Customized event in Flex 2

Events are a way of sharing data between two components. Events also acts as a asynchronous message suggesting that some input has arrived to the receiver of the event. For e.g. when a person clicks on the button control the event click is generated signifying that a user input in the form of a click has occurred. Once the event has occurred we need to take some action like on button click display the “hello world” in a text box. The action that we take is known as handling of the event. Or in other words we need to specify the even handler. Event handler is a function that specifies what needs to be done on that particular event.

The following code snippet makes it clear

<?xml version="1.0" encoding="utf-8"?>

<mx:Application xmlns:mx=="http://www.adobe.com/2006/mxml"

layout="vertical">

<mx:Script>

<![CDATA[

//Event handler for the button click function

private function buttonEventHandler(evt:Event):void{

taTest.text="Hello World";

}

]]>

</mx:Script>

<mx:Button id="btTest" label="Click Me" click="buttonEventHandler(event)" />

<mx:TextArea id="taTest" />

</mx:Application>

buttonEventHandler is the event handler that will be called when the button is clicked. One salient point to note is the argument passed in the handler. We pass the object of type Event i.e event. The object “event” is always available in the application and we need to use it at the time of calling an event handler.

Sometimes it happens that we need to transfer some data between the two components.

For e.g. there may be a component which is responsible to get the data from the server using a httprequestserver. Component populates the data in a collection and now wants to send to another component which implements the list or data grid to display the information to the end user. The events are the way through which this can be done. The catch here is that the event thrown by one component should be available in the other component. So the component1 when it is done with populating the collection will throw an event indicating which needs to be handled in the component2 responsible to display the collection in the datagrid. It happens because of the way the events are transmitted from the component to the Application. Before getting into the transmission concept we need to understand one more concept known as event listener. Event Listener signifies that the component wants to receive a particular event and so it wants to register itself as one of the recipients.

We use the function addEventListener(event name , event handler) to do that. Now you must be wondering that the above button code did not had any such function call then how were we able to handle the event. The simple answer is that the event that we talked about was provided by the flex framework and so every component has got an implicit event listener for such events.

But the events that we are now going to see are the one that user has created himself (customized events). The customized events are used to transfer the data. These are created by the users and so the components will not be aware of them. Event Listener explicitly registers the component to listen to these customized events.

HOW TO WRITE A CUSTOMIZED EVENT

There are 3 stages in order t use a customized event

1. Create customized event

2. Dispatch the customized event

3. Listen the customized event.

Create Stage

Let us see how we actually write a customized event. We are just discussing the salient points here which are required to write a customized event. Syntactical and theoretical details are well covered in the Adobe documents.

Customized event can be declared by writing a class that inherits from Event.

package controls

{

import flash.events.Event;

public class ItemAddedEvent extends Event

{

var itemDescription:String; //an item to be added to the cart

public static const ITEMADDEDEVENT:String ="ItemAddedEvent";

public function ItemAddedEvent(description:String)

{

super(ITEMADDEDEVENT,true,true);

itemDescription=description;

}

override public function clone():Event{

return new ItemAddedEvent(itemDescription);

}

}

}

These are the steps that need to follow for defining any customized event

1. The class has to extend Event

2. Declare what ever data type you want to transfer by this event.

var itemDescription:String; //an item to be added to the cart

3. One can give a name to the event for easy reference. In our case it is ITEMADDEDEVENT. We can give multiple names to the same event. This is required when we want to use the same event in different places. For e.g. in a shopping cart application we need to perform add item to cart, delete item from the cart and update the quantity of the item in the cart. In all the operations the same item type data is required to be passed. So we can define a single customized class that favors for the transfer of a item type data. But the same event class can be used for all the 3 functions. We will give 3 names so that we remember for what all purpose we are calling this class.

public static const ADD:String = "addtocart";

public static const REMOVE:String = "removefromcart";

public static const UPDATE:String = "updatecart";

4. In the Constructor we need to call super(). The parameters required to be passed are the type of the event (ADD,REMOVE,UPDATE) and the optional parameters specifying the properties of the event.

Super(type, isbubble,iscancellable)

The second argument signifies whether the event is allowed to bubble. i.e. an event thrown by a component is allowed to be listened by the components that are up in the hierarchy.

5. clone() . This is required as per the Adobe document. We need to override the clone function and in the body just instantiate the customized event and return the object so created.

override public function clone():Event

{

return new ItemAddedEvent(itemDescription);

}

That is all to be done for a creation of an event.

Dispatch stage:

The event is thrown by the component in this stage. There is a well defined procedure to be followed to do that.

Lets take an example and do that

<?xml version="1.0" encoding="utf-8"?>

<mx:VBox xmlns:mx="http://www.adobe.com/2006/mxml">

<mx:Metadata>

[Event(name="addItemEvent", type="controls.ItemAddedEvent")]

</mx:Metadata>

<mx:Label text="Item name:"/>

<mx:TextInput id="enteredItem"/>

<mx:Button width="104" height="28" cornerRadius="10" fillColors="[#00ff00, #00B000]" label="Add Item" fontSize="12" click="greenClickEventHandler()"/>

<mx:Script>

<![CDATA[

import controls.ItemAddedEvent;

private function greenClickEventHandler():void{

trace("Ouch! I got clicked! Let me tell this to the world.");

// throws the event and the passes the string to the customized event object

dispatchEvent(new ItemAddedEvent(enteredItem.text)); // passing the string

}

]]>

</mx:Script>

In the example

<mx:Metadata>

[Event(name="addItemEvent", type="controls.ItemAddedEvent")]

</mx:Metadata>

The metadata section is an optional section that tells that these are the events this component is going to dispatch. It is more for giving the information to the developer that this particular code dispatches some event. This is useful since the other components by looking at the metadata section can find out the events they can listen to.

We give the name of the event (addItemEvent) and the location of the class (controls.ItemAddedEvent). Note we could have also given ITEMADDEDEVENT in place of addItemEvent.

private function greenClickEventHandler():void{

trace("Ouch! I got clicked! Let me tell this to the world.");

// throws the event and the passes the string to the customized event object

dispatchEvent(new ItemAddedEvent(enteredItem.text)); // passing the string

}

When the greenClickEventHandler() is getting called on the click event the component is calling the dispatchEvent function. By calling this function the component is actually throwing the customized event. The dispatchEvent function requires the object of the Customized event class as an argument. So essentially we have passed the data structure (in this case string) to the event. This data will be available to all the component who will listen to this event.

Listening

This is the process of registering for the event. The component has to explicitly tell the application that it wants to get the event thrown by the other component and handle the data that is also being passed along with the event. This is done with the help of addEventListener function. The next section will make this concept clearer. For now we can assume that the function allows the component to receive the event and then handle it.

For e.g.

Component called

addEventListener(ItemAddedEvent.ITEMADDEDEVENT,addItemtoCartEventHandler);

to register the event and tell which handler function to call

private function addItemtoCartEventHandler(event:ItemAddedEvent):void{

sample.text+="yes a book has been added"+ event.itemDescription;

}

In the event handler pass an object (event:ItemAddedEvent) of the customized event class type and then use the data passed through the event using the object passed in the argument.(event.itemDescription)

Now the final step is to instantiate the component that will generate the event which is

<controls:newlargebutton />

The way it will work is

Instantiate ----:comp1 will throw the customized event on click----: comp2 will listen to it and call its event handler

Now lets get into the intricacies of how the events are propagated between the components.

In Flex in order to use any component, that component must be instantiated in the application(the application tag).

<mx:Application xmlns:mx="http://www.adobe.com/2006/mxml" layout="vertical">

comp 1

comp 2

…

comp n

</mx:Application>

The Application is the root of all the components. So if make a hierarchy tree it will look like

<Application>

| |

<Comp1> <comp2>

So we can say that Application is the parent node of comp1 and comp2

Again there can be a situation where we instantiate a component inside another component. Like

<hbox>

<list>

<button>

<hbox>

so list and button are getting instantiated in hbox

so in these cases the hierarchy tree will look like

<Application>

|

<comp1>

|

<comp2>

But in this case also <application> is related to comp2 in a ancestral relationship.

We need to understand this point because of the nature of event propagation. When any component generates an event it can be listened by all the components that are higher in the hierarchy. So an event generated by comp2 can be listened by comp1 and the Application. This means comp1 can have an event listener to register for the event generated by comp2, similar is the case with Application.

So if comp2 has to send any data to comp1, it will simply generate an event which will send the data to the comp1 which would be listening for this event.

There may be a situation where the components are not in the direct hierarchy but are in a sibling relation

<Application>

| |

<comp1> <comp2>

Now an event produced by comp2 will be easily listened by Application, but not by the comp1. But comp2 wanted to make comp1 listen to the event. Now events that are listened by the parent can be listened by the children also through following command

parent.addEventListener(event type,event handler)

This means that comp1 can listen to the events that reaches upto the parent. So the event sent by comp2 reaches Appication. Now this event can be heard by the comp1.

So in order to pass the events from one component to another one has to look at the way they are related to each other and depending on that call the addEventListener().

Lets look at some of the examples for understanding the concepts

E.g.1 There are two components related to each other in a parent child relationship. Comp1: It’s a VBOX containing a button and a textbox.

Comp2: A VBOX containing TextArea

Now the requirement is that in comp1 what ever is entered in textbox should be displayed in Comp2 textarea on the button click. So basically we want to send text string from comp1 to comp2. This is a classical situation where we will write a customized event to transfer the string between the two components.

The way we have arranged the components in the implementation is

<Application>

|

<comp2>

|

<comp1>

so addeventlistener will be called without parent prefix in comp1

Custom Event class (ItemAddedEvent.as)

package controls

{

import flash.events.Event;

public class ItemAddedEvent extends Event

{

var itemDescription:String; //an item to be added to the cart

public static const ITEMADDEDEVENT:String ="ItemAddedEvent";

public function ItemAddedEvent(description:String)

{

super(ITEMADDEDEVENT,true,true);

itemDescription=description;

}

override public function clone():Event{

return new ItemAddedEvent(itemDescription);

}

}

}

Comp1 (newlargebutton.mxml)

<?xml version="1.0" encoding="utf-8"?>

<mx:VBox xmlns:mx="http://www.adobe.com/2006/mxml">

<mx:Metadata>

[Event(name="addItemEvent", type="controls.ItemAddedEvent")]

</mx:Metadata>

<mx:Label text="Item name:"/>

<mx:TextInput id="enteredItem"/>

<mx:Button width="104" height="28" cornerRadius="10" fillColors="[#00ff00, #00B000]" label="Add Item" fontSize="12" click="greenClickEventHandler()"/>

<mx:Script>

<![CDATA[

import controls.ItemAddedEvent;

private function greenClickEventHandler():void{

trace("Ouch! I got clicked! Let me tell this to the world.");

// throws the event and the passes the string to the customized event object

dispatchEvent(new ItemAddedEvent(enteredItem.text)); // passing the string

}

]]>

</mx:Script>

</mx:VBox>

Comp2: (BlindShoppingCart.mxml)

<?xml version="1.0" encoding="utf-8"?>

<mx:VBox xmlns:mx="http://www.adobe.com/2006/mxml"

xmlns:controls="controls.*"

creationComplete="init()">

<mx:Script>

<![CDATA[

import flash.events.Event;

import controls.ItemAddedEvent;

private function init():void

{

// eventlistener

addEventListener(ItemAddedEvent.ITEMADDEDEVENT,addItemtoCartEventHandler);

}

private function addItemtoCartEventHandler(event:ItemAddedEvent):void{

sample.text+="yes a book has been added"+ event.itemDescription;

}

]]>

</mx:Script>

<!—instantiated the comp1 -- >

<controls:newlargebutton addItemEvent="addItemtoCartEventHandler(event)" />

<mx:TextArea id="sample" />

</mx:VBox>

Application (greenbutton.mxml)

<?xml version="1.0" encoding="utf-8"?>

<mx:Application xmlns:mx="http://www.adobe.com/2006/mxml"

layout="vertical"

xmlns:ctrl="controls.*" >

<ctrl:BlindShoppingCart width="350" height="200" />

</mx:Application>

Same example but instead of child parent relation change it to sibling

So the instantiation of the comp1 will not happen inside comp2 instead it will happen in the Apllication

<Application>

| |

<comp1> <comp2>

Custom Event class (ItemAddedEvent.as)

package controls

{

import flash.events.Event;

public class ItemAddedEvent extends Event

{

var itemDescription:String; //an item to be added to the cart

public static const ITEMADDEDEVENT:String ="ItemAddedEvent";

public function ItemAddedEvent(description:String)

{

super(ITEMADDEDEVENT,true,true);

itemDescription=description;

}

override public function clone():Event{

return new ItemAddedEvent(itemDescription);

}

}

}

Comp1 (newlargebutton.mxml)

<?xml version="1.0" encoding="utf-8"?>

<mx:VBox xmlns:mx="http://www.adobe.com/2006/mxml">

<mx:Metadata>

[Event(name="addItemEvent", type="controls.ItemAddedEvent")]

</mx:Metadata>

<mx:Label text="Item name:"/>

<mx:TextInput id="enteredItem"/>

<mx:Button width="104" height="28" cornerRadius="10" fillColors="[#00ff00, #00B000]" label="Add Item" fontSize="12"

click="greenClickEventHandler()"/>

<mx:Script>

<![CDATA[

import controls.ItemAddedEvent;

private function greenClickEventHandler():void{

trace("Ouch! I got clicked! Let me tell this to the world.");

// throws the event and the passes the string to the customized event object

dispatchEvent(new ItemAddedEvent(enteredItem.text)); // passing the string

}

]]>

</mx:Script>

</mx:VBox>

Comp2: (BlindShoppingCart.mxml)

<?xml version="1.0" encoding="utf-8"?>

<mx:VBox xmlns:mx="http://www.adobe.com/2006/mxml" xmlns:controls="controls.*"

creationComplete="init()">

<mx:Script>

<![CDATA[

import flash.events.Event;

import controls.ItemAddedEvent;

private function init():void

{

// eventlistener

//addEventListener(ItemAddedEvent.ITEMADDEDEVENT,addItemtoCartEventHandler); --wont work

parent.addEventListener(ItemAddedEvent.ITEMADDEDEVENT,addItemtoCartEventHandler);

}

private function addItemtoCartEventHandler(event:ItemAddedEvent):void{

sample.text+="yes a book has been added"+ event.itemDescription;

}

]]>

</mx:Script>

<mx:TextArea id="sample" />

</mx:VBox>

Application (greenbutton.mxml)

<?xml version="1.0" encoding="utf-8"?>

<mx:Application xmlns:mx="http://www.adobe.com/2006/mxml"

layout="vertical" xmlns:ctrl="controls.*" >

<ctrl:newlargebutton width="350" height="82" />

<ctrl:BlindShoppingCart width="350" />

</mx:Application>

In case the parent.addeventListener is not given , the event is not dispatched to the comp2 and so on button click the text in the text are is not written.

Since in the example we have given parent.addeventListener, so comp1 and interact to comp2 via parent and we get

Reading remote XML data in Flex

We all know that Flex 2 provides ArrayCollection , XMLListCollection to read XML and consume the data.
However in case the XML contains only a single set of Data entries ArrayCollection are not the right solution.
Assuming the XML is
<dataroot>
<Capital>
<city>Delhi</city>
<country>India</country>
</Capital>
</dataroot>


In order to read this XML you have two ways to do

<mx:HTTPService id="rpcPatHTTP" url="sample.xml" resultFormat="object" result="handlerCapital(event)" />

<mx:Script>
<![CDATA[

[Bindable]
var myAC:ArrayCollection = new ArrayCollection();
private function handlerCapital(event:ResultEvent):void
{
myAC = event.result.dataroot.Capital as ArrayCollection


}

]]>
</mx:Script>

In the above case if you try to get the length of the myAC ArrayCollection, you will get 0
so
myAC.length will give 0

In such cases you should use the XMLListCollection. So the above example will be rewritten as

<mx:HTTPService id="rpcPatHTTP" url="sample.xml" resultFormat="e4X" result="handlerCapital(event)" />

<mx:Script>
<![CDATA[

[Bindable]
var myXML:XMLListCollection;
private function handlerCapital(event:ResultEvent):void
{
myXML = new XMLListCollection(event.result.Capital);

}

]]>
</mx:Script>


Now if you try to find the length of the collection you will get 1 so myXML.length will give 1, which is correct.

So my recommendation is use XMLListCollection over ArrayCollections to store the incoming remote data since you never know the number of entries in the XML file

Display Images from remote server in flex2.

I had a tough time in finding the resources that give working examples on this topic. Let me set the context of the problem then this discussion will make more sense.
There are times when you want to display images from various sources on the internet on your flex page, or you have a image server and you want to display the images from the server on to your page.
There are two ways of doing this
1. The server has sent you the image in the form of a byte array and you just want to render that byte array on your flex page
2. To mention the URL of the image and ask flex to display the image from that URL

Note: Only jpg,png,gif and swf files can be displayed.

Method 1 (byte array)

This is quite simple. (courtesy http://www.kineticz.net/blahg/2007/07/26/rending-a-bitmap-from-a-bytearray-in-flash/ )

I am giving the pieces of the code. Basically Assuming that you have your byte array ready with you from the server. I will explain how to render


<?xml version="1.0" encoding="utf-8"?>
<mx:Application xmlns:mx="http://www.adobe.com/2006/mxml" layout="absolute" creationComplete="init()">
<mx:Script>
<![CDATA[

private function renderImage():void
{
var byteLoader:Loader = new Loader();
byteLoader.loadBytes(Your_Byte_Array_Name);
imageCanvas.rawChildren.addChild(byteLoader);
}
</mx:Script>
<mx:Canvas id="imageCanvas" left="10" top="189" right="528" bottom="10"/>

</mx:Application>

We essentially have to do the following three steps:
step1 : declare an instance of Loader class.

var byteLoader:Loader = new Loader();

step2: Now load the bytes of the byte array in the loader. This is going to load the bytes in the security context of your application
byteLoader.loadBytes(your_byte_array);

step3: You need to have a canvas control where the image will be displayed so you must have something like
<mx:Canvas id="imageCanvas" left="10" top="189" right="528" bottom="10"/>

step4: Finally load the Byte Array in the canvas
imageCanvas.rawChildren.addChild(byteLoader);


Method 2: Directly accessing the images using Image URL
In this method I will give a sample code that will access the image using the image URL. First I will give the sample implementation and then explain whats need to be done. The code has been taken from http://onflex.org/flexapps/applications/ProgressiveImageLoading/srcview/index.html


<?xml version="1.0" encoding="utf-8"?>
<mx:Application
xmlns:mx="http://www.adobe.com/2006/mxml"
initialize="appInit()"
layout="absolute"
viewSourceURL="srcview/index.html"
backgroundGradientAlphas="[1.0, 1.0]"
backgroundGradientColors="[#FFFFFF, #FFFFFF]">

<mx:Script>
<![CDATA[

// Progressive Image Loading

import flash.net.URLStream;
import flash.net.URLRequest;
import flash.utils.getTimer;

// loader to hold the loaded bytes
public var loader:Loader;

// URLStream to load the image bytes
public var imageStream:URLStream;

// ByteArray to hold aggregate image data
public var imageData:ByteArray;

// Set up initial conditions when the application initializes
public function appInit():void
{
//create a URLStream to load data in. The incoming data will be obtained using URLStream.
imageStream = new URLStream();
//add some event listeners for PROGRESS and COMPLETE. This will help in order to determine when the image download is complete
imageStream.addEventListener( ProgressEvent.PROGRESS , imageStreamProgress );
imageStream.addEventListener( Event.COMPLETE , imageStreamComplete );

//create a fresh loader instance
// this is similar to the Method 1 . Only thing is that the byte Array has not been obtained till now.
loader = new Loader();
imageCanvas.rawChildren.addChild( loader );
}

public function imageStreamProgress( event:Event ):void
{
// if there are no bytes do nothing
if( imageStream.bytesAvailable == 0 ) return

// ooo bytes process the image data
this.processImageData();
}

public function imageStreamComplete( event:Event ):void
{
// if connected, stop that.
if ( imageStream.connected ) imageStream.close();

// lets refresh the displayList after rendering cycle
imageCanvas.callLater( this.processImageData );
}

public function processImageData():void
{
// if connected, read all the bytes that have been loaded into the aggregate bytearray. The byte Array is filled with the incoming data
if ( imageStream.connected ) imageStream.readBytes( imageData , imageData.length );

// clean up the loader
loader.unload();

//push the aggregate bytearray of loaded image data in there.
loader.loadBytes( imageData );
}

public function loadImage( input:String ):void
{
//if connected we need to stop that
if ( imageStream.connected ) imageStream.close();

//lets load a new image url
imageStream.load( new URLRequest( input + '?' + getTimer() ) );

// clean out all the crud in that loader
loader.unload();

//create a fresh bytearray to store aggregate image data
imageData = new ByteArray();
}

]]>
</mx:Script>


<!-- this is the place that we are accessing the jpg image from our server-->
<mx:Button click="loadImage('http://sampleserver/scripts/sample.jpg')" y="10" label="Load sample" x="371" />

<mx:Canvas id="imageCanvas" left="10" top="40" right="10" bottom="10"/>

</mx:Application>


A rough algorithm

1. Create a URLStream to access the remote image URL
2. Create a loader object
3. Create a byte Array
4. Create a canvas or some container where to render the image. Associate the loader object as a child of this canvas
5. Listen to the events ProgressEvent.PROGRESS,Event.COMPLETE. These will help in knowing if the download is over or still progressing
6. We load the URL to the URLStream
7. Clear the loader
8. initialize the ByteArray
9. Now listen to the events.
If the Progress event occurs then
10. Read the bytes into the byteArray and then load the bytes to the loader. So we will see the some portion of the image on the client
If the Complete Event occurs then
11 close the URLStream.

Books to read for Ruby on Rails

I will suggest the following books to learn ruby on rails
One should first learn ruby since rails is based on this. There are two interesting books I would recommend
Learning Ruby, Oreilly Publishers
The book talks about the syntactical details of Ruby in a simple and practical manner with lots of example

The other book is
Ruby For Rails,David Black,Manning Publication
The book explains Ruby as well as Rails. Lot of examples are given so that will help in easy understanding of the topic. The advantage of this book is that its a complete deal it talks about Ruby, Rails and gives an example Rails application.

For Rails:
1. http://www.onlamp.com/pub/a/onlamp/2005/01/20/rails.html
Refer to the above URl. This is like the first guide that every Rails beginner must refer to. The page explains how to install rails, How to write the hello world in rails. It also gives an example to complete the overview of Rails

2. Agile Web Development with Rails,second Edition:
This is the book of Rails. Written by the creator of rails (DHH) himself is the most authenticated account of the Rails framework. Usually the books written by the language creators are little tough to understand e.g. Kernighan &Richie 's C, Bjarne Stroustrup's C++, But this book is very simple to understand. The book talks about the internals of Rails framework and also develops an application. Its a must to read book

One thing I have noticed in the books that are being written these days, the authors are more inclined towards explaining the practical aspects of the language giving examples rather than just a theoretical overview.

Logger in rails

By default log files are created in the log directory. But if you want to control you logs then you can specify it in the config/environment.rb file

# Include your application configuration below
#Added to include the logging capability in the code
MY_LOGGER = Logger.new("c:/logging_rail.log")
MY_LOGGER.level = Logger::DEBUG

From the code use lgger as
MY_LOGGER.info("-----------------------------");
MY_LOGGER.info("INFO: Inside function");
MY_LOGGER.info("INFO: At:"+1.hour.from_now.to_s);
MY_LOGGER.debug("DEBUG:executeQueryByID--queryid:"+params[:queryid]);

Web as a platform --my views

Web2.0 is all about the platform.

In layman terms we call OS as a platform , this is because it gives us the necessary support that is required to develop applications and to use them. OS alone is of no use if there are no applications. The popularity of a OS (Windows,Linux) is based on the number of user/system applications it provides and the support it provides to develop new applications. Linux is quite popular among students since it provides vast support for development of applications. Another reason why we want platform is because it allows thousand of users to contribute to the development of new applications without worrying about the low level details of the system. Also the developers of the OS can concentrate on developing the platform while other people can develop the applications which they think are useful.

Think it this way If there was no OS then you would be required to perform all the memory management, task scheduling all by yourself. Even a Hello World program could then become a 1 month project !!. In simple words the kinds of application which we can develop now a days in few hours would have taken months.

Now a days all the current programming paradigms are based on frameworks .NET, J2EE, Ruby on Rails rather than on a particular language specifications. Users first look what all support a particular framework gives in terms of architecture, classes, language and then start using it. The reason why Rails is gaining so much popularity is because of the framework (the richness of the platform it provides)

The bottom line is that users want ability to extend the systems rather than just be a dumb user.

Web2.0 supports this philosophy in a big way.

The new websites/web portals are coming with the idea of providing the platform/framework. They provide the user various APIs that help him to develop the applications and plug them in their framework.E.g. facebook. The users are developing mashups that are the result of remixing multiple APIs together to create a new application. The development in web had never been so prominent as it is happening now. Like OS is running various processes which can share some information among themselves, similarly in web the portals can share information in the form on RSS feeds. The web is full of resources : Photos from Flickr, Videos from YouTube, Social Networking APIs from Facebook, Bookmarks from delicious, Storage and Computing power form Amazon, Financial services from Yahoo, Maps from Google and list goes on (This is like the drivers, Communication libraries a typical OS provide) you can use them and come up with wonderful applications.

The main reason for this success is because when you provide an application rather than a platform you have defined an intended use which has limited use. But when you provide a platform you open the opportunities for developing applications which even you had never thought about before. This is how Facebook has become so popular.Its a win-win situation for the facebook and its users

1. The popularity of the facebook is touching sky (Benefit for facebook)

2. The users can do experiments and come up with really cool applications which the creator may not have thought about. (user are happy so is facebook)

3. It creates an interest in the tech savvy users to use more and more of the facebook and hence the user faithfulness (user are happy so is facebook)

4. Overall it is helping in the growth of the craze for the web. (Most important)

I will say that the philosophy of Web2.0 to provide a platform has started the web revolution which is going to change the way the applications were developed forever.

require_gem error

c:/ruby/bin/rake.bat:24: undefined method `require_gem' for main:Object (NoMethodError)

I encountered this error when I installed rails version 2.0.2. I wanted to use

call rake db:migrate

when I encountered the error
"c:/ruby/bin/rake.bat:24: undefined method `require_gem' for main:Object (NoMethodError)"

I was able to remove this error by following way:

1. open the rake.bat file. This file is present in the bin folder of ruby. For me the path was c:rubybinrake.bat
2. The file will look like the following:

@echo off
if not "%~f0" == "~f0" goto WinNT
ruby -Sx "%0" %1 %2 %3 %4 %5 %6 %7 %8 %9
goto endofruby
:WinNT
"%~d0%~p0ruby" -x "%~f0" %*
goto endofruby
#!/bin/ruby
#
# This file was generated by RubyGems.
#
# The application 'rake' is installed as part of a gem, and
# this file is here to facilitate running it.
#

require 'rubygems'
version = "> 0"
if ARGV.size > 0 && ARGV[0][0]==95 && ARGV[0][-1]==95
if Gem::Version.correct?(ARGV[0][1..-2])
version = ARGV[0][1..-2]
ARGV.shift
end
end
require_gem 'rake', version
load 'rake'

__END__
:endofruby

3. You need to replace the line

require_gem 'rake'version

with

gem 'rake',version

4. The file rake.bat will now look like

@echo off
if not "%~f0" == "~f0" goto WinNT
ruby -Sx "%0" %1 %2 %3 %4 %5 %6 %7 %8 %9
goto endofruby
:WinNT
"%~d0%~p0ruby" -x "%~f0" %*
goto endofruby
#!/bin/ruby
#
# This file was generated by RubyGems.
#
# The application 'rake' is installed as part of a gem, and
# this file is here to facilitate running it.
#

require 'rubygems'
version = "> 0"
if ARGV.size > 0 && ARGV[0][0]==95 && ARGV[0][-1]==95
if Gem::Version.correct?(ARGV[0][1..-2])
version = ARGV[0][1..-2]
ARGV.shift
end
end
gem 'rake', version
load 'rake'

__END__
:endofruby

Hope that this should solve your problem