CORBA Overview; Developing Basic CORBA Applications

CORBA Overview;Developing Basic CORBA

Applications

DS 520 - Distributed Systems Frameworks

Distributed Systems Frameworks 2

Review Anatomy and Requirements of Distributed Computing Advantages and Features of DOC Systems General Architecture of DOC Systems Overview of Object Management Architecture and CORBA


Distributed Computing Can think of DC as:

breaking down an application into individual computing agents distributed over a network work together on a cooperative task

Motivation for DC: parallel processing: can solve larger problems without larger computers applications and data may be difficult to relocate redundant processing agents for fault tolerant systems

Flavors of Distributed Programming Messaging

Asynchronous, Open-loop: programmer must maintain context Examples; socket programming, cgi (based on http), etc.

Remote Invocation Synchronous (usually), Closed-loop, Location transparency (looks like a local call) Examples: Remote Procedure Call (RPC), Distributed Objects


General Pattern for Remote Invocation

Call: marshal arguments convert to network format locate server transmit data

Client CodeClient Code

Stub

Server CodeServer Code

Skeleton

InfrastructureInfrastructure

Serve: receive data convert & unmarshal invoke method marshall return value transmit data


Anatomy of a Distributed Application

3 layers of a distributed application: network level (TCP/IP, network protocols) higher-level services (directory services, security protocols, etc.) application level

At the application level: processes threads objects: processes can be made up of one or more objects which can be accessed

by one or more threads within the process agents: an “independent” functional element (e.g., in a banking application we

may have a customer agent, a transaction agent, an information brokerage agent)

Example (the customer agent): object1: running on client machine (2 threads: listening for data; updating display) object2: running on the bank server (issuing queries; sending data back to client)


Requirements for a Distributed Application

Partitioning and Distributing Data and Functions data-driven distribution functional distribution object-based distribution

Flexible, Extendible Communication Protocols allocation of tasks to agents has a direct influence on the complexity of the

communication protocol (type of data, amount of data, persistence of connections) may also be dictated by legacy systems that need to be incorporated

Multi-threading Requirements server object may need to service multiple remote agents at once effective way to optimize resources

Security Requirements authentication of agent identities define resource access levels data encryption


Features of DOC Systems Object Interface Specification

to allow clients to access objects regardless of implementation details

Object Manager The core of a distributed object system (e.g., ORB, or Registry in RMI) Manages object skeletons and object references on the server When a client requests a new object, the object manager

locates the skeleton for the class of the requested object creates new instance based on skeleton; stores new object in the object storage sends a reference to the new object back to the client

Registration / Naming Service Acts as an intermediary between the object client and the object manager Once the interface to an object is defined, an implementation of the interface

must be registered with the service so that it can be addressed by clients

Object Communication Protocol to handle remote object requests Must support a means of transmitting and receiving object and method

references, and data in the form of objects or basic data types


General Architecture for a DOC System

IDL

Com

pil

ers

Registration Service

Object Skeleton

Object Storage

Object Manager

Naming Service

Client Application

ServerImplementation

Client StubInterface

Object Interface

Specification


Remote Object Transactions at Runtime

ObjectStubs

Server ObjectImplementation Object

Skeleton

Object Manager Naming Service

Client Application

2. Resolve Object

4. Object Interactions

1. Request Object

3. Object Handle


OMG’s Mission Develop a single architecture, using object technology, for

distributed application integration, guaranteeing: reusability of components; interoperability & portability; basis in commercially available software

Consensus-based approach

Focus on swiftly-developed, easily usable (“off the shelf”) component standards: Single terminology for object-orientation. Common abstract framework. Common reference model. Common interfaces & protocols

WEB Server

OMG IDLProgramsIIOP (CORBA)

IIOP-- HTML-- GIF, JPEG-- AV, WAV-- Java

Sea of Objects(CORBA)

WEB BrowserJava Enabled

Hot JavaJava BeansActiveX

Java Interpreter

ActiveX

APP

LegacyDCE APP

DCE

COM/CORBA

PROGRAMS

DCE

IIOP

Long Term Vision


Non-standardizedapplication-specific

interfaces

Verticaldomain-specific

interfaces

Horizontalfacility interfaces

Application Interfaces Domain Interfaces CORBAfacilities

General service interfaces

CORBAservices

Object Request Broker

Object Management Architecture


Overview of CORBA Objects CORBA objects differ from typical programming language objects:

CORBA objects can be located anywhere on a network. CORBA objects (like Java objects) can run on any platform. CORBA objects can be written in any of several languages.

CORBA object developers need know nothing of where their clients will be, what hardware or OS they will run on, or what language they will be written in. CORBA objects approach universal accessibility.

A client of an object has access to an object reference for the object, and invokes operations on the object. A client knows only the logical structure of the object according to its interface and

experiences the behavior of the object through invocations. Client code has no knowledge of the implementation of the object or which ORB is used

to access the implementation.

An object implementation provides the semantics of the object, usually by defining data for the object instance and code for the object's methods.


ClientClient Object ImplementationObject Implementation

ORB

RequestRequest

A Request A request consists of:

Target object (target object identified by a unique object reference) Operation Parameters (the input, output and in-out parameters defined for the operation;

may be specified individually or as a list) Optional request context

Results (the result values returned by the operation)

Skeletoncode

Skeletoncode

Client Proxy(stub code)

Client Proxy(stub code)


CORBA Framework Elements

Object Request Broker (ORB) This is the object manager in CORBA

Mechanisms for specifying interfaces Interface Definition Language (IDL) - for static interface definitions Dynamic Invocation Interface (DII) - lets clients access interfaces as first-class

objects at run-time from an Interface Repository.

Internet Inter-Orb Protocol (IIOP) A binary protocol for communication between ORBs. Was added in CORBA 2.0


Today CORBA Framework and Components CORBA IDL Steps in Developing CORBA Applications Examples


Object Request Broker (ORB) The Object Manager in CORBA On client side the ORB is responsible for

accepting requests for a remote object finding implementation of the object accepting client-side reference to the remote object(converted to a language

specific form, e.g., a Java stub object) routing client method calls through the object reference to the object

implementation

On server side the ORB lets object servers register new objects receives requests from the client ORB uses object’s skeleton interface to invoke object’s activation method creates reference for new object and sends it back to client


OMG IDL OMG Interface Definition Language (IDL):

mappings for many languages/compilers; independent of any particular language/compiler; multiple-inheritance, public interface-structured specification language; not for implementation. primary support for interoperability between static and dynamic requests

mechanisms.

IDL Structure Module

a namespace

Interface abstract type multiple inheritance

Struct structured data

Module auction { exception NotAllowed {};

struct Sale { int price; string item; }

interface Auction { void bid (in long price) raises NotAllowed; }}

Module auction { exception NotAllowed {};

struct Sale { int price; string item; }

interface Auction { void bid (in long price) raises NotAllowed; }}


Internet Inter-Orb Protocol (IIOP) CORBA specification is neutral with respect to network protocols

the CORBA standard specifies what is known as the General Inter-ORB Protocol (GIOP) GIOP is a high-level standard protocol for communication between ORBs not used directly; instead, it is specialized by a particular protocol that would then be

used directly

Internet Inter-ORB Protocol (IIOP) IIOP is the GIOP-based protocol for TCP/IP networks As of the 2.0 version of the CORBA specification, vendors are required to implement the

IIOP protocol

CORBA Networking Model CORBA applications are built on top of GIOP-derived protocols such as IIOP these protocols, in turn, rest on top of TCP/IP, DCE, or other underlying transport

protocol the network uses an application architecture can be designed to use a bridge that would interconnect, for

instance, DCE-based application components with IIOP-based ones.


Passing Objects by Reference In a distributed application, there are two possible methods for

an application component to obtain access to an object in another process: When an object is passed by reference, the object itself remains "in place"

while an object reference for that object is passed. Operations on the object through the object reference are actually processed by the object itself.

When an object is passed by value, the object's state is copied and passed to its destination (via object serialization), where a new copy of the object is instantiated. Operations on that object's copy are processed by the copy, not by the original object.

Note: in CORBA, objects are generally passed by reference (however, CORBA 3.0 specification has now added pass-by-value semantics).


Object References An Object Reference is the information needed to specify an object

within an ORB. The representation of an object reference handed to a client is only valid for the lifetime

of that client. The language mapping also provides additional ways to access object references in a

typed way for the convenience of the programmer. There is a distinguished object reference, the null reference, guaranteed to be different

from all object references, that denotes no object. In Java, this is a Java null.

To invoke a CORBA object, you need a reference for the object. There are two ways to get a reference for a CORBA object: from another object, such as a factory or a name service

from a string that was specially created from an object reference

Interoperable Object References CORBA uses IOR as a “pointer” to a specific instance of a class in a distributed

environment encodes host, port, object identity may be externalized (using object_to_string)


CORBA Components Client stub

Each stub represents (it is a proxy) an object operation (a possible request) which a client invokes in a language-dependent manner (e.g., by calling a subroutine which represents the operation).

The stubs make calls on the rest of the ORB using interfaces that are private to JavaIDL. Alternatively, a client may dynamically construct and invoke request objects which can

represent any object operation.

Implementation Skeleton Each skeleton provides the interface through which a method receives a request (dynamic

and static skeletons)

Object Adapter Purpose is to interface an object's implementation with its ORB Each object adapter provides access to those services of an ORB (such as activation,

deactivation, object creation, object reference management) used by a particular type of object implementation.

ORB Interface The interface to the small set of ORB operations common to all objects, e.g., the

operation which returns an object's interface type.


ObjectAdapter

ImplementationSkeletons

ORB Core

Client Object Implementation

ClientStubs

ORBInterface

DynamicInvocation

standard interface

One interface per object operation

One interface per object adaptor

Proprietary ORB interface

Normal call interfaceUp call interface

CORBA Components


Client

ClientStubs

ORBInterface

DynamicInvocation

Clients perform requests using object references.

Clients may issue requests through object interface stubs (static) or dynamic invocation interface.

Clients may access general ORB services:

• Interface Repository.• Context Management.• List Management.• Request Management.

Client Side


Object Implementation

ORBInterface

ObjectAdapter

Implementations receive requests through skeletons (without knowledge of invocation approach).

The Object Adapter provides for:• management of references;• method invocation;• authentication;• implementation registration;• activation/deactivation.

ImplementationSkeletons

Implementation Side


Static v. Dynamic Invocation Static Invocation

Static interfaces are generated in form of client stubs by the IDL (pre-) compiler.

This means that the structure of the object has to be known before hand (at compile time).

Allows for better type checking; less runtime overhead; self-documentation.

Dynamic Invocation Dynamic Invocation Interface (DII) allows clients to invoke operations on

remote objects without having access to object stubs (another way to do this without dynamic invocation is to download static client stubs via a Java applet).

Clients must discover interface-related information at runtime (e.g., using the interface repository)

Servers can offer new services anytime without the need for recompilation on the client side.


Dynamic Requests The Dynamic Invocation Interface (DII) allows clients to

dynamically: discover objects; discover objects’ interfaces; create requests; invoke requests; receive responses.

Major features of Dynamic Invocation Interface: requests appear as objects themselves; requests are reusable; invocation may be synchronous or asynchronous; requests may be generated dynamically, statically or in combination approach.


CORBA Interface Repository The Interface Repository is a service that provides persistent

objects that represent the IDL information in a form available at runtime. Note: The JavaIDL runtime does not include an implementation of an Interface

Repository and one is not generally required by clients at runtime. Using the IR, it is possible for a program to encounter an object whose

interface was not known at compile time, yet be able to determine what operations are valid on the object and make invocation on it.

Interface Repository provides: Dynamic client access to interface definitions to construct a request. Dynamic type-checking of request signatures. Traversal of inheritance graphs. ORB-to-ORB interoperability.


CORBA Implementation Repository The Implementation Repository contains information that allows the

ORB to locate and activate implementations of objects. Ordinarily, installation of implementations and control of policies

related to the activation and execution of object implementations is done through operations on the Implementation Repository.

In addition to its role in the functioning of the ORB, the Implementation Repository is a common place to store additional information associated with implementations of ORB objects. (e.g., debugging information, administrative control, resource allocation, security, etc)

The Implementation Repository supports the implementation of object servers. It is not needed by clients in order to access servers.


IDL InterfaceDefinitions

ImplementationInstallation

ClientStubs

InterfaceRepository

ImplementationRepositoryImplementation

Skeletons

Client Object Implementation

Accesses Includes DescribesIncludes

Summary of CORBA Interfaces

All objects are defined in IDL by specifying their interfaces. Object definitions (interfaces) are manifested as objects in the Interface

Repository, as client stubs, and as implementation skeletons. Descriptions of object implementations are maintained as objects in the

Implementation Repository.


Summary: CORBA Remote Method Invocation

Clients use “object interfaces” through language mapping Java clients should work on any ORB that supports the Java language bindings. Clients can call any object instance remotely, so long as the object instance

implements the interface.

Clients can call remote objects statically or dynamically The server cannot tell whether the client is using static or dynamic invocation.

Objects are identified using a unique id: Interoperable Object Reference (IOR) CORBA passes objects by reference IOR was Introduced in CORBA 2.0 Object references can be converted to strings and back to “live” objects via

ORB interface functions.


Steps in Developing CORBA Applications

Write specification for each object using IDL Use IDL Compiler (e.g., idl2java) to generate:

Client Stub code Server Skeleton code

Write the client (in Java, can be applications or applets) Write the server object implementation code (the “servant”) Compile the client and server code Start the server Run the client application


The Structure of CORBA IDL(See Section 2.3.4 of Brose, Vogel, Duddy)

Modules provide namespaces for a set of related class descriptions map to Java packages with the same name

Interfaces specification of the operations a client can invoke on a remote object in addition to operations, interfaces include constructs such as:

constant declarations attributes (can be read/write or readonly; implementation automatically creates get

and set operations for these attributes) exceptions (raised if the operations do not perform successfully)

Operations CORBA equivalent of methods in Java IDL defines the operations’ signature: parameters and return values/types parameters can be in, out, or inout IDL also defines what exceptions can be raised by the operation


Data Types in CORBA IDL Basic Types

short, long, unsigned long, unsigned short, float, double, long double, char, wchar, boolean, string, octet, etc.

Constructed Types struct and union (similar to C++; can be used in conjunction with a typedef) sequence (variable sized arrays of objects) any (generic type which represents any possible IDL types; similar to the Java

Object type) enum (enumerated type with named integer values) arrays valuetypes (similar to interfaces; preceded with keyword valuetype to provide

pass-by-value semantics)

Each CORBA IDL data type gets mapped to a native data type via the appropriate language binding (e.g, IDL-to-Java mapping).


IDL Syntaxmodule <identifier> {

<type declarations>;

<constant declarations>;

<exception declarations>;

interface <identifier> [:<inheritance>] {



<attribute declarations>;


[<op-type>]<identifier>(<parameters>)

[raises <exceptions>]



. . .

};

interface <identifier> [:<inheritance>] {. . .};

. . .

};

module <identifier> {




interface <identifier> [:<inheritance>] {



<attribute declarations>;






. . .

};

interface <identifier> [:<inheritance>] {. . .};

. . .

};


An IDL Examplemodule MyAnimals {

interface Dog: Pet, Animal {

attribute short age;

exception NotInterested{string explanation};

void Bark(in short how_long)

raises (NotInterested);

void Sit(in string where)


. . .

};

interface Cat: Animal {

void Eat();

. . .

};

};

module MyAnimals {

interface Dog: Pet, Animal {

attribute short age;

exception NotInterested{string explanation};

void Bark(in short how_long)


void Sit(in string where)


. . .

};

interface Cat: Animal {

void Eat();

. . .

};

};


module RoomBooking {

interface Meeting { readonly attribute string purpose; readonly attribute string participants; oneway void destroy(); };

interface MeetingFactory { Meeting CreateMeeting(in string purpose,in string participants); };

interface Room { // A Room provides methods to view, make,cancel bookings.

enum Slot { am9, am10, am11, pm12, pm1, pm2, pm3, pm4 }; const short MaxSlots = 8; typedef Meeting Meetings[ MaxSlots ];

exception NoMeetingInThisSlot {}; exception SlotAlreadyTaken {};

readonly attribute string name;

Meetings View(); void Book( in Slot a_slot, in Meeting a_meeting ) raises(SlotAlreadyTaken); void Cancel( in Slot a_slot ) raises(NoMeetingInThisSlot); };};

module RoomBooking {

interface Meeting { readonly attribute string purpose; readonly attribute string participants; oneway void destroy(); };

interface MeetingFactory { Meeting CreateMeeting(in string purpose,in string participants); };

interface Room { // A Room provides methods to view, make,cancel bookings.

enum Slot { am9, am10, am11, pm12, pm1, pm2, pm3, pm4 }; const short MaxSlots = 8; typedef Meeting Meetings[ MaxSlots ];

exception NoMeetingInThisSlot {}; exception SlotAlreadyTaken {};

readonly attribute string name;

Meetings View(); void Book( in Slot a_slot, in Meeting a_meeting ) raises(SlotAlreadyTaken); void Cancel( in Slot a_slot ) raises(NoMeetingInThisSlot); };};

Another IDL Example


The CORBA Count Example

// Count.idl

module Counter

{

interface Count

{

attribute long sum;

long increment();

};

};

The Count Program the client will invoke a method increment() on the remote object Count; increment() adds 1 to the value of the attribute sum and returns the value

to the client program; the initial value of sum is set by the client (sum is a read/write attribute, so the

client can use CORBA’s built-in accessor functions to manipulate it.

First we must start with an IDL definition for the Count object:


The IDL Compiler Before using the IDL compiler or running the program make sure that

the PATH and CLASSPATH environment variables are properly set up:

Your PATH must point to the Java runtime environment directory as well as the ORB runtime directory ( \inprise\vbroker\bin);

Your CLASSPATH variable must include the current directory (“.”), the usual Java classes and libraries (including the directory where your program classes will be stored) as well as the following jar files required by Visibroker:

\inprise\vbroker\lib\vbjcosnm.jar\inprise\vbroker\lib\vbjorb.jar\inprise\vbroker\lib\vbjapp.jar\inprise\vbroker\lib\vbjtools.jar\inprise\vbroker\lib\vbjgk.jar

\inprise\vbroker\lib\vbjcosnm.jar\inprise\vbroker\lib\vbjorb.jar\inprise\vbroker\lib\vbjapp.jar\inprise\vbroker\lib\vbjtools.jar\inprise\vbroker\lib\vbjgk.jar

In Visibroker 3.x

\inprise\vbroker\lib\vbdev.jar\inprise\vbroker\lib\vbjdev.jar\inprise\vbroker\lib\vbjorb.jar\inprise\vbroker\lib\vbjmigration.jar\jdk1.2.2\lib\servlet.jar

\inprise\vbroker\lib\vbdev.jar\inprise\vbroker\lib\vbjdev.jar\inprise\vbroker\lib\vbjorb.jar\inprise\vbroker\lib\vbjmigration.jar\jdk1.2.2\lib\servlet.jar

In Visibroker 4.x


The IDL Compiler Now you can compile the IDL definition:

> idl2java -no_tie Count.idl

Important Notes: if you are using Visibroker 3.3, you need JDK 1.1.8; as VBJ 3.3 is not

compatible with Java 2 platform

in VBJ 4.x, idl2java compiler generated code is based on the POA (portable object adapter) semantics; in order to use BOA semantics, must use:

> idl2java -boa Count.idl

also some additional casting needs to be done in BOA initialization (more on this later, but see Chapters 30 and 31 of the VBJ 4.x Programmer’s Guide)

in VBJ 3.x, the idl2java compiler generated code is based on BOA semantics (also the API for the stub and skeleton code is different than those in VBJ 4.x)


What Does the IDL Compiler Generate?

idl2java

Count.idl

_st_Count.java

CountHelper.java CountHolder.java

_CountImplBase.java

Count.java _example_Count.java

Client-Side Server-Side

Client Stub Implementation Skeleton

Note: these source files will be a part of the Java package Counter and will be placed ina directory with the same name.

VBJ 3.xVBJ 3.x


What Does the IDL Compiler Generate?

idl2java

Count.idl

_CountStub.java

CountHelper.java CountHolder.java

CountPOA.java

Count.java

Client-Side Server-Side

Client Stub Implementation Skeleton

Note: if -boa option is used with idl2java compiler, _CountImplBase.java is alsogenerated to be used in the object implementation.

VBJ 4.x using POAVBJ 4.x using POA


Generated Classes & Interfaces Counter._st_Count: implements the client-side stub for the Count object; it

provides the marshaling functions for the client. [_CountStub in VBJ 4.x]. Counter._CountImplBase: implements the server-side skeleton for Count; it

unmarshals the arguments passes by the client. [CountPOA in VBJ 4.x using POA]. Counter.CountHelper: provides some additional utility functions for the client,

including the narrow operation for downcasting CORBA object references, and (in this case) the Visibroker specific bind operation for finding the remote object.

Counter.CountHolder: it is used as a wrapper for passing out and inout parameters of type Count (Java only natively supports in parameters).

Counter.Count: contains the Java interface corresponding to the IDL interface for Count; the object implementation for Count must implement this interface.

Counter._example_Count: a sample class for the Count object implementation which could be filled-in by the programmer. [not generated in VBJ 4.x].

package Counter;public interface Count extends org.omg.CORBA.Object { public void sum(int sum); public int sum(); public int increment();}


Class Hierarchy for BOA Count Interface

classorg.omg.CORBA.portable.Skeleton(provided by the ORB)

classorg.omg.CORBA.portable.Skeleton(provided by the ORB)

abstract classCounter._CountImplBase(generated by id2java)

abstract classCounter._CountImplBase(generated by id2java)

class CountImpl(written by programmer and used by the server)

class CountImpl(written by programmer and used by the server)

interfaceCounter.Count(generated by id2java)

interfaceCounter.Count(generated by id2java)

Extends

Extends

Implements

classorg.omg.CORBA.portable.ObjectImpl(provided by the ORB)

classorg.omg.CORBA.portable.ObjectImpl(provided by the ORB)

abstract classCounter._st_Count(generated by id2java)

abstract classCounter._st_Count(generated by id2java)

Extends

Note: in case of POA, the object skeleton “CountPOA”extends org.omg.CORBA.PortableServer.Servant

Note: in case of POA, the object skeleton “CountPOA”extends org.omg.CORBA.PortableServer.Servant


Steps in Developing the Count Application

Write a client program in Java: this can be an applet or an application; in this case we write a Java application CountClient.java.

Write the server-side code for the Count application: write a server application which creates and serves instance(s) of the Count object

(instances of the class CountImpl); write the object implementation (CountImpl).

Object Implementation: CORBA inheritance model: the implementation class is always derived from the

corresponding _XXXImplBase class generated by IDL compiler (in this case from _CountImplBase; [Note: with POA this is XXXPOA, e.g., CountPOA]

this inheritance allows the servant class to obtain the properties of both CORBA and Java object models;

the alternative to inheritance model is to use delegation (in CORBA terminology the Tie method); if -no_tie option is not used with idl2java relevant classes will be generated;

in this case we create our CountImpl by modifying the file _example_Count.java.


_example_Count.java

package Counter;

public class _example_Count extends Counter._CountImplBase {

public _example_Count(java.lang.String name) {

super(name);

}

public _example_Count() {

super();

}

public int increment() {

// implement operation...

return 0;

}

public void sum(int sum) {

// implement attribute writer...

}

public int sum() {

// implement attribute reader...

return 0;

} }

package Counter;

public class _example_Count extends Counter._CountImplBase {

public _example_Count(java.lang.String name) {

super(name);

}

public _example_Count() {

super();

}

public int increment() {

// implement operation...

return 0;

}

public void sum(int sum) {

// implement attribute writer...

}

public int sum() {

// implement attribute reader...

return 0;

} }


// CountImpl.java: The Count Implementation

class CountImpl extends Counter._CountImplBase{ private int sum;

// Constructors CountImpl(String name) { super(name); System.out.println("Count Object Created"); sum = 0; }

// get sum public int sum() { return sum; }

// set sum public void sum(int val) { sum = val; }

// increment method public int increment() { sum++; return sum; }}

The BOA Count Object Implementation


Tasks Performed by the Main Server Class Initialize the Object Request Broker

done by creating an instance of the ORB pseudo-object; the static Java method init() on the class org.omg.CORBA.ORB returns

an instance of an ORB.

Initialize the Basic Object Adaptor a reference to an instance of the BOA is obtained via the method BOA_init().

Create the object (an instance of the class implementation) in this case we create an instance of CountImpl class

Activate the newly created object in BOA this is done by calling the obj_is_ready() method of the ORB

instance on the object; this action exports the object to the ORB.

Wait for incoming requests in BOA this is done by calling the impl_is_ready() method of the ORB

instance on the object;


// CountServer.java: The Count Server main program

class CountServer { static public void main(String[] args) { try { // Initialize the ORB org.omg.CORBA.ORB orb = org.omg.CORBA.ORB.init(args, null);

// Initialize the BOA org.omg.CORBA.BOA boa = orb.BOA_init();

// Note: if using VBJ 4.x with -boa option the BOA // initialization must be done with following cast: // com.inprise.vbroker.CORBA.BOA boa = // ((com.inprise.vbroker.CORBA.ORB)orb).BOA_init();

// Create the Count object CountImpl count = new CountImpl("My Count");

// Export to the ORB the newly created object boa.obj_is_ready(count);

// Ready to service requests boa.impl_is_ready(); }

catch(org.omg.CORBA.SystemException e) { System.err.println(e); } }}

TheBOA Count Server


Tasks Performed by the Client

Initialize the ORB done the same way as for the server.

Locate the Remote Object in this case client binds to the object via the Visibroker bind() method.

Perform Operations via Remote Methods set the remote sum attribute to zero; calculate the start time; invoke the increment method 1000 times; calculate the elapsed time; print the results.


// CountClient.java Static Client, VisiBroker for Java

class CountClient{ public static void main(String args[]) { try { // Initialize the ORB org.omg.CORBA.ORB orb = org.omg.CORBA.ORB.init(args, null);

// Bind to the “persistent” Count Object reference Counter.Count counter = Counter.CountHelper.bind(orb, "My Count");

// Set sum to initial value of 0 counter.sum((int)0);

// Calculate Start time long startTime = System.currentTimeMillis();

// Increment 1000 times for (int i = 0 ; i < 1000 ; i++ ) { counter.increment();}

// Calculate stop time; print out statistics long stopTime = System.currentTimeMillis(); System.out.println("Avg Ping = " + ((stopTime-startTime)/1000f)+" msecs"); System.out.println("Sum = " + counter.sum()); } catch(org.omg.CORBA.SystemException e) { System.err.println("System Exception"); System.err.println(e); } } }

The BOA Count Client


Building and Running the Count Example Compile the client, server, and the implementation classes

vbjc -d \CorbaJavaBook.2e\classes CountClient.java

vbjc -d \CorbaJavaBook.2e\classes CountServer.java

vbjc -d \CorbaJavaBook.2e\classes CountImpl.java

Note: \CorbaJavaBook.2e\classes is where the Java bytecode classes will be stored (without -d option classes are generated in the current directory).

Start the Visibroker Smart Agent (OSAgent) osagent -c OR start osagent -c OSAgent may be installed as an NT Service or started using the icon in the

program menu

Start the Count Server: vbj CountServer

Execute the Client Application: vbj CountClient


The Count Program in

Action


Next time More Examples of CORBA Applications CORBA Object References The POA version of Count Application More on IDL elements

read/readonly attributes oneway methods parameter passing mechanisms exceptions and exception handling

Discussion of Project Phase I

CORBA Overview; Developing Basic CORBA Applications

Documents

Transcript of CORBA Overview; Developing Basic CORBA Applications