Harshana Eranga Martin's Blog

FA Community Shield 2011 - Epic Manchester United

noreply@blogger.com (Harshana Eranga Martin) — Sat, 03 Sep 2011 11:06:00 +0000

FA community shield 2011 was taken place on 7th August 2011 at Wembley Stadium, London. This time the parties involved was Manchester United as usual and Manchester city.

Manchester United started the game in a brilliant way and keep creating chances but failed to materialize any of 'em. Finally city got 2 goals out of nothing.

In the second half United started where they left it in the first half and this time we managed to score and game on from there. Then Nani scored the best team goal I have seen for last few years by Manchester United. Cleverly was instrumental in attacking and in the midfield. Finally Nani scored the injury time winner to mark a yet another remarkable comeback win for Manchester United.

Official FA Highlughts of the match:

Official FA Pitchside highlights of the match:

Glory Glory Man Utd! #MUFC #RedDevils

Apache Maven Incremental Build support for WSO2 Carbon

noreply@blogger.com (Harshana Eranga Martin) — Sat, 21 May 2011 07:43:00 +0000

Apache Maven is one of the most widely used software build systems today. If you have a codebase size of multiple hundreds of megabytes or even few gigabytes, then Maven will be the best option in terms of easiness of modularity and management of components.

However when you have a codebase of multiple hundreds of megabytes or few gigabytes, it leads to several issues even you build your system with Maven. One such pressing matter is build time for your codebase. If I explain it more clearly, when you change a single component, you need to build all the systems depending on that component. But currently maven does not support this dependency based component building. Therefore you need to build the whole codebase after each commit to your codebase. This means you build a set of components you are really never changed by your commit or which are never affected by the change you made, which is clearly a waste of time and resources. One way of reducing this wastage is, introducing Build Profiles. But still, this will build whole lot of components which never changed or affected by your commit. So you will understand build profile is not the ideal solution for this. Then what is the ideal solution for this issue? It is INCREMENTAL BUILDING support for the codebase.

What is Incremental Building?

Incremental Build means rebuild only the minimal set components of the source code after making a change to the source code. This means avoid rebuilding the complete codebase when a change made in a component and build only the required set of components which are,

1. Changed Component

2. The dependent components of the changed component.

Why we cannot simply use "mvn install"?

Let's say we have a component-parent with 2 child components, component-api and component-impl where component-api defines the API and component-impl contains the implementation of the API defined in the component-api.

component-parent

|- component-api

|- component-impl

As you can see in the above simple diagram, component-api is a dependency for the component-impl since component-impl implements the APi defined in the component-api.

Let's think of a scenario where we modify the component-api but not component-impl accordingly which should lead to a compilation error . Then we build the component-parent with "mvn install". When we do this what happens is, Maven will rebuild the component-api and it will skip building component-impl displaying "Nothing to compile - all classes are up to date" message. Hence the build is successful despite the compilation errors at the component-impl. This means, though we have modified the dependency of the component-impl, Maven only looks at the current component. Therefore using mvn install is not the solution for Incremental Build with Maven.

Is Maven useless for Incremental Building?

The answer is NO. Maven does support Incremental Building but the support which comes by default is useless and inaccurate since the inability of Maven to look at dependencies and identify the changes. Therefore when this issue is fixed, Maven supports incremental building.

Is there a solution which is already Implemented?

Yes. Lucky for us, there is a solution implemented called Maven-Incremental build plugin. This is a maven plugin for Apache Maven and we can use it to enable fully functional true incremental build for Maven.

How does it work?

Maven-Incremental build plugin executed in the first phase (Validate) of the default maven Lifecycle. What it does is, it verifies the current component source (src/main/java) /resources (src/main/resources) or it's dependencies are not being updated after last build. If there is a change at least in one of them, maven-incremental-build plugin automatically cleans the target so that it enforces to recompile the project.

If you think of the scenario, you will understand this is the perfect solution. If I explain it here, when we change the component-api, while building the component-impl, maven-incremental build plugin identifies that the dependency component-api is changed and so that the component-impl target is cleaned. Therefore the component-impl is recompiled and in the process, it will detect the compilation error.

How to enable the Maven-Incremental build for your project?

All you have to do is declare the maven-incremental build plugin in your root pom file.

<build>
<plugins>
<plugin>
<groupId>org.jvnet.maven.incrementalbuild</groupId>
<artifactId>incremental-build-plugin</artifactId>
<version>1.3</version>
<executions>
<execution>
<goals>
<goal>incremental-build</goal>
</goals>
</execution>
</executions>
</plugin>
</plugins>
</build>

The above xml element declares the maven-incremental build plugin for your project. But you need to add the following section to your pom to enable automatic download of the maven-incremental build plugin from it's plugin repository.

<pluginRepositories>
<pluginRepository>
<id>repository.dev.java.net-maven2</id>
<name>Java.net Repository for Maven</name>
<url>http://download.java.net/maven/2/</url>
<layout>default</layout>
</pluginRepository>
</pluginRepositories>

After adding these 2 elements to your pom file. you are ready to use maven-incremental-plugin.

Is there anyone use this plugin? If Yes, How?

Yes. There are lot of people who uses this plugin to improve their productivity. WSO2 is one of them.

When WSO2 Carbon 3.2.0 release work started, introducing incremental build support to Carbon was one of the major task scheduled for that release. Hence I took the ownership of this task [1] and integrated the Maven Incremental Build plugin to WSO2 Carbon 3.2.0 build. Hence WSO2 Carbon now supports build using Maven Incremental builder plugin.

As I have explained previously in this post, First you need to add Maven Incremental plugin and java.net plugin repository to your root pom file. You can find how I have done it for WSO2 Carbon here [2].

Then you need to check whether all your maven modules are inherit from the root pom. This has to be done since there can be dependencies which are not directly extend your root pom file. This is the case for our WSO2 Carbon dependencies. These dependencies do extend their respective Apache product poms such as Axis2 parent pom is org.apache:apache:8. In that case you need to add those plugin repository entry and plugin entry to Axis2 pom as well. You can see how we have done it for Axis2 from here [3]. This is true for other dependencies such as ODE as well. You can see all our dependencies from here [4].

Everything other component in our WSO2 Carbon do extend our Carbon root pom. For example Carbon Core [5] inherits from Carbon root pom. Therefore we don't need to add Incremental plugin entries to it since Carbon Core inherits those plugins from root pom.

Now you are ready to use maven Incremental build plugin.

How to execute the Maven Incremental build?

Maven Incremental build plugin is mapped to the Validate phase of the default lifecycle [6]. Hence it is executed by default when you execute any phase which comes after Validate. Hence you can execute it with a maven command such as "mvn compile" or "mvn package" or "mvn install" or "mvn deploy" since all these phases (compile, package, install, deploy) comes after validate phase.

Since Incremental build handles the cleaning of targets, you need to avoid mentioning "clean" phase in your maven command.

Hence you can use "mvn compile" or "mvn package" or "mvn install" or "mvn deploy" commands to execute Incremental build.

[1]. http://markmail.org/message/nsj7extfmn4mjuqs

[2]. https://svn.wso2.org/repos/wso2/branches/carbon/3.2.0/pom.xml

[3]. https://svn.wso2.org/repos/wso2/branches/carbon/3.2.0/dependencies/axis2/1.6.1-wso2v1/pom.xml

[4]. https://svn.wso2.org/repos/wso2/branches/carbon/3.2.0/dependencies/

[5]. https://svn.wso2.org/repos/wso2/branches/carbon/3.2.0/core/pom.xml

[6].http://maven.apache.org/guides/introduction/introduction-to-the-lifecycle.html#Build_Lifecycle_Basics

Manchester United Vs Chelsea - UEFA Champions League - Home Leg

noreply@blogger.com (Harshana Eranga Martin) — Thu, 14 Apr 2011 05:40:00 +0000

Manchester United v Chelsea by fifirbo

Chelsea humiliated at The Theater of Dreams - Old Trafford, Manchester. 2-1 to Manchester United. Man United responds to the chelsea goal with in 20 seconds to end their hope. Ji scores in front of Stretford End. Can there be a better way to score the winner? :)

Manchester United Vs Chelsea - UEFA Champions League - Away Leg

noreply@blogger.com (Harshana Eranga Martin) — Thu, 14 Apr 2011 05:30:00 +0000

Chelsea v Manchester United by jordanip

Manchester United ends Chelsea dreams of euro glory in their home.

Git: Why it stands above other revision control systems and why we should be using it

noreply@blogger.com (Harshana Eranga Martin) — Sun, 20 Feb 2011 10:03:00 +0000

Linus talks about why he designed and implemented Git and why Git has been so superior over other existing revision control systems. This is a very interesting tech talk.

WSO2 Carbon Studio: Tools for WSO2 Middleware Platform

noreply@blogger.com (Harshana Eranga Martin) — Sat, 19 Feb 2011 14:34:00 +0000

WSO2 is an open source middleware company with a complete and comprehensive SOA middleware platform which is well known as Carbon and a PaaS which is known as Stratos.

But without proper tools, developers who develop their own solutions based on these SOA middleware will not get maximum benefits of them. Therefore, tools can play a significant role in reaching new sights. WSO2 Carbon Studio plays this significant role for WSO2 Carbon stack and Stratos PaaS by allowing new developers to try and evaluate Carbon platform with ease and making tasks easier for existing users.

WSO2 Carbon Studio is a collection of Eclipse plugins which enhances Eclipse IDE functionalities by extending IDE features to support WSO2 Carbon products and Stratos with many other SOA features. Therefore Eclipse users can download Carbon Studio install it on your Eclipse installation. We decided to use Eclipse because it is completely free and open source and it is the most popular and widely used Java IDE and it will allow us to reach a much wider audience than any other IDE.

Following are the features that are included in the latest Carbon Studio 1.0.4 release.

Application Server Tools

Create and Edit Apache Axis2 Web Service
- Contract first (Top down approach)
- Code first (Bottom-up approach)
Create WSDL for Apache Axis2 Web Service archive (aar file)
Generate Web Service client
- From aar file
- From WSDL
Web Applications
- Create and edit web applications
Test and debug Apache Axis2 Services and Web Applications
Deploy Apache Axis2 services and web applications
Deploy JAX-WS services

Enterprise Service Bus Tools

View, Create and Edit
- Endpoints
- Proxy Services
- Sequences
- Local Entries
Create custom mediators
Registry Referencing
Test and debug custom mediators and other ESB artifacts
Deploy custom mediators and other ESB artifacts
- containing as hot deployable file
- containing as registry resource

Governance Registry Tools

Create and deploy registry resource artifacts
- from a local file or a folder
- Import from a registry or as a registry dump
Create, edit, debug and deploy registry handlers and filters
Registry Management
- Working with a registry online
  - Adding multiple remote registries at once
  - View, add, edit and delete registry resources and collections
  - Import (drag-drop) resources from registry and file system
  - View, add, edit and delete Properties, Associations, Dependencies, Comments and Tags
  - Modify permission of a resource or collection
  - Easily modify resources through configured Eclipse editors
  - Check-out registry content to Eclipse workspace
- Working with a registry in the offline mode
  - Add resources in Eclipse workspace to the registry
  - Add, modify and delete checked-out resources in the workspace and commit the changes back
  - Sync the checked-out resources with the online registry
- User Management
  - Add, modify and delete users
  - Modify the permissions of a given role
  - Modify permission for a selected registry resource

Business Process Server Tools

View, create and edit BPEL projects
Test and deploy BPEL artifacts

Gadget Server Tools

Create and edit gadgets
Test and deploy gadget artifacts

Data Services Server Tools

Create and edit data services (XML configurations)
Create and edit data services validators
Test and deploy data services artifacts and data services validators

Carbon Tools

Create, edit, debug and deploy Carbon UI bundles
Deploy third party libraries as bundles
Test and deploy data service artifacts and data service validators

The recommended configuration for WSO2 Carbon Studio is as follow.

1. Eclipse Helios (3.6) or Helios SR1 (3.6.1) for Java EE developers.

2. Oracle JDK 1.6

3. Any Operating System

4. Around 100Mb of Hard Disk space

In order to install Carbon Studio, there are 2 methods of doing that. You can follow any of them according to your preference.

1. Offline installation via Downloaded WSO2 Carbon Studio P2 features

- If you prefer download first and install later, this option is for you.

2. Online installation via WSO2 P2 feature repository at http://dist.wso2.org/p2/carbon-studio/releases/1.0.4/

- If you have a fast internet connection, you can try this.

You can find more information regarding installation from http://wso2.org/library/knowledge-base/2010/10/quick-start-wso2-carbon-studio.

If you are using WSO2 Carbon platform in your development environment, now it is time to try Carbon Studio.

If you have any questions regarding WSO2 Carbon Studio or any other Carbon product, feel free drop a mail to carbon-dev@wso2.org. If your problem is specifically about Carbon Studio, you can use tools-user@wso2.org.

You can reach WSO2 Tooling team via WSO2 Carbon Studio forum as well.

Creating and Applying a patch for Subversion

noreply@blogger.com (Harshana Eranga Martin) — Sat, 19 Feb 2011 08:59:00 +0000

If you are working on a source code managed by Subversion, usually you deal with providing patches for your local changes against the base version in the source control. So it is important to know about, how to create a patch using Subversion command line tools.

To create a patch, follow the synatx

svn diff YOUR_FILE_OR_FOLDER > PATCH_FILE_NAME

For example, if I need to create the patch for MyClass.java and I want the patch file name to be MyClass.patch, then the command is,

svn diff MyClass.java > MyClass.patch

Next you might want to apply a patch provided by someone to your source code. Then the Syntax for applying patch is,

patch -p0 < PATCH_FILE_NAME

For the above example of MyClass.patch the command is,

patch -p0 < MyClass.patch

Hope this helps.

Watch your favorite sport online

noreply@blogger.com (Harshana Eranga Martin) — Sun, 23 Jan 2011 07:37:00 +0000

If you are a sports fan or enjoys watching sports, then the best option for you is to subscribe with a dedicated sports channel such as ESPN, Star Sports, Ten Sports. But if you are unable to do that, then the next option is watching these channels in Internet. There are lot of sites who are providing this facility. I found the following web sites to be quite useful than the other sites.

1. http://myp2p.eu/
2. http://www.livescorehunter.com

These sites offer channels with various video quality, languages and various software including flash on your browser. So if you want to watch sports with your 512 kbps broadband link, I think these sites offer better options for you.

Hope you will enjoy the coverage.

WSO2 Carbon

noreply@blogger.com (Harshana Eranga Martin) — Sun, 21 Nov 2010 11:41:00 +0000

WSO2 Inc. is a US based open source enterprise middle-ware company. WSO2 has more than 10 products which are completely inter-operable and these products are widely known as the WSO2 Carbon product stack.

WSO2 Carbon is the core of the each product in WSO2 Carbon product stack and it provides key aspects of all the products such as security, clustering, logging, statistics, and management. WSO2 Carbon is powered by OSGi technology. Hence it poses all the capabilities provided OSGi framework including Modularization, Dynamic Component Installing, Starting, Updating, Stopping and Uninstalling and etc. This Dynamic modularization behavior of OSGi makes it really easy to build rich set of products on top of WSO2 Carbon core. In simple words, WSO2 takes the full advantage of OSGi framework and provide the rich set of middleware products by encompassing capabilities such as service hosting and management, message routing and transformation, governance and identity management, business process management and business activity monitoring, and etc to the WSO2 Carbon Core.

The Figure 1 shown below explains the overall architecture of WSO2 Carbon product stack. This shows how the Carbon core is implemented using a set of OSGi services including security, UI, clustering, etc and how the rest of the WSO2 Carbon Features are built on top of WSO2 Carbon Core and provide rich set of services to users such as Message Mediation, SOA Governance, etc.

Figure 1: WSO2 Carbon Core Architecture

As it is explained in the Figure 1, WSO2 Carbon features are build on top of WSO2 Carbon Core and implement a rich set of products such as WSO2 Enterprise Service Bus (WSO2 ESB), WSO2 Application Server (Previously known as WSO2 WSAS) , WSO2 Data Services Server, etc. The following Figure 2 shows how each of these products are been built on top of the WSO2 Carbon Core and what are the WSO2 Carbon Features comprises each of these products.

Figure 2: WSO2 Products and their composition

As I have been explaining throughout this post, the specialties of these products in the WSO2 Carbon product stack are, they are inter-operable and they nicely fit together to create a complete and comprehensive middle-ware platform. The Figure 3 shows how these products fit together and create this middle-ware platform.

Figure 3: WSO2 Complete middle-ware platform

You can make use these products in your business by building your business solutions on top of the WSO2 products. They are distributed completely free of charge under the Apache Software License and you can use these products to build your commercial applications without worrying about proprietary software licensing issues.

If you have any questions about WSO2 Carbon product stack, please visit WSO2 Developer Portal and raise your voice in the WSO2 Carbon mailing list.

Apache Maven is now available in Eclipse

noreply@blogger.com (Harshana Eranga Martin) — Sun, 21 Nov 2010 07:06:00 +0000

I'm sure all most all of you know about Apache Maven. In a nutshell, Apache Maven is a software project management tool developed and distributed by Apache Foundation. It is widely used for project dependency management, project compilation and build artifact management.

Normal Maven tool is used with your operating system console (E.g: Command Prompt- In Windows and Bash Console - In Linux). But it is a serious limitation in a world where programmers work with Integrated Development Environments (IDEs) such as Eclipse, JIdea. Therefore these IDE providers are providing support for Maven in their IDEs. There is a Maven integration project under The Eclipse Foundation which provides the Maven integration to Eclipse IDE. This project is widely known as M2E or M2Eclipse. This project was initiated at Sonatype by the founder of Apache Maven project, Jason van Zyl.

M2E is managed under the Eclipse Tools top level project and it is still in the incubation phase. The project is led by Igor Fedorenko from Sonatype and the project summery is here. The latest M2E release is 0.12 and it is available at M2E p2 update site. This provides the Maven 3 support but it is still incompatible with Maven 2 related release including 0.10 release, If you are managing your project using Maven 2 please install M2E 0.10 release.

English Premier League 2010/11 Match Day Intro ESPN StarSports

noreply@blogger.com (Harshana Eranga Martin) — Sun, 31 Oct 2010 13:30:00 +0000

English Premier League 2010/11 Team Logo Animation

noreply@blogger.com (Harshana Eranga Martin) — Sun, 31 Oct 2010 13:17:00 +0000

Project epZilla research paper published at ICTer 2010 held at Colombo, Sri Lanka.

noreply@blogger.com (Harshana Eranga Martin) — Sat, 02 Oct 2010 11:06:00 +0000

After carrying out the successful research project "Project epZilla", we came up with a research paper which covers every aspect of the research including our system architecture, design, implementation and results. It was named as "Scalable Fault Tolerant Architecture for Complex Event Processing Systems". We were informed about the ICTer 2010 international conference to be held at Sri Lanka which is organized by University of Colombo School of Computing, by Mr. Sarves (Our internal supervisor) and Mrs. Vishaka Nanayakkara (Head, Dept of Computer Science & Engineering, University of Moratuwa), we decided to submit our paper to this conference. Our paper was selected to be published under the Full Paper category. Our paper presentation was scheduled to be held on 1st of October 2010 at Cinnamon Lakeside Hotel's Kings Court room. Dishan Metihakwala has presented the paper to the audience and it was our uttermost pleasure to present our paper in front of the large professionals including Prof. Athula Ginige from University of Western Sydney, Prof. Gihan Wicramanayake from UCSC, Dr. Elizabeth Sparrow - President BCS , Mrs. Vishaka Nanayakkara, Dr. Shahani Weerawarana, Dr. Shantha Fernando, etc. I, Rajeev Sampath and Chathura Randika were also among the audience and Dishan did a great presentation to the audience. In the Q&A session one question raised by the audience regarding how do we achieve fault tolerance in the system and Dishan tackled the question wisely and managed to give a complete answer for that question and later came to know that we are the only group who managed to tackle the questions raised by the same person during all 3 days of the conference. Several professionals among the audience came to us and praised the way Dishan did his presentation and tackled the question during the time we had our lunch. So the overall event was a great success for us and we will be celebrating this success in the future as well.

The published paper is now available in the ICTer 2010 web site at http://www.icter.org/UCSCConf/index.php/icter/icter2010/paper/view/90 and it will be indexed in the IEEE research publication section very soon since ICTer conference is sponsored by IEEE and all the papers presented at the conference are indexed as IEEE publications. I personally believe that University of Colombo School of Computing has done an outstanding job in organizing this whole event and wish them greater success in the future.

How to read a Research Paper?

noreply@blogger.com (Harshana Eranga Martin) — Sun, 15 Aug 2010 19:48:00 +0000

If you have involved in a research project, first thing you might have done is reading a pile of research papers. If you have done it properly, you might have encountered the hardest issue of identifying the papers which are going to be useful and interesting for you. Yeah it happened to me :) . I involved in a research project for last year and collected about 80-90 papers which contains relevant key words for my project. Obviously It is not practical or useful to read all these papers since most of them are not proving useful information and most of them are boring, But how can you identify these types of papers without wasting time on reading them and understanding it at the last? This document provides a fantastic solution for your problem. If you are going to involve in a research project, first read the above document. Then start reading the research papers.

FIFA 2010: From My Point of View Part 1: England

noreply@blogger.com (Harshana Eranga Martin) — Wed, 04 Aug 2010 18:24:00 +0000

It's true that I'm not an expert on Football. But still I have a good idea of the game and love the game a lot. Therefore I decided to express my opinion about the FIFA World Cup 2010 RSA.

FIFA Soccer World Cup is the most exciting sporting tournament in the planet. Lots of people do love football and lots of them watching and enjoying the game. This time it was hosted by the Republic of South Africa. This is the first time the biggest sporting event held in the African continent. Therefore there were lots of rumours that it is going to fail because RSA or any other African country is incapable of handling such a huge task. But as we all have seen, they have achieved and faced the tough challenge with great success. So all the Africans should be proud of it. We have witnessed a fantastic opening and closing ceremonies. Now towards the performance of the teams.

First I think it is better try to study a team at once. So first we start with one of the tournament favourites before the tournament kick off. Of course England ;).

I believe, they are one of the most unfortunate and most disappointing team in the tournament. If we look at their qualification round, they were in the same group with Ukraine and Croatia which are good in soccer and they managed to win 9 out of 10 qualifying games, loosing to Ukraine at the last qualifying round match, which was took place even after they qualified to South Africa for finals. In the qualifying game, England has shown some superb performance. Wayne Rooney scored 9 goals in the process and he was the second highest goal scorer for the UEFA region. Before going to south Africa, they had several international friendly fixtures with Mexico and Japan. They managed to win both fixtures with ease. These performance made them one of the tournament favourites.

Before going to RSA, they practised in Austria to get used to the high altitude conditions in the RSA. So it iis logical to assume, they had a good preparation for the tournament. Only issues they had was last minute injuries to Rio Ferdinand and Rooney's form after suffering an ankle injury at the UEFA champions league quarter finals against Bayern Munich. He was expected to get back to his form after scoring few goals.

First match was against USA and they had the dream start when Captain Steven Gerrard scored in to the 5th minute of the kick off. But every thing started to go backward when USA leveled when Robert Green made a mistake. There onward they lost the momentum in the tournament. There was no motivation or activeness in the pitch from them. They lost the possession frequently and surprisingly, they never even bothered to win the ball back. They expected rivals to come and hand over the ball to them. The midfield was disastrous. Lampard, Barry or Gerrard were not able to win the ball from the opponents in most games, specially against Algeria. Algerians were the better side on that day. They were able to keep the ball and pass the ball frequently and accurately. In contrast, England did not managed to keep the possession or the pass the ball around. Finally they managed to advanced to the knockout stage with the help of Defoe scoring against Slovenia.

In the knockout stage they faced the Germany, the best counter attackers in the tournament. As a result of the weak midfield, they managed to penetrate to the defence of the England so easily. Meanwhile English defence also followed the same thing as English midfield. John Terry and Matthew Upson were the central defenders but none of them were able to run with 32 years old Klose to stop scoring 1st goal. Then again few minutes later attack came breaching the English defence and non of them marked Podolski who was running on the left wing. Result was 2-0 up Germany. Finally England managed to score a goal in the 37th minutes Upson heading the cross of Gerrard. The build up process was done by the James Milner and Steven Gerrard in the right wing. Then England started to play with some confidence and attacked Germans more often. As a result They managed to Score another goal which was a chip by Lampard, first hit the cross bar and then pitching about 1 meter inside the goal post and eventually came back to the out side. But the Uruguayan Referee and Assists marked the most embarrassing moment of the whole tournament when they disallowed that goal. England Coach and his coaching staff saw the ball crossing the line from their dugout in the middle of the ground. But the referees who had more closer view, did not see it. It was utterly embarrassing when thinking, Referee and his assistant were the only people in the ground who did not see it while over 40,000 of the crowd saw it. I think it is the end for his refereeing career for the referee and his assistant. I sincerely do hope that for the sake of the game. From there onward England did not able to come back and ended up with their biggest defeat in the World Cup final.

Though there was a huge mistake by the Referee i believe England players should take the responsibility of their defeat because they really did not managed to impress. They did not perform in the same manner they did in qualifying. Some said they are tired. But it is their job to play football. They are not supposed to do anything else. So they should mange their schedule or whatever and should play with fullest determination for the country, So I think they let down their coach and fans.

I hope they will do better in the upcoming international friendly fixtures and Euro 2012.

Name of the next Eclipse Release

noreply@blogger.com (Harshana Eranga Martin) — Thu, 03 Jun 2010 10:48:00 +0000

Since now we are at the final step of the Helios release, now it is time for us to know about the name of the next Eclipse Release. It is "Indigo". "Eclipse Indigo". Yikes!! After reading up to this point you might be wondering like "Indigo?? really??".To be frank, I have the same feeling. It does not sounds like Calisto, Europa, Galileo. Helios. But most of the Eclipse committers think it is the best. So the next release of Eclipse will be called "Indigo".

You can download the latest stable builds of the Eclipse from here. Eclipse Helios will be release in 23rd of June along with 39 project including the Mylyn and ECF which are i'm also contributing. Please find more information about the Helios release from here.

Running RMI Applications in Eclipse IDE

noreply@blogger.com (Harshana Eranga Martin) — Sat, 08 May 2010 15:55:00 +0000

If you are using Eclipse IDE and develop RMI applications, the main problem you are facing is running and/or Debugging the RMI applications. If you have searched the Eclipse Plugin central, you may already know about the Genady RMI plugin for Eclipse. But it is not Free!! So here is a way to run RMI applications in Eclipse IDE without using any other Plugin.

1. Run => Run Configurations

2. Specify the main class in the Main tab

3. Specify "-Djava.rmi.server.codebase=file:${workspace_loc}/YourProjectName/bin/ -Djava.security.policy=file:${workspace_loc}/YourProjectName/Policy.txt" in the VMParameter section.

You have to replace "YourProjectName" in the above string with the name of your RMI Application. If you are not using a file based security policy you don't need to enter "-Djava.security.policy=file:${workspace_loc}/YourProjectName/Policy.txt". Pointer to Server code base location is sufficient in that case.

Before running the application using Eclipse IDE, you need to start the RmiRegistry. To do that you can start command prompt and type "rmiregistry" in the command prompt.

So now you can run and debug RMI applications using your Eclipse IDE without any problem.

Why I use Eclipse IDE?

noreply@blogger.com (Harshana Eranga Martin) — Sat, 08 May 2010 14:05:00 +0000

Almost all of us do our programming in Integrated Development Environments aka as IDEs. IDEs are making life easier for Programmers. IDEs have language specific editor, compiler, static code analysis tools, etc integrated in to one product and they work in a collaborative manner so that the integrated product users can access the functionalities of individual application in the complete product.

There are IDEs for many of the programming languages such as Microsoft Visual Studio for .NET application development, Eclipse for Java, C++, Php etc, JIdea for Java. Most of these IDEs are proprietary applications and the users have to pay some money to use it. As an example MS Visual Studio and JIdea commercial version. But Eclipse IDE is completely free of charge and it is open source IDE. So most of the developers use Eclipse for their application development because they can use it for Free. But it is not the ONLY reason for the huge customer base of the Eclipse IDE. It is very much user friendly, feature rich and the features pluggable while running the IDE.

Eclipse APIs are completely open and any programmer can access the APIs. These APIs are provided with respective projects and there are lot of guidance provided via examples by the project committers and consumers. We can ask questions in respective project mailing lists, IRC channels, News groups. If we have better ideas for projects, you can create an enhancement request in the Eclipse issue tracker aka Bugzilla. So any one can provide the ideas to the developers. Even anyone can develop applications as OSGi plugins and submit to projects.

Basically Eclipse provides every think I need for Free and it is very easy to use and full of features. So I use Eclipse IDE (JDT) for my Java development. Currently I'm working as a committer for Eclipse Foundation and I'm helping them to make the Eclipse IDE better. You also can become a contributor for Eclipse Foundation and help them. So Why not you?

A Happy and Prosperous Sinhala and Hindu New Year!!

noreply@blogger.com (Harshana Eranga Martin) — Wed, 14 Apr 2010 02:12:00 +0000

Wish you all a Happy, Prosperous, Wealthy and Healthy New Year!! ඔබ සැමට කිරියෙන් පැණියෙන් බත බුලතින් ඉතිරෙන නිදුක් නිරෝගී සම්පන්න සුබ නව වසරක් වේවා!!

Simple Financial Operation Simulator

noreply@blogger.com (Harshana Eranga Martin) — Wed, 31 Mar 2010 12:43:00 +0000

This week i have been working on a Simple financial operation simulator. I completed this in few hours and may contain lot of programming malpractices. But i was interested in completing it because it uses a xml based schemas for information storage. This is the first formal application i have worked on this type of approach. This is an example for a theory we learnt under Advanced Databases course to use XML based databases. So if you are not very much familiar with xml based database schemes, this would be a good example for you.

Here is the code. Codebase is published under EPL 1.0. You are abide by the laws of IP and terms of EPL license.

Java Remote Method Invocation - Part 3

noreply@blogger.com (Harshana Eranga Martin) — Sun, 07 Mar 2010 07:47:00 +0000

In this post I'm going to discuss about another useful thing about Java RMI. The topic I'm going to cover in this post is running the Java RMI when the byte code is packaged to a jar file. Basically this means running RMI when the source code is compiled and packaged to a java archive (jar).

In this case also, the basics we have discussed so far applies and have to follow the same steps.

1. Start the RMI registry.

2. Start the RMI Server.

3. Start the client.

Example commands to run the Server.

java -classpath C:\Eclipse\LCRRMIServer\target\LCRRMIServer.jar -Djava.security.policy=C:\Eclipse\LCRRMIServer\target\policy.txt -Djava.rmi.server.codebase=file:/C:/Eclipse/LCRRMIServer/target/LCRRMIServer.jar org.epzilla.leader.rmi.LeaderRMIServer

In this command what we are doing is, first we set the jar file to the classpath. We do this even we run using the batch script. Then we specify the security policy and next we say where is our RMI interface can be found. In this case it is the jar file. So we set the codebase path to the jar file. In this case we do not specify "/" at the end of the codebase path since that uses to specify the root folder of the compiled source and in this case it is the jar file and we have given the jar file path. So do not include that "/" because it will specify a wrong path for RMI Interface and it will lead to a ClassNotFoundException and UnmarshalError.

Running the client can be done using same command we have used before. Here is the example command for the client.

java -classpath C:\Eclipse\LCRRMIServer\target\LCRRMIServer.jar org.epzilla.leader.rmi.LeaderMessageClient

In this also we set the classpath to the jar file and run the client.

Why don't we just dance and The bourne trilogy

noreply@blogger.com (Harshana Eranga Martin) — Mon, 01 Mar 2010 20:37:00 +0000

Great song which i heard in the Hank FM via ITunes...

This is the theme song of the Bourne Trilogy...

Both from youtube.com

Generating Javadoc for your project using Eclipse

noreply@blogger.com (Harshana Eranga Martin) — Mon, 22 Feb 2010 19:15:00 +0000

Java language API is usually exposed using the java documentation html files which we call as Javadocs. Eclipse provide the facility to create the javadoc for your java project. It's very easy but i found that most of the Eclipse users do not know how to do it. Follow these simple steps and generate your javadoc for your API using Eclipse.

1. Project-> Generate javadoc

Now you will get the following window.

2. Select the path of the file "javadoc.exe" which comes with your jdk at the "Javadoc Command" text box. For me, it is "C:\Program Files\Java\jdk1.6.0\bin\javadoc.exe".

3. Just click finish and generate or you can continue to so more customizations using next. If you want to make more and more adjustments, then read the instructions carefully and move forward with next.

4. When you click on finish, Eclipse will execute the javadoc command for you and generate javadoc for your API in few seconds.

Java Remote Method Invocation - Part 2

noreply@blogger.com (Harshana Eranga Martin) — Sat, 20 Feb 2010 20:02:00 +0000

In my last post on Remote Method Invocation, I have explained how to run the simple RMI example of printing hello world which was given in the SDN Getting started article.

In that article I have shown how to run the hello world example with the example commands to be given to the console. In this article I'm going to explain about a bit advanced example. This example is used to a Text file from Server to Client via Java RMI.

You may find this example article also from the SDN Developer article section.

In this File Transfer example also I'm using the namespace org.epzilla since it is my project's namespace. But the article does not use any namespace. So please be careful when you are following the steps. Make sure to customize the instructions appropriately.

This application is consists of 4 items. They are

1. FileInterface - Remote Interface

2. FileImpl - Remote Interface Implementation

3. FileServer - RMI Server

4. FileClient - RMI Client

Document gives a solid understanding of how to implement these things and why we do it in those ways.

Apart from the above items, we need an extra item for this ( and many more) application. That is Java Security Policy document. Example Java Security policy implementation also given in the document. So there won't be any trouble. But that security policy is "All in and Anything allowed" type, so the security is very weak. If you are using this code for some enterprise application, I strongly recommend you to use some tighter security policy.

As usual complete the implementation of the source code by following the example document.

Step 1

Go to your source directory from Console.

Step 2

Compiling the source code.

javac -d destDir SourceFile1.Java SourceFile2.java so on

E.g.:

javac -d C:\test FileInterface.java FileImpl.java FileServer.java FileClient.java

Output: Class files will be created at the given destination "c:\test". This is my destDir and classDir.

Step3

Go to the c:\test folder from the console

Create the stub using following command.

rmic namespace.RemoteInterfaceImplementation

E.g.:

C:\test> rmic org.epzilla.FileImpl

Output:This has created a file as FileImpl_stub.class

Step4

invoking RMI Registry

start rmiregistry

Output: a new blank console

Step5

Place your policy.txt file in the classDir. In my case place the policy.txt file in c:\test directory.

Step6

Running the Server.

start java -classpath classDir -Djava.security.policy=classDir\policy.txt -Djava.rmi.server.codebase=file:classDir/ namespace.Server

E.g.:

start java -classpath c:\test -Djava.security.policy=c:\test\policy.txt -Djava.rmi.server.codebase=file:c:\test/ org.epzilla.FileServer

Output: New Console will be open, If you haven't change the example code, it is blank. If you printed a message on successful deployment, it will show.

You may run this command from any place/directory in the console.

Step7

Running the client

java  -classpath classDir namespace.Client

E.g.:

java -classpath C:\test org.epzilla.FileClient

Output: Since I have changed the source code bit with some hard coded file to be downloaded to client, I don't pass the filename and machine as argument in this program.

This command also can be executed from any where in the console.

Things to keep in mind when working with RMI:

1. When compiling the source code, you need to be at the exact location of source code.

2. What ever the place you name as your Destination Directory(destDir), you always deals with this place.

3. When generating stub using rmic, you need to be there at the destDir in console.

4. Place your security policy document also in the destDir.

5. Launching RMI Registry, RMI Server, RMI Client can be done from any place in the file system. All you need is setting the classDir(destDir) correctly.

If you follow these instructions, you can write and execute your Java RMI applications as easy as executing normal J2SE application.

Eclipse ECF 3.2 Release Available

noreply@blogger.com (Harshana Eranga Martin) — Fri, 19 Feb 2010 19:26:00 +0000

Yesterday( 19/02/2010) we(ECF community) have released our latest version of ECF. Now you can get it from the [1] location.

This version of ECF supports only Eclipse Galileo ( 3.5.x) and Eclipse Helios M5(3.6M5) versions and higher versions only. The release was originally scheduled on 23/06/2010 along the release of Helios. But we thought it would be better if we can release this minor version of ECF before the API freeze for the Helios version and after this release also we can continue some implementation, API changes till the end of April 2010. So we have scheduled this release to be launched on 19/02/2010.

The major changes and contributions for this release are

1. Initial implementation of OSGi 4.2 Remote Services specification.

2. SOAP and REST APIs for ECF Remote Services.

3. Load balancing for ECF Remote Services using JMS queue via ECF ActiveMQ provider implementation of JMS.

4. New Remote Services examples.

5. Initial implementation of Google Wave Provider for ECF.

You may find more details from ECF Wiki[2] and ECF Blog.[3]

Many thanks should go to ECF Project Lead and all the committers and contributers.

courtesy:

Mr. Scott Lewis, ECF Project Lead

[1]. http://www.eclipse.org/ecf/downloads.php

[2]. http://wiki.eclipse.org/ECF

[3]. http://eclipseecf.blogspot.com/