- •preface
- •acknowledgments
- •about this book
- •Who should read this book?
- •Roadmap
- •Code conventions
- •Code downloads
- •Author Online
- •About the title
- •About the cover illustration
- •Rethinking the web application
- •A new design for the Web
- •1.1 Why Ajax rich clients?
- •1.1.1 Comparing the user experiences
- •1.1.2 Network latency
- •1.1.3 Asynchronous interactions
- •1.1.4 Sovereign and transient usage patterns
- •1.1.5 Unlearning the Web
- •1.2 The four defining principles of Ajax
- •1.2.1 The browser hosts an application, not content
- •1.2.2 The server delivers data, not content
- •1.2.3 User interaction with the application can be fluid and continuous
- •1.2.4 This is real coding and requires discipline
- •1.3 Ajax rich clients in the real world
- •1.3.1 Surveying the field
- •1.3.2 Google Maps
- •1.4 Alternatives to Ajax
- •1.4.2 Java Web Start and related technologies
- •1.5 Summary
- •1.6 Resources
- •First steps with Ajax
- •2.1 The key elements of Ajax
- •2.2 Orchestrating the user experience with JavaScript
- •2.3 Defining look and feel using CSS
- •2.3.1 CSS selectors
- •2.3.2 CSS style properties
- •2.3.3 A simple CSS example
- •2.4 Organizing the view using the DOM
- •2.4.1 Working with the DOM using JavaScript
- •2.4.2 Finding a DOM node
- •2.4.3 Creating a DOM node
- •2.4.4 Adding styles to your document
- •2.4.5 A shortcut: Using the innerHTML property
- •2.5 Loading data asynchronously using XML technologies
- •2.5.1 IFrames
- •2.5.2 XmlDocument and XMLHttpRequest objects
- •2.5.3 Sending a request to the server
- •2.5.4 Using callback functions to monitor the request
- •2.5.5 The full lifecycle
- •2.6 What sets Ajax apart
- •2.7 Summary
- •2.8 Resources
- •Introducing order to Ajax
- •3.1 Order out of chaos
- •3.1.1 Patterns: creating a common vocabulary
- •3.1.2 Refactoring and Ajax
- •3.1.3 Keeping a sense of proportion
- •3.1.4 Refactoring in action
- •3.2 Some small refactoring case studies
- •3.2.2 Managing event handlers: Observer pattern
- •3.2.3 Reusing user action handlers: Command pattern
- •3.2.4 Keeping only one reference to a resource: Singleton pattern
- •3.3 Model-View-Controller
- •3.4 Web server MVC
- •3.4.1 The Ajax web server tier without patterns
- •3.4.2 Refactoring the domain model
- •3.4.3 Separating content from presentation
- •3.5 Third-party libraries and frameworks
- •3.5.2 Widgets and widget suites
- •3.5.3 Application frameworks
- •3.6 Summary
- •3.7 Resources
- •Core techniques
- •The page as an application
- •4.1 A different kind of MVC
- •4.1.1 Repeating the pattern at different scales
- •4.1.2 Applying MVC in the browser
- •4.2 The View in an Ajax application
- •4.2.1 Keeping the logic out of the View
- •4.2.2 Keeping the View out of the logic
- •4.3 The Controller in an Ajax application
- •4.3.1 Classic JavaScript event handlers
- •4.3.2 The W3C event model
- •4.3.3 Implementing a flexible event model in JavaScript
- •4.4 Models in an Ajax application
- •4.4.1 Using JavaScript to model the business domain
- •4.4.2 Interacting with the server
- •4.5 Generating the View from the Model
- •4.5.1 Reflecting on a JavaScript object
- •4.5.2 Dealing with arrays and objects
- •4.5.3 Adding a Controller
- •4.6 Summary
- •4.7 Resources
- •The role of the server
- •5.1 Working with the server side
- •5.2 Coding the server side
- •5.2.1 Popular implementation languages
- •5.3 The big picture: common server-side designs
- •5.3.1 Naive web server coding without a framework
- •5.3.2 Working with Model2 workflow frameworks
- •5.4 The details: exchanging data
- •5.4.2 Introducing the planet browser example
- •5.5 Writing to the server
- •5.5.1 Using HTML forms
- •5.5.2 Using the XMLHttpRequest object
- •5.5.3 Managing user updates effectively
- •5.6 Summary
- •5.7 Resources
- •Professional Ajax
- •The user experience
- •6.1 Getting it right: building a quality application
- •6.1.1 Responsiveness
- •6.1.2 Robustness
- •6.1.3 Consistency
- •6.1.4 Simplicity
- •6.1.5 Making it work
- •6.2 Keeping the user informed
- •6.2.1 Handling responses to our own requests
- •6.2.2 Handling updates from other users
- •6.3 Designing a notification system for Ajax
- •6.3.1 Modeling notifications
- •6.3.2 Defining user interface requirements
- •6.4 Implementing a notification framework
- •6.4.1 Rendering status bar icons
- •6.4.2 Rendering detailed notifications
- •6.4.3 Putting the pieces together
- •6.5 Using the framework with network requests
- •6.6 Indicating freshness of data
- •6.6.1 Defining a simple highlighting style
- •6.6.2 Highlighting with the Scriptaculous Effects library
- •6.7 Summary
- •6.8 Resources
- •Security and Ajax
- •7.1 JavaScript and browser security
- •7.1.1 Introducing the “server of origin” policy
- •7.1.2 Considerations for Ajax
- •7.1.3 Problems with subdomains
- •7.2 Communicating with remote services
- •7.2.1 Proxying remote services
- •7.2.2 Working with web services
- •7.3 Protecting confidential data
- •7.3.1 The man in the middle
- •7.3.2 Using secure HTTP
- •7.3.3 Encrypting data over plain HTTP using JavaScript
- •7.4 Policing access to Ajax data streams
- •7.4.1 Designing a secure web tier
- •7.4.2 Restricting access to web data
- •7.5 Summary
- •7.6 Resources
- •Performance
- •8.1 What is performance?
- •8.2 JavaScript execution speed
- •8.2.1 Timing your application the hard way
- •8.2.2 Using the Venkman profiler
- •8.2.3 Optimizing execution speed for Ajax
- •8.3 JavaScript memory footprint
- •8.3.1 Avoiding memory leaks
- •8.3.2 Special considerations for Ajax
- •8.4 Designing for performance
- •8.4.1 Measuring memory footprint
- •8.4.2 A simple example
- •8.5 Summary
- •8.6 Resources
- •Ajax by example
- •Dynamic double combo
- •9.1 A double-combo script
- •9.2 The client-side architecture
- •9.2.1 Designing the form
- •9.2.2 Designing the client/server interactions
- •9.3 Implementing the server: VB .NET
- •9.3.1 Defining the XML response format
- •9.4 Presenting the results
- •9.4.1 Navigating the XML document
- •9.4.2 Applying Cascading Style Sheets
- •9.5 Advanced issues
- •9.5.2 Moving from a double combo to a triple combo
- •9.6 Refactoring
- •9.6.1 New and improved net.ContentLoader
- •9.7 Summary
- •Type-ahead suggest
- •10.1 Examining type-ahead applications
- •10.1.2 Google Suggest
- •10.2.1 The server and the database
- •10.3 The client-side framework
- •10.3.1 The HTML
- •10.3.2 The JavaScript
- •10.3.3 Accessing the server
- •10.5 Refactoring
- •10.5.1 Day 1: developing the TextSuggest component game plan
- •10.5.3 Day 3: Ajax enabled
- •10.5.4 Day 4: handling events
- •10.5.6 Refactor debriefing
- •10.6 Summary
- •11.1 The evolving portal
- •11.1.1 The classic portal
- •11.1.2 The rich user interface portal
- •11.2 The Ajax portal architecture using Java
- •11.3 The Ajax login
- •11.3.1 The user table
- •11.4 Implementing DHTML windows
- •11.4.1 The portal windows database
- •11.4.3 Adding the JS external library
- •11.5 Adding Ajax autosave functionality
- •11.5.1 Adapting the library
- •11.5.2 Autosaving the information to the database
- •11.6 Refactoring
- •11.6.1 Defining the constructor
- •11.6.2 Adapting the AjaxWindows.js library
- •11.6.3 Specifying the portal commands
- •11.6.4 Performing the Ajax processing
- •11.6.5 Refactoring debrief
- •11.7 Summary
- •Live search using XSLT
- •12.1 Understanding the search techniques
- •12.1.1 Looking at the classic search
- •12.1.3 Examining a live search with Ajax and XSLT
- •12.1.4 Sending the results back to the client
- •12.2 The client-side code
- •12.2.1 Setting up the client
- •12.2.2 Initiating the process
- •12.3 The server-side code: PHP
- •12.3.1 Building the XML document
- •12.3.2 Building the XSLT document
- •12.4 Combining the XSLT and XML documents
- •12.4.1 Working with Microsoft Internet Explorer
- •12.4.2 Working with Mozilla
- •12.5 Completing the search
- •12.5.1 Applying a Cascading Style Sheet
- •12.5.2 Improving the search
- •12.5.3 Deciding to use XSLT
- •12.5.4 Overcoming the Ajax bookmark pitfall
- •12.6 Refactoring
- •12.6.1 An XSLTHelper
- •12.6.2 A live search component
- •12.6.3 Refactoring debriefing
- •12.7 Summary
- •Building stand-alone applications with Ajax
- •13.1 Reading information from the outside world
- •13.1.1 Discovering XML feeds
- •13.1.2 Examining the RSS structure
- •13.2 Creating the rich user interface
- •13.2.1 The process
- •13.2.3 Compliant CSS formatting
- •13.3 Loading the RSS feeds
- •13.3.1 Global scope
- •13.3.2 Ajax preloading functionality
- •13.4 Adding a rich transition effect
- •13.4.2 Implementing the fading transition
- •13.4.3 Integrating JavaScript timers
- •13.5 Additional functionality
- •13.5.1 Inserting additional feeds
- •13.5.2 Integrating the skipping and pausing functionality
- •13.6 Avoiding the project’s restrictions
- •13.6.1 Overcoming Mozilla’s security restriction
- •13.6.2 Changing the application scope
- •13.7 Refactoring
- •13.7.1 RSS reader Model
- •13.7.2 RSS reader view
- •13.7.3 RSS reader Controller
- •13.7.4 Refactoring debrief
- •13.8 Summary
- •The Ajax craftsperson’s toolkit
- •A.1 Working smarter with the right toolset
- •A.1.1 Acquiring tools that fit
- •A.1.2 Building your own tools
- •A.1.3 Maintaining your toolkit
- •A.2 Editors and IDEs
- •A.2.1 What to look for in a code editor
- •A.2.2 Current offerings
- •A.3 Debuggers
- •A.3.1 Why we use a debugger
- •A.3.2 JavaScript debuggers
- •A.3.3 HTTP debuggers
- •A.3.4 Building your own cross-browser output console
- •A.4 DOM inspectors
- •A.4.1 Using the Mozilla DOM Inspector
- •A.4.2 DOM inspectors for Internet Explorer
- •A.4.3 The Safari DOM Inspector for Mac OS X
- •A.5 Installing Firefox extensions
- •A.6 Resources
- •JavaScript for object-oriented programmers
- •B.1 JavaScript is not Java
- •B.2 Objects in JavaScript
- •B.2.1 Building ad hoc objects
- •B.2.2 Constructor functions, classes, and prototypes
- •B.2.3 Extending built-in classes
- •B.2.4 Inheritance of prototypes
- •B.2.5 Reflecting on JavaScript objects
- •B.2.6 Interfaces and duck typing
- •B.3 Methods and functions
- •B.3.1 Functions as first-class citizens
- •B.3.2 Attaching functions to objects
- •B.3.3 Borrowing functions from other objects
- •B.3.4 Ajax event handling and function contexts
- •B.3.5 Closures in JavaScript
- •B.4 Conclusions
- •B.5 Resources
- •Ajax frameworks and libraries
- •Accesskey Underlining Library
- •ActiveWidgets
- •Ajax JavaServer Faces Framework
- •Ajax JSP Tag Library
- •Ajax.NET
- •AjaxAC
- •AjaxAspects
- •AjaxCaller
- •AjaxFaces
- •BackBase
- •Behaviour
- •Bindows
- •BlueShoes
- •CakePHP
- •CL-Ajax
- •ComfortASP.NET
- •Coolest DHTML Calendar
- •Dojo
- •DWR (Direct Web Remoting)
- •Echo 2
- •FCKEditor
- •Flash JavaScript Integration Kit
- •Google AjaxSLT
- •Guise
- •HTMLHttpRequest
- •Interactive Website Framework
- •Jackbe
- •JPSpan
- •jsolait
- •JSON
- •JSRS (JavaScript Remote Scripting)
- •LibXMLHttpRequest
- •Mochikit
- •netWindows
- •Oddpost
- •OpenRico
- •Pragmatic Objects
- •Prototype
- •Qooxdoo
- •RSLite
- •Ruby on Rails
- •Sack
- •SAJAX
- •Sarissa
- •Scriptaculous
- •SWATO…
- •Tibet
- •TinyMCE
- •TrimPath Templates
- •Walter Zorn’s DHTML Libraries
- •WebORB for .NET
- •WebORB for Java
- •XAJAX
- •x-Desktop
- •XHConn
- •index
- •Symbols
- •Numerics
Using the framework with network requests |
237 |
|
|
6.5 Using the framework with network requests
In chapter 5 we introduced the ContentLoader object as a common encapsulation for network traffic. We can make use of our notification framework to provide status reports on any data request that we make automatically. Let’s scope out the requirements first.
When a request is made to the server, we would like to indicate that it is in progress, with a low-priority notification. To differentiate network requests from other low-level notifications, we would like to use a different icon. We have the image of the earth from our planetary viewer example in chapters 4 and 5, so let’s use that.
When the network request completes, we would like the notification to be cleared and replaced by a low-level notification if everything went okay or by a medium-level notification if there was an error.
To implement this, we simply need to create Message objects at appropriate points in the lifecycle of the request, such as when the request is fired, when it completes or errors, and so on. The modified code for our ContentLoader object is given in listing 6.9.
Listing 6.9 ContentLoader with notifications
net.ContentLoader=function( ... ){ ... }; net.ContentLoader.msgId=1; net.ContentLoader.prototype={
loadXMLDoc:function(url,method,params,contentType){ if (!method){
method="GET";
}
if (!contentType && method=="POST"){ contentType='application/x-www-form-urlencoded';
}
if (window.XMLHttpRequest){ this.req=new XMLHttpRequest();
} else if (window.ActiveXObject){
this.req=new ActiveXObject("Microsoft.XMLHTTP");
}
if (this.req){ try{
var loader=this; this.req.onreadystatechange=function(){
loader.onReadyState.call(loader);
}
this.req.open(method,url,true); if (contentType){
this.req.setRequestHeader('Content-Type', contentType);
238CHAPTER 6
The user experience
|
} |
|
b Notify request has started |
|
this.notification=new msg.Message( |
||
|
"net00"+net.ContentLoader.msgId, |
|
|
|
"loading "+url, |
|
|
|
msg.PRIORITY_LOW.id, |
|
|
|
-1, |
|
|
|
"img/ball-earth.gif" |
|
|
|
); |
|
|
|
net.ContentLoader.msgId++; |
|
|
|
this.req.send(params); |
|
|
}catch (err){ |
|
|
|
|
this.onerror.call(this); |
|
|
} |
|
|
|
} |
|
|
|
}, |
|
|
|
onReadyState:function(){ |
|
|
|
var |
req=this.req; |
|
|
var |
ready=req.readyState; |
|
|
if (ready==net.READY_STATE_COMPLETE){ |
|
||
var httpStatus=req.status; |
|
|
|
if (httpStatus==200 || httpStatus==0){ |
|
||
|
this.onload.call(this); |
c Clear initial notification |
|
|
this.notification.clear(); |
||
}else{ |
|
|
|
|
this.onerror.call(this); |
|
|
} |
|
|
|
} |
|
|
|
}, |
|
|
|
defaultError:function(){ |
d Notify on error |
var msgTxt="error fetching data!" +"<ul><li>readyState:"+this.req.readyState +"<li>status: "+this.req.status
+"<li>headers: "+this.req.getAllResponseHeaders() +"</ul>";
if (this.notification){ this.notification.clear();
}
this.notification=new msg.Message( "net_err00"+net.ContentLoader.msgId, msgTxt,msg.PRIORITY_DEFAULT.id
);
net.ContentLoader.msgId++;
}
};
Using the framework with network requests |
239 |
|
|
When we make a network request using loadXMLDoc(), we create a low-level notification band attach a reference to it to the ContentLoader object. Note that we set the lifetime to -1, so that the notification won’t expire by itself.
In the onReadyState() method, we clear the notification if everything has gone well c. In the case of an error, we call the defaultError() method, which now generates a notification of its own d. The message for this notification uses HTML markup to create a richer report than plain text could.
Listing 6.10 demonstrates an example page using this modified ContentLoader.
Listing 6.10 Notifications sample page
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
<html>
<head>
<title>Notifications test</title>
<link rel=stylesheet type="text/css" href="msg.css"/>
<script type="text/javascript" src="x/x_core.js"></script> <script type="text/javascript" src="extras-array.js"></script> <script type="text/javascript" src="styling.js"></script> <script type="text/javascript" src="msg.js"></script>
<script type="text/javascript" src="net_notify.js"></script> <script type="text/javascript">
window.onload=function(){
msg.render('msgbar');
}
var msgId=1;
function submitUrl(){ |
|
|
var url=document.getElementById('urlbar').value; |
b Make request |
|
var loader=new net.ContentLoader(url,notifyLoaded); |
||
} |
|
to server |
function notifyLoaded(){ |
c Notify that resource is loaded |
|
var doneMsg=new msg.Message( |
||
"done00"+msgId, |
|
|
"loaded that resource you asked for: "+this.url, msg.PRIORITY_LOW.id
);
msgId++;
msg.render('msgbar');
}
</script>
</head>
240CHAPTER 6
The user experience
<body>
<div class='content'>
<p>here is some content. This is what the web application is up to when not being bugged silly by all these messages and notifications and stuff. <p>Type in a URL in the box below (from the
same domain, see Chapter 7), and hit 'submit'. A souped-up contentloader that understands the notification system will be invoked.
<input id='urlbar' type='text'/>
<a href='javascript:submitUrl()'>submit</a> </div>
<div id='msgbar' class='msgbar'></div> </body>
</html>
The page (as seen in figures 6.6 and 6.7) renders a simple HTML form into which the user can type URLs. Clicking the submit link will attempt to load that URL b and fire the notifyLoaded() callback function if successful. notifyLoaded() doesn’t actually do anything with the resource, other than report that it has fetched it by creating another Message object c.
Note that the behavior on a successful request is not written into the framework but provided by a custom onload handler function. This allows the framework to be adapted to differing requirements. In the example in listing 6.9, we have hard-coded the default behavior in case of error. In a real application, every network failure may not be sufficiently important to warrant a big in-your-face dialog. We leave it as an exercise for the reader to add a parameter to the ContentLoader to denote the urgency of the notification required on failure (or else provide an overridden onError handler with a gentler notification policy).
Figure 6.6
A successfully loaded resource will be displayed as a tooltip on the status bar.