Thursday, September 27, 2012

Creating OSGi Bundles with Maven Bundle Plugin

OSGi is a framework specification to create java applications based on a complete and dynamic component model. That means your application can be entirely consisted of cohesive and loosely-coupled modules. These modules are called "bundles" in OSGi jargon.

There are several implementations including Apache Felix and Eclipse Equinox. These implementations use existing capabilities of Java Archive (jar) files to implement OSGi bundles. If you look inside a general jar file (which is basically a zip file with .jar extension), there will be a file called MANIFEST.MF inside a directory named META-INF. If the jar was built by maven it will contain some meta data like: 
Manifest-Version: 1.0
Archiver-Version: Plexus Archiver
Created-By: Apache Maven
Built-By: amilas
Build-Jdk: 1.6.0_35


There are manifest headers specific to OSGi so that runtime can recognize the jar file as a bundle ( and more). Bnd is a tool used to insert those headers into jar files.

While there's built-in support for OSGI in famous IDEs such as Eclipse and IntelliJ IDEA, if your project is maven-based, it's convenient to use maven for creating bundles too.
Maven Bundle Plugin can be used for this purpose. We can specify OSGi parameters in the module POM and the bundle plugin will use Bnd to automatically insert them and create an OSGi bundle.

This is a introductory post to show how to use Maven with IntelliJ IDEA to create OSGi bundles.

Goto File-> New Project... and select "Create  project from scratch"
At the next screen, enter give a project name and a location as shown. Make sure to select the type as Maven Module.



Finally enter a group id as shown, and click finish.

This will create a new project and a module from maven archtype.

Now let's create two sub modules for our project. these will be the resulting OSGi bundles.

Goto File-> Add Module... again "Create  project from scratch" and name the module as "module-a". In the final screen "Add module to" and "parent" fields should be automatically filled as shown.


Similarly add another module named module-b

 Create two packages "api" and "impl" inside org.amila.sample.osgi.a in module-a, and similarly in module-b so the project pane will show the project structure as follows.

You can use Bundle activator to perform some tasks when the OSGi framework starts or stops a certain bundle. We can implement org.osgi.framework.BundleActivator interface to utilize this function.
Create a new class Adapter in your impl package in module A. Use the following code:

package org.amila.sample.osgi.a.impl;

import org.osgi.framework.BundleActivator;
import org.osgi.framework.BundleContext;

public class Activator implements BundleActivator {

    /**
     * Implements BundleActivator.start().
     * @param bundleContext - the framework context for the bundle.
     **/
    @Override
    public void start(BundleContext bundleContext) throws Exception {
        System.out.println("Module A is starting");
    }

    /**
     * Implements BundleActivator.stop().
     * @param bundleContext - the framework context for the bundle.
     **/
    @Override
    public void stop(BundleContext bundleContext) throws Exception {
        System.out.println("Module A is shutting down");
    }
}

Add the felix dependency to the module POM.


Set the packaging to "bundle" and add the maven bundle plugin.

After these changes, pom.xml of module-a should look like this:

<?xml version="1.0" encoding="UTF-8"?>
<project xmlns="http://maven.apache.org/POM/4.0.0"
         xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
         xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
    <parent>
        <artifactId>OsgiDemo</artifactId>
        <groupId>org.amila.sample.osgi</groupId>
        <version>1.0</version>
    </parent>
    <modelVersion>4.0.0</modelVersion>
    <artifactId>module-a</artifactId>
    <packaging>bundle</packaging>

    <build>
        <plugins>
            <plugin>
                <groupId>org.apache.felix</groupId>
                <artifactId>maven-bundle-plugin</artifactId>
                <version>2.3.7</version>
                <extensions>true</extensions>
                <configuration>
                    <instructions>
                        <Bundle-SymbolicName>${project.groupId}.${project.artifactId}</Bundle-SymbolicName>
                        <Bundle-Name>${project.artifactId}</Bundle-Name>
                        <Bundle-Version>1.0.0</Bundle-Version>
                        <Bundle-Activator>org.amila.sample.osgi.a.impl.Activator</Bundle-Activator>
                        <Private-Package>org.amila.sample.osgi.a.impl</Private-Package>
                   </instructions>
                </configuration>
            </plugin>
        </plugins>
    </build>
    <dependencies>
        <dependency>
            <groupId>org.apache.felix</groupId>
            <artifactId>org.osgi.core</artifactId>
            <version>1.4.0</version>
        </dependency>
    </dependencies>

</project>

Similarly add Activator and bundle configuration for module B.

 Now, you can build the project using maven. Resulting jar files will be OSGi bundles. If you open MANIFEST.MF file of a bundle now it will contain bundle headers as we configured in the module POM.

You don't have to specify packages to import. If your module uses a package from another module, its import header will be automatically added to the manifest.mf file.

Sunday, September 2, 2012

Obtaining a List of IP Addresses

During our final year project I wanted to get active ip addresses of a node. This is easily doable using java.net.NetworkInterface
The method below returns a list of IP addresses of a machine except those belonging to loopback or inactive network interfaces. The addresses can be further classified as IPv4 or IPv6 using instance check with java.net.Inet4Address and java.net.Inet6Address
    public static List<InetAddress> getIPAddress() throws SocketException {

        List<InetAddress> ipAddresses = new ArrayList<InetAddress>();
        Enumeration e;
        e = NetworkInterface.getNetworkInterfaces();
        while (e.hasMoreElements()) {
            NetworkInterface ni = (NetworkInterface) e.nextElement();
            if (ni.isLoopback() || !ni.isUp()) continue;

            for (Enumeration e2 = ni.getInetAddresses(); e2.hasMoreElements(); ) {
                InetAddress ip = (InetAddress) e2.nextElement();
                ipAddresses.add(ip);
            }
        }
        return ipAddresses;
    }

Sunday, July 1, 2012

Motion Controlled 3D Flight Simulator

I came across this today while going through my PC's hard drive. This is an exhibit we made back in 2010 for exhibition "Exmo" held in University of Moratuwa. It was also exhibited in Techno, Infotel and Deyata Kirula Exhibitions. This kind of things may be quite common these days. However I'm happy that this was an original idea at that time.


Since we had a limited amount of time, we had to reuse some of the segments. 3d world is mostly based on Riemer's tutorials. A 2D array is used to create the buildings of the city. Heightmap technique is used to model the terrain. It looked quite good with mixed textures, flare and water effects.



The program processes webcam feed frame by frame. We used blob counter technique along with several filters to locate an area of predefined color, with the help of AForge. The rectangle covering that area is considered and its center point is calculated. The displacement of this point relative to center of the captured screen is used to maneuver the aircraft in the game. 

Source Code


Written in C#. XNA is used as the gaming framework. AForge is used as the image processing library. I'm sharing the source hoping someone may find it useful, though its quality and readability is not very high.
Install XNA 3.1 and AForge 2.1. You should use Visual Studio 2008.


Get the source code here:
https://github.com/amilamanoj/FlightSim

Dependencies


You need Windows and following installed for this to work

  • .NET Framework 3.5
  • XNA 3.1
  • AForge 2.1

Installers 


I've made an installer with above dependencies. XNA and AForge are embedded in the installer. .NET Framework isn't embedded due to its size. It will be automatically downloaded and installed. You can get the installer here. (~50MB)


If you have these components already installed or prefer to install them manually, use this installer (~17MB)

Playing


Aircraft can be controlled using mouse, keyboard or movement of an object (or your hands with gloves).
Default controller is mouse. You can switch between controllers using Tab during gameplay. Press Q to quit during gameplay.
You can select the webcam and configure the color in options screen. Default color is red. For example, you can use a red colored plastic ball.
If you haven't played a flight simulator before, you should try playing with mouse or keyboard before trying with motion.

Friday, May 25, 2012

Compiler Construction and JavaCC

Scanning and Parsing are two important phases of compilation process.While it is possible to code a scanner and a parser all by hand, it becomes tedios, specially when the grammar of the language gets more complex. There are Lexer and Parser generators so that we can generate them from a specified grammer.

JavaCC is one such compiler generator. Here are key points about JavaCC.
  • It can generate both scanner and parser.
  • TopDown parsing: generate the parse tree in a top-down manner.
  • It generates Recursive Descent parsers. The parser will execute a set of recursive methods to process the input (not based on parse tables)
  • Predictive (non-backtracking). It looks ahead a number of symbols to decide which grammar rule to use.
  • Supports LL(K) grammars. That means there must not be left recursion, and the grammar should be left refactored.
  • Detects left-recursion, possible ambiguities and warns during compiler generation.
  • Can use different look-ahead values in different local sections
  • Can use syntactic look-ahead. Hence can be considered as a LL(*) parser generator.
  • Gammar can be provided in EBNF notation. Removing left recursion is easy and the grammar is more readable.
  • Provides debugging support: tokenizing / parsing / looking ahead.
  • Supports panic-mode error recovery.
  • Supports Unicode.
  • Lots of good documentation / resources to follow.
  • Good tooling support, including an Eclipse plugin, IntelliJ IDEA plugin and a Maven plugin.
Having unicode support means you can build your compiler using any (natural) language you prefer! You'll only have to change the (programming) language keywords for the new (natural) language. You can also change the (context-free) grammar a bit if you want to make your (programming) language closer to the new (natural) language.

Here is my attempt to create a compiler in my own language, සිංහල (Sinhala). This uses a simplified version of c grammer.
https://github.com/amilamanoj/CMinus/



Recording Music from a Musical Instrument

You can record high quality audio by direct cable connection between your computer and musical instrument (such as musical keyboard, electric guitar).

You need a two way audio cable. (with 3.5mm TRS connectors). Just connect it to audio out of the instrument and line-in/mic of the computer.
Most of the instruments have a 6.35mm socket for audio output. In that case you also need a 6.35mm male to 3.5mm female jack stereo audio adapter. You can get these from a local electronics store or a site like amazon, ebay.

Moving on to software, you don't need expensive professional sound editing software. Audacity is a free and open source cross-platform audio recorder/editor and perfectly fits our purpose.
 
Next comes the tricky part. During recording, you hear what you play through the computer. If there's a delay between what you play and what you hear, it's very difficult to play. Let's look at solutions for this.

Windows

When it comes to Windows, drivers pass sound through the system kernel, resulting considerable amount of latency.
This won't be a big issue on sound cards that support ASIO. ASIO is a sound card driver  protocol that minimizes the latency by bypassing operating system layers and connecting directly to the soundcard. However, not many soundcards support ASIO (specially integrated audio chipsets).

ASIO4ALL may help you in this case. It is an independent universal emulated ASIO driver and brings ASIO support to virtually all consumer-grade soundcards and integrated audio chipsets.
The problem here is that, you need a recording program that supports ASIO. Though Audacity can support ASIO, they don't distribute it with ASIO support since it's proprietary and has an incompatible license. You maybe able to find an unofficial version with ASIO support to download, or if you are a software geek, you can compile Audacity yourself with ASIO support.

Linux

Most modern distributions (Ubuntu, Fedora, Mint, OpenSUSE)  use PluseAudio with ALSA. Usually it results in lower latencies. If you are lucky, you may be able to record without any tweaking.
Within audacity, just goto Edit -> Preferences -> Recording and select Software Playthorugh.

If you experience some delay, you can route the input at PulseAudio level by engaging loopback module. Goto terminal and execute:
pactl unload-module module-loopback
In my experience, this can also have some latency, and if your computer (laptop) has a built in microphone, you might get a loud noise.

What worked best for me is using the pacat command. What we do here is capturing the raw audio steam from the specific device we want and playing it back from the output device on the fly, specifying minimum latency. You can quickly get this working by following these instructions. Make sure you have selected "pulse" as both playback and recording device in Audacity Preferences -> Devices, otherwise you won't be able to record while pacat is active.

After you have recorded, you can easily get the mp3 by using Audacity's Export command.

Here is a recording I made. This is a part of the song "Never Forget" in Halo3. Played with Yamaha PSR-E323. Recorded with Audacity in Linux with the last technique above.
Never Forget (cover) by Amila Manoj

Wednesday, April 25, 2012

AS2 Protocol

AS2 is a protocol that describes how to exchange structured business data securely using the HTTP transfer protocol.
Structured business data exchanged using AS2 protocol can be,
  • Electronic Data Interchange (EDI) in either the
    • UN Electronic Data Interchange for Administration, Commerce, and Transport (UN/EDIFACT) format or
    • The American National Standards Committee (ANSI) X12 format
  • XML or any other structured data formats.


Major application of AS2 is to exchange data in EDI formats.

Security is achieved using digital certificates and encryption. Exchanged messages can be signed using to provide security requirements such as authenticity and preventing non-repudiation. They can be also encrypted to provide confidentiality and integrity. Those are optional requirement according to AS2 specification. An AS2 message can have its content in plain text and without a digital signature.

S/MIME is a format and protocol for adding cryptographic signature and/or encryption services to Internet MIME messages. In AS2, files are encoded as attachments in an S/MIME message. This is what we call an AS2 message. Those messages are sent using the HTTP or HTTPS, usually as POST.



Content Types and AS2

 

There can be several content types for AS2 messages.
  • When there is no encryption, no signature
    • application/EDIxxxx or application/xml
  • When there is no encryption, but signature is present
    • multipart/signed – message contains two MIME parts
      • application/EDIxxxx or application/xml
      • application/pkcs7-signature
  • When there is Encryption, but no signature is present
    • application/pkcs7-mime
      • application/EDIxxxx or application/xml (in the decrypted message)
  • When there is both encryption and signature
    • application/pkcs7-mime
      • multipart/signed(encrypted) – decrypted message contains 2 parts
        • application/EDIxxxx or /xml)
        • application/pkcs7-signature)

Usually AS2 clients are called "Trading partners". When sending a message, they can request an acknowledgement message called MDN (Message Disposition Notification).
There are several options of requesting a MDN. They are Synchronous MDN, Asynchronous MDN and No MDN.

If there are problems receiving or interpreting the original AS2 message, a "failed" MDN may be sent back. Both "failed" MDN and not receiving MDN (when it is requested) are considered as failures according to specs.

Mendelson (GPL) and OpenAS2 (BSD) are open source implementations of AS2 for java, These can be used to send and receive AS2 messages. Specifically, Mendelson has a user-friendly GUI.

Sunday, March 18, 2012

Ruby Notes

I recently started learning Ruby as a part of "Software Engineering for Software as a Service" online course offered by  University of California, Berkeley.

Since I am mostly experienced with java, I found some differences to be interesting.
  • Ruby is object oriented, similar to Java. 
    • But everything is an object. Even primitive types such as integers are objects. 
    • Almost everything is a method call on some object. Even most operators are instance methods. You can do method calls on anything. 
      • 6.methods will return a list of methods that it will respond to.
      • 1+2 means we pass the objects operator + and the number 2 to the send method of object 1. 1.send(:+, 2)
  • Dynamically typed
    • Even though objects have types, variable don't have types. 
    • Also, there are no declarations either
    • That means, you just can use a variable without declaring first, and use it to store any type of object.
  • Identifier Conventions
    • Class names should UpperCamelCase similar to Java
    • But methods and variable names should use snake_case unlike java's camelCase
  • Ruby is pass-by-reference, since everything is an object. (whereas Java is pass-by-value.)
  • Metaprogramming
    • this basically means you can do some of the programming during the runtime, such as method definitions. 
    • Say there's an instance variable name. Instead of writing it's getter and setter, you can specify to create them during runtime using metaprogramming by: attr_accessor :name
  • Iterators
    • Iterations play a big role in Ruby. 
    • In Java, we usually run a loop with an index and do something with the object for the index in each iteration.
    • In Ruby, iterating with an index is discouraged. Rather, we let objects manage their own traversal.
    • my_array.each do |elt| { } end
  • Duck Typing
    • An object's current set of methods and properties determines the valid semantics, rather than its inheritance from a particular class or implementation of a specific interface.
    • For example, you can call sort on arrays with different type of objects, strings, hashes, if they respond to the method somehow, without considering their types.
  • Mix-ins
    • This is used to achieve duck typing.
    • In Ruby, there are things called Modules, A module is a collection of class & instance
      methods that are not actually a class. Therefore you cannot instantiate it
    • But by including modules in your class you can resuse (mix) their methods
    • class A &lt; B ; include MyModule ; end . A.foo first search A, MyModule and finally B.
Here are some of the resources for getting started with Ruby:
  1. Ruby in Twenty Minutes
  2. Try Ruby
Ruby Installer is a nice packaging that makes it easy to install Ruby on Windows

Wednesday, February 1, 2012

Using Apache Thrift with Maven

When you use Thrift for your project, you have to manually generate the sources, put them inside your source folder and build.

This plugin does the work for you when you execute the maven build.

However, this plugin is not yet available on maven central repositories, therefore you have to add developer's repo. to your project pom.

 

Then, as described in the plugin's project page, you have to:

1. Use Java 1.5 or newer due to the usage of Generics

You can specify it in the configuration of maven compiler plugin:

 


2. Have Thrift executable is in your PATH or set the parameter of the plugin to the correct location.

 

3. Include the dependency for libthrift

 

Then place your *.thrift files to the directory: src/main/thrift.

Now you can execute mvn clean install as usual.

Saturday, January 14, 2012

Volunteer Computing: An Introduction

As the technology advances and research areas widen, demand for computational power increases day by day. Such computational demands can be mainly observed in several categories. Physical simulations from molecular level to universe level, analysis of large data from optical telescopes, gene sequencers, gravitational wave detectors, particle colliders and biology-inspired algorithms are some of those categories.
These tasks require high-performance computing (HPC). So one solution is to use supercomputers. But typically, the rate of job completion is more important than the turnaround time of individual jobs since overall result is what’s useful. The term to refer to that idea is high-throughput computing.

To achieve high-throughput computing, distributed computing is a better approach since individual job can be processed in parallel in large quantities.
Available distributed computing options are:
○ Cluster computing - dedicated computers in a simple location.
○ Desktop grid computing - PCs within an organization as a computing resource.
○ Grid computing - sharing computing resources by separate organizations.
○ Cloud computing - a company selling access to computing power
○ Volunteer computing

Volunteer computing (also sometimes referred to as global computing) uses computational power volunteered by the general public to perform distributed scientific computing. Volunteers may include individuals as well as organizations such as universities.
This approach allow ordinary Internet users to volunteer their computer resources on idle time by forming parallel computing networks easily, quickly and inexpensively without needing expert help.

Typically when it comes to volunteer computing, the volunteers who contribute with their resources are considered to be anonymous, although some volunteer computing frameworks may collect information like a nickname and email address of the volunteers for the usage of credit system, etc.

Each of the distributed computing paradigms have different resources pools. For example, number of computers owned by a particular university when it comes to grid computing, and the number of servers owned by a company in cases of cloud computing. The number of total possible personal computers is the resource pool in the case of volunteer computing.
To understand the importance of volunteer computing, we have to consider its resource pool.
The number of privately-owned PCs around the globe is currently estimated as 1 billion and is expected to grow to 2 billion by 2015. Also, the resource pool is self-financing, self-updating and self-maintaining. Users buy and maintain their own computers. Therefore various costs associated with other types of grid computing do not apply to volunteer computing. Another important point is that consumer market adopts the latest technology quickly. A supercomputer or a computing grid cannot be replaced or upgradedeasily as newer technologies emerge. But the typical PC user can. For example, the fastest processors today are GPUs developed with computer games in mind. Due to these factors, we can state that volunteer computing has a huge potential for world computational needs.
Berkeley Open Infrastructure for Network Computing (BOINC) is the predominant volunteer computing framework in use.
Some of the other volunteer computing frameworks are:
○ Bayanihan Computing Group
○ JADIF - Java Distributed (volunteer / grid) computing Framework
○ Javelin Global Computing Project
○ XremWeb Platform
○ Entropia

Here is a list of most active volunteer computing projects as of January 2012.
SETI@home: Search for extra-terrestrial life by analyzing radio frequencies emanating from space
Einstein@home: Search for pulsars using radio signals and gravitational wave data
World Community Grid: Humanitarian research on disease, natural disasters, and hunger
Climateprediction.net: Analyse ways to improve climate prediction model
Folding@home: Computational molecular biology
LHC@home: Improve the design of the Large Hadron Collider and its detectors
Milkyway@home: Create a highly accurate three-dimensional model of the Milky Way galaxy using data collected from the Sloan Digital Sky Survey
Spinhenge@home: Study nano-magnetic molecules for research into localized tumor chemotherapy and micro-memory
PrimeGrid: Generate a list of sequential prime numbers, search for particular types of primes
Malariacontrol.net: Simulate the transmission dynamics and health effects of malaria

(This post includes citations from several sources and aims to summarize volunteer computing)

Friday, January 6, 2012

First Steps of Apache Thrift with Java in Linux


Apache Thrift is a software framework for scalable cross-language services development. It was originally developed by Facebook before it was donated to Apache Software Foundation.  

Download the stable release

Unpack the tar.gz archive to a directory you prefer
(say home /home/amila/apacheThrift)


You need JDK and Apache Ant at least to run Thrift's Java tutorial.
(You can refer to my previous posts to find how to install JDK on ubuntu.)

Use apt-get to install Ant

 

We first need to install Thrift compiler and language variables before we start developing we Thrift.

There are several required packages to install Thrift that are not installed on a linux distribution by default.
To install those
 


Goto top level directory of unpacked thrift distribution
(eg: /home/amila/apacheThrift/thrift-0.8.0)
 
 

During this process, thrift will scan and list the different language found.
It should say:

..along with other languages found on your computer.

However, to configure Thrift for all those languages, you may need to install additional packages

Now you can make Thrift:
 

You might get some error if all required libraries for the languages configured in above step are not present.

In that case, you can deselect the packages you don't need when configuring
For example, say you don't need the support for Ruby. When configuring, you can use:
 
(I had to deselect erlang libraries to get it working on ubuntu 11.04)

After make is completed successfully, install Thrift by,

 

To check if the installation is successfully completed:
 

You should get an output like:
 

Tutorial are located at ./tutorials directory.

There you will find two files tutorials.thrift and shared.thrift

.thrift files describe the interfaces (IDL) in terms of the classes, methods they include.


 

This will create a directory named "gen-java" inside your current directory which will include generated Java classes according to specified thrift file.

Now goto the directory "java" inside the current directory (tutorial) and execute ant.
The ant script will compile both generated source files and the source file inside java directory and build a jar file.

Finally, run the tutorial by:
 

You may also find this page useful.