+``String`` representation of processor's path without the address part and can be obtained via the ``processorId``
+method.
 
 .. includecode:: code/docs/persistence/PersistenceDocSpec.scala#processor-id
 
 Applications can customize a processor's id by specifying an actor name during processor creation as shown in
-section :ref:`processors`. This works well when using local actor references but may cause problems with remote
+section :ref:`processors`. This changes that processor's name in its actor hierarchy and hence influences only
-actor references because their paths also contain deployment information such as host and port (and actor deployments
+part of the processor id. To fully customize a processor's id, the ``processorId`` method should be overridden.
-are likely to change during the lifetime of an application). In this case, ``Processor`` implementation classes
-should override ``processorId``.
 
 .. includecode:: code/docs/persistence/PersistenceDocSpec.scala#processor-id-override
 
-Later versions of Akka persistence will likely offer a possibility to migrate processor ids.
-
 .. _channels:
 
 Channels
@@ -295,10 +292,9 @@ Identifiers
 -----------
 
 In the same way as :ref:`processors`, channels also have an identifier that defaults to a channel's path. A channel
-identifier can therefore be customized by using a custom actor name at channel creation. As already mentioned, this
+identifier can therefore be customized by using a custom actor name at channel creation. This changes that channel's
-works well when using local actor references but may cause problems with remote actor references. In this case, an
+name in its actor hierarchy and hence influences only part of the channel identifier. To fully customize a channel
-application-defined channel id should be provided as argument to ``Channel.props(String)`` or
+identifier, it should be provided as argument ``Channel.props(String)`` or ``PersistentChannel.props(String)``.
-``PersistentChannel.props(String)``.
 
 .. includecode:: code/docs/persistence/PersistenceDocSpec.scala#channel-id-override
 

akka-persistence/src/main/scala/akka/persistence/Persistence.scala

@@ -86,5 +86,5 @@ class Persistence(val system: ExtendedActorSystem) extends Extension {
     system.asInstanceOf[ActorSystemImpl].systemActorOf(Props(pluginClass).withDispatcher(pluginDispatcherId), pluginType)
   }
 
-  private def id(ref: ActorRef) = ref.path.toStringWithAddress(system.provider.getDefaultAddress)
+  private def id(ref: ActorRef) = ref.path.elements.mkString("/", "/", "")
 }