Skip to main content

Notice: this Wiki will be going read only early in 2024 and edits will no longer be possible. Please see: https://gitlab.eclipse.org/eclipsefdn/helpdesk/-/wikis/Wiki-shutdown-plan for the plan.

Jump to: navigation, search

Difference between revisions of "ECF/Asynchronous Remote Services"

< ECF
Line 42: Line 42:
 
</source>
 
</source>
  
<Under Construction>
+
With an ECF created asynchronous proxy, the consumer thread that calls helloAsyncService.helloAsync is guaranteed '''not''' to block, and the appropriate IAsyncCallback method will be executed when the remote call completes (with either success or failure).  The callback is called (by an arbitrary ECF thread) sometime '''after''' the helloAsync method completes. 
 +
 
 +
A nice thing is that neither the consumer nor the host have to actually implement the asynchronous proxy.  The proxy is automatically constructed by the ECF remote services implementation when accessed by a consumer.
 +
 
 +
At this point, you might ask:  But how is the asynchronous proxy defined?  i.e. where does the the '''helloAsyncService''' come from?
 +
 
 +
The answer to this is that it is defined in a new/second service interface that is related to the IHello service interface:
 +
 
 +
<source lang="java">
 +
public interface IHelloAsync extends IAsyncRemoteServiceProxy {
 +
 
 +
public void helloAsync(String from, IAsyncCallback callback);
 +
}
 +
</source>
 +
 
 +
Notice that this async interface declaration resembles the IHello service interface declaration, but differs from it in a couple of ways:
 +
 
 +
1) The name is '''IHelloAsync''' rather than '''IHello'''
 +
2) It extends '''IAsyncRemoteServiceProxy'''
 +
3) The method name is '''helloAsync''' rather than '''hello'''
 +
4) The method arguments are String and IAsyncCallback rather than just String
 +
 
 +
With ECF 3.3 remote services, when a proxy is created, '''iff''' an interface class with the name '''[fq service interface name]Async''' can be loaded and it extends IAsyncRemoteServiceProxy then the proxy will implement that interface.  So, for our hello example the proxy service reference will implement both the IHello and IHelloAsync methods, and the consumer can use '''either''' of the methods declared.  So, for example
 +
 
 +
<source lang="java">
 +
IHello helloService = ...get service reference via declarative services/injection, or ServiceTracker, or other...
 +
if (helloService instanceof IHelloAsync) {
 +
  IHelloAsync helloServiceAsync = (IHelloAsync) helloService;
 +
  // call it asynchronously
 +
  helloAsyncService.helloAsync("slewis", new IAsyncCallback() {
 +
      public void onSuccess(Object result) {
 +
          // Do something with result
 +
      }
 +
      public void onFailure(Throwable t) {
 +
          // Deal with network failure in appropriate manner
 +
      }
 +
  });
 +
}
 +
</source>
 +
 
 +
This gives the consumer maximum flexibility in determining how a given invocation will occur (i.e. synchronously or asynchronously.  The caller can use the synchronous proxy, or simply cast to the asynchronous proxy.  Or both invocation methods can be used as desired.
 +
 
 +
 
 +
 
 +
===Futures===

Revision as of 18:27, 15 April 2010

ECF's Remote Services API, which is used to implement OSGi 4.2 remote services, has had the ability for consumers/clients to use asynchronous/non-block remote method calls for at least the last 2 years.

For background on the discussion about asynchronous remote services invocation in general and possible future standardization by the OSGI standards organization see Peter Kriens blog entry, and Scott Lewis' blog entry.

Normal/Synchronous Proxies

When an OSGi service is actually invoked today, this is done by the caller calling a method on the service. For example, consider a simple 'hello' service:

package org.eclipse.ecf.examples.remoteservices.hello;
 
public interface IHello {
 
	public void hello(String from);
 
}

Once a consumer of this service gets a valid service reference (i.e. through ServiceTracker, declarative services, getServiceReference or however), it can actually invoke/use the service via a method call on 'hello':

helloService.hello("slewis");

With OSGi remote services, this 'helloService' may actually be a proxy. If invoked as above, under the covers, a proxy will marshall the method arguments (i.e. 'slewis' String in this case), and then communicate with the remote service host via some network (i.e. via some protocol). If there is a result of the call, then it will then be unmarshalled and return to the caller thread. With synchronous invocation the thread that calls 'hello' will block until this entire process is complete.

Asynchronous Proxies

ECF has just added the ability to declare asynchronous access to a remote method, so that the consumer can ge guaranteed that calling the service will not block indefinitely. So, for example, it's now possible to make a call to the hello service like this

helloAsyncService.helloAsync("slewis", new IAsyncCallback() {
    public void onSuccess(Object result) {
        // Do something with result
    }
    public void onFailure(Throwable t) {
        // Deal with network failure in appropriate manner
    }
});

With an ECF created asynchronous proxy, the consumer thread that calls helloAsyncService.helloAsync is guaranteed not to block, and the appropriate IAsyncCallback method will be executed when the remote call completes (with either success or failure). The callback is called (by an arbitrary ECF thread) sometime after the helloAsync method completes.

A nice thing is that neither the consumer nor the host have to actually implement the asynchronous proxy. The proxy is automatically constructed by the ECF remote services implementation when accessed by a consumer.

At this point, you might ask: But how is the asynchronous proxy defined? i.e. where does the the helloAsyncService come from?

The answer to this is that it is defined in a new/second service interface that is related to the IHello service interface:

public interface IHelloAsync extends IAsyncRemoteServiceProxy {
 
	public void helloAsync(String from, IAsyncCallback callback);
}

Notice that this async interface declaration resembles the IHello service interface declaration, but differs from it in a couple of ways:

1) The name is IHelloAsync rather than IHello 2) It extends IAsyncRemoteServiceProxy 3) The method name is helloAsync rather than hello 4) The method arguments are String and IAsyncCallback rather than just String

With ECF 3.3 remote services, when a proxy is created, iff an interface class with the name [fq service interface name]Async can be loaded and it extends IAsyncRemoteServiceProxy then the proxy will implement that interface. So, for our hello example the proxy service reference will implement both the IHello and IHelloAsync methods, and the consumer can use either of the methods declared. So, for example

IHello helloService = ...get service reference via declarative services/injection, or ServiceTracker, or other...
if (helloService instanceof IHelloAsync) {
   IHelloAsync helloServiceAsync = (IHelloAsync) helloService;
   // call it asynchronously
   helloAsyncService.helloAsync("slewis", new IAsyncCallback() {
       public void onSuccess(Object result) {
           // Do something with result
       }
       public void onFailure(Throwable t) {
           // Deal with network failure in appropriate manner
       }
   });
}

This gives the consumer maximum flexibility in determining how a given invocation will occur (i.e. synchronously or asynchronously. The caller can use the synchronous proxy, or simply cast to the asynchronous proxy. Or both invocation methods can be used as desired.


Futures

Back to the top