Machine Learning Authors: Yeshim Deniz, Zakia Bouachraoui, Elizabeth White, Liz McMillan, Carmen Gonzalez

Related Topics: Machine Learning , Java IoT

Machine Learning : Article

JavaServer Faces and AJAX for Google Fans

Create your own custom components and build RIAs

This is our last article in a series of four that have been introducing the concepts of creating AJAX-enabled JavaServer Faces (JSF) components. In this article we are going to summarize and encapsulate the concepts that were introduced in the three previous JDJ articles starting with the "Rich Internet Components with JavaServer Faces" (Vol. 10, issue 11), and design a Google-like JDJ InputSuggest component.

We will show you how to use Mabon to create a simple and powerful input component with built-in suggest functionality similar to what Google Suggest provides. To make it easy for application developers to use our JDJ InputSuggest component, we are going to use the Weblets open source project to bundle external resources, such as icons and JavaScript libraries, into a Java archive (JAR) that represents our JSF component bundle.

Creating an AJAX-Enabled JSF Input Suggest Component
The JSF AJAX input suggest solution consists of four classes as shown in Figure 1.

These classes are as follows:

  • The HtmlInputSuggest is the renderer-specific subclass.
  • The HtmlRenderer superclass provides some convenient methods for encoding resources.
  • The HtmlInputSuggestRenderer is your new custom Renderer, which is in charge of the markup rendered to the client, including resources needed such as JavaScript libraries and style sheets.
  • The HtmlInputSuggestTag is the tag handler.
The part of our input suggest solution that will provide the richness is a JavaScript library - inputSuggest.js - that contains functions needed to leverage Mabon to retrieve data from the application developer's backing bean. We'll focus on the following artifacts - the inputSuggest.js file and the HtmlInputSuggestRenderer - both impacted by Mabon and provide the input field with type-ahead and suggest list functionality.

The Input Suggest JavaScript Library
Since we use Mabon, there is no need to worry about fetching data from the backing bean. We can leave this to Mabon. What we do need to be concerned about, though, is how to handle the returned data on the XMLHttpRequest object, how to populate the actual suggest list, and how to handle user interactions. The inputSuggest.js library contains a number of functions that are used to handle keyboard navigation and mouse interactions, and for space limitations we'll be focusing on the functions that have the most impact on the JSF HtmlInputSuggest component.

The doKeyPress Function
The doKeyPress function, as shown in Listing 1, handles keypress events and checks whether the user pressed the TAB key or not. The TAB key would, under normal circumstances, navigate out of the input field and raising the blur event. For an input suggest solution, a TAB key can also be used to select an active row in the list of suggestions and as such we need to trap the TAB key and select a row in the suggest list and add the value to the input field, or, if no list is available, navigate out of the input field. If navigation occurs, the doBlur() function will be invoked and close the list of suggestions.

The doKeyUp Function
This function is invoked on any keyup event, and, depending on which key was activated, it will perform certain actions. Even though the TAB key has been managed by the doKeyPress function, we still have to make sure to catch it and terminate the process.

Another aspect of key strokes is the UP and DOWN arrow keys. Normally these keys will cause the cursor to navigate left and right in a regular input field. With an input suggest component a user is expecting these keys to navigate the list of suggestions.

The doKeyUp function's most important part, for this article, is to catch all key strokes that end up with a new character entered in the input suggest field, and is not a Backspace (see Listing 2). In this case the doKeyUp function will invoke the blur() function, which evaluates the value entered by the user and if changed raise a change event. We also have to reset the focus to the input suggest field, input.focus(), so the user can continue to enter more text.

If a user enters a new value, the doChange() function is invoked (see Listing 3). The doChange() function will call the mabon.send() function, passing a Map containing information about the value entered by the user, the JSF managed bean to invoke, and the callback function for this solution - _callback().

The _callback function, as shown in Listing 4, is responsible for handling the returned result from the response object and creating a list with suggestions according to the value entered by the user. Since we asynchronously communicate with the server, there is a chance that the user has entered a new character before the response comes back. To ensure that we can handle this, we have added a timer that delays the type-ahead feature until a certain time has passed.

If the value has changed since the request was made or the value is the same as the suggested value, we do nothing (see Listing 5). If the value is the same as the initial value entered, we add the first suggested value in the list to the input field and highlight the part that is appended to the value entered by the user.

The HtmlInputSuggestRenderer
It's time to have a look at how we can leverage the client-side functionality provided by the inputSuggest.js library and encapsulate it into one single JSF component. In this sample the main player is the JSF Renderer - HtmlInputSuggestRenderer. The Renderer is responsible for making sure that the correct markup is written to the client including any references to additional resources such as JavaScript libraries, CSS, icons, etc.... As we described in our previous JDJ article - "JSF and AJAX: Introducing a new open source project" (Vol. 11, issue 1) - you can use Weblets to package these resources into the same library as your JSF components.

Using Weblets
The open source Weblets project (http://weblets.dev.java.net) aims to solve the resource-packaging problem in a generic and extensible way so that all JSF component writers can leverage it, and it places no additional installation burden on the application developer.

A Weblet acts as a mediator that intercepts requests from the client and uses short URLs to serve resources from a JAR file. Unlike the servlet or filter approach, a Weblet can be registered and configured inside a JAR file, so the component library renderers, their resource files, and the Weblet configuration file (weblets-config.xml) can all be packaged together in the same JAR file. You don't need to separately deploy additional installables when the component libraries are upgraded to new versions. For the application developer, no configuration steps are needed.

It's important to note that all resources served up by Weblets are internal resources, used only by the Renderer. Any resources, such as images, that are provided by the application are supplied as component attribute values and loaded from the context root as external resources.

The HtmlInputSuggestRenderer Class
The two most important methods in this HtmlInputSuggestRenderer are the encodeBegin() and encodeEnd() methods. The encodeBegin() method, as shown in Listing 6, is responsible for writing out the needed resources for this component. The writeScriptResource() method and writeStyleResource() method are convenience methods provided by the HtmlRenderer and provide "write-only-once" semantics, preventing the same library from being written multiple times in case the application developer adds more than one input suggest component to the page.

Using Mabon
Mabon is an open source project hosted on the http://mabon.dev.java.net Web site. Mabon offers a convenient way to hook in a specially designed life cycle that is ideal for AJAX-enabled components that need to fetch data directly from a backing bean, without the overhead of a full JSF life cycle. It also provides a Mabon protocol - mabon:/ - that is used to reference the backing bean and a JavaScript convenience function that is used to send the target URL and any arguments needed and then asynchronously receive data from the managed bean.

Mabon and JSON
As you know, the XMLHttpRequest provides two response types - responseText and responseXML - that can be used to fetch data. The question to ask is, when should I use each? Answers to this question can differ depending on whom you ask, but we can recommend one rule. Ask yourself whether you control the syntax of the response.

The responseXML type returns a complete DOM object (which gives you ample ways of walking the DOM tree), allowing you to find the information needed and apply changes to the current document. This is useful when your component will impact surrounding elements and you don't control the response (for example, when you're communicating with a Web service).

For the input suggest component, you do control the response and you are looking at only fetching data for your component, not modifying the whole page's DOM structure. The responseText type returns plain text, which allows you to leverage JSON syntax for the response. For components leveraging AJAX, JSON is an extremely useful data-interchange format, since it can be easily parsed with the eval() function.

The eval() function takes one argument, a string of JavaScript code, and parses and executes this string in one go rather than trying to process each part separately. This is significantly faster than any other type of parsing, such as XML DOM parsing.

This is the reason why Mabon implements JSON - you control the response, and JSON syntax is easy and fast to parse.

The encodeEnd() Method
The real "work" is done in the encodeEnd() method, as shown in Listing 7. In the encodeEnd() method we get the Map of attributes from the HtmlInputSuggest component. One of the attributes on this component is the doSuggest attribute. From this attribute we can get hold of the MethodBinding, if any, and from the MethodBinding object we can get hold of the actual MethodBinding expression defined by the application developer, for example, #{backingBean.doSuggest}. We then trim the #{} from the expression and concatenate the remainder of the string with the mabon:/ protocol-like syntax. The MabonViewHandler will recognize the string and return a resource URL that will be written to the client (for example, /context-root/mabon-servlet-mapping/backingBean.doSuggest).

Using the Input Suggest Component
Creating an AJAX solution is not a simple task, although there are several AJAX toolkits available, such as the Dojo Toolkit (www.dojotoolkit.org), that make it a lot easier. What JSF can offer is an even simpler programming model and one that is known by millions of developers: JSP and Java. To finish this article off in a fashion that suits a proper Ajax solution, let's look at how you can use this input suggest component in a JSF application, as shown in Listing 8.

This page contains one HtmlInputSuggest component, <jdj:inputSuggest>, that has the value attribute set to a value binding expression. This expression is pointing to a value property on the backing bean. The doSuggest attribute contains a method binding expression pointing to a doSuggest() method on the same backing bean. Let's have a look at the backing bean, as shown in Listing 9.

The value property is just a plain old JavaBean property. The doSuggest() method, as shown in Listing 10, is a bit more interesting. This method takes the initial value entered by the user and passed to it via Mabon from the doChange() function (see Listing 3). The doSuggest() method then returns an Array filtered based on the users initial value to the client. It is important that the returned value conforms to the supported JSON syntax.

The end result of this HtmlInputSuggest component looks like Figure 2.

From this article, we hope you have gained an understanding of how to Ajax-enable data fetch for your JSF components using Mabon, and how you can package external resources needed for your JSF component into the same archive as your Java classes leveraging the Weblets project.

Finally, now that you know how to create reusable rich Internet components with JSF and AJAX, we hope you will apply the techniques you have learned in this article series to create your own custom components and build Rich Internet Applications (RIAs) with JSF.


More Stories By Kaazing Blog

Kaazing is helping define the future of the event-driven enterprise by accelerating the Web for the Internet of Things.

More Stories By John Fallows

John brings to Kaazing his 17 years’ experience in technology development and software design, and is considered a pioneer in the field of rich and highly interactive user interfaces. As CTO he formulates Kaazing Corporation’s vision of enabling mobile users, marketplaces and machines to connect and communicate in real-time, more reliably and at unprecedented scale. He defines the architecture of the Kaazing product suite and oversees its development. Prior to co-founding Kaazing, John served as Architect for Brane Corporation, a startup company based in Redwood City, California. Before joining Brane, he was a Consulting Member of Technical Staff for Server Technologies at Oracle Corporation. During his last five years at Oracle, he focused on designing, developing, and evolving Oracle ADF Faces to fully integrate Ajax technologies. Originally from Northern Ireland, he received his MA in Computer Science from Cambridge University in the United Kingdom and has written several articles for leading IT magazines such as Java Developer’s Journal, AjaxWorld Magazine, and JavaMagazine (DE), and is a popular speaker at international conferences. He is co-author of the bestselling book Pro JSF and Ajax: Building Rich Internet Components (Apress).

Comments (3)

Share your thoughts on this story.

Add your comment
You must be signed in to add a comment. Sign-in | Register

In accordance with our Comment Policy, we encourage comments that are on topic, relevant and to-the-point. We will remove comments that include profanity, personal attacks, racial slurs, threats of violence, or other inappropriate material that violates our Terms and Conditions, and will block users who make repeated violations. We ask all readers to expect diversity of opinion and to treat one another with dignity and respect.

CloudEXPO Stories
Cloud is the motor for innovation and digital transformation. CIOs will run 25% of total application workloads in the cloud by the end of 2018, based on recent Morgan Stanley report. Having the right enterprise cloud strategy in place, often in a multi cloud environment, also helps companies become a more intelligent business. Companies that master this path have something in common: they create a culture of continuous innovation. In his presentation, Dilipkumar Khandelwal outlined the latest research and steps companies can take to make innovation a daily work habit by using enterprise cloud computing. He shared examples from companies that have benefited from enterprise cloud computing and took a look into the future of how the cloud helps companies become a more intelligent business.
Never mind that we might not know what the future holds for cryptocurrencies and how much values will fluctuate or even how the process of mining a coin could cost as much as the value of the coin itself - cryptocurrency mining is a hot industry and shows no signs of slowing down. However, energy consumption to mine cryptocurrency is one of the biggest issues facing this industry. Burning huge amounts of electricity isn't incidental to cryptocurrency, it's basically embedded in the core of "mining." In this winner-takes-all game, burning the most electricity increases the chances of winning. The Bitcoin Energy Consumption Index states that the global energy usage of all bitcoin mining already is equivalent to the power uptake of the country of the Czech Republic. Mining equipment for a larger operation can exceed 100 megawatts (MWs) - similar to what a 1 million-square-foot Google ...
CloudEXPO has been the M&A capital for Cloud companies for more than a decade with memorable acquisition news stories which came out of CloudEXPO expo floor. DevOpsSUMMIT New York faculty member Greg Bledsoe shared his views on IBM's Red Hat acquisition live from NASDAQ floor. Acquisition news was announced during CloudEXPO New York which took place November 12-13, 2019 in New York City. Our Silicon Valley 2019 schedule will showcase 200 keynotes, sessions, general sessions, power panels, and hands on tutorials presented by 150 rockstar speakers in 10 hottest conference tracks of 2019:
The dream is universal: heuristic driven, global business operations without interruption so that nobody has to wake up at 4am to solve a problem. Building upon Nutanix Acropolis software defined storage, virtualization, and networking platform, Mark will demonstrate business lifecycle automation with freedom of choice and consumption models. Hybrid cloud applications and operations are controllable by the Nutanix Prism control plane with Calm automation, which can weave together the following: database as a service with Era, micro segmentation with Flow, event driven lifecycle operations with Epoch monitoring, and both financial and cloud governance with Beam. Combined together, the Nutanix Enterprise Cloud OS democratizes and accelerates every aspect of your business with simplicity, security, and scalability.
Andrew Keys is co-founder of ConsenSys Enterprise. He comes to ConsenSys Enterprise with capital markets, technology and entrepreneurial experience. Previously, he worked for UBS investment bank in equities analysis. Later, he was responsible for the creation and distribution of life settlement products to hedge funds and investment banks. After, he co-founded a revenue cycle management company where he learned about Bitcoin and eventually Ethereum.