Skip to main content

Creating a custom component

The primary responsibility of a component is to decode and encode data.
encode = convert request parameters to component attributes
decode = render component attributes to the view (render html mostly...)

The rendering can be directly implemented by the component or can be delegated to a renderer. We use a special renderer(s) when we want different types of rendering for that component (maybe for different display devices).

So a JSF component has 2 parts: the component + renderer

Image by Rick Hightower explaining on where rendering fits in the JSF lifecycle - Ref

All custom components have to be subclasses of UIComponent. However, we subclass UICOmponentBase which is an abstract class that gives us a skeletal implementation of UIComponent.

Creating the custom component:
As an exercise, I will create a simple custom component that renders a 5x5 grid (as an html table). Here are the steps:
  1. Create a subclass of UIComponent (I chose UIOutput, since this is an output only component, ie. we do not modify the component from the view). This class does not have any properties at the moment, since the component is a simple display component that spits out a simple table. Since we do not have any properties I have not overriden the methods for saving and restoring state. Overriding encodeEnd() and decode() and returning null from them.
  2. Implemented LabelComponentTag as a subclass of UIComponentTag. The responsibility of this class is to return the component_family and component_renderer.
  3. Wrote the TLD file defining the spanningTable tag
My custom component is throwing a NullPointerException. This is because the server was not able to locate the projects tld file. When the app reads the JSP pages, it will try to locate and load all referenced tld files. Our JSP pages, declare tags with their corresponding URI's, and web.xml maps the URI's to tld files. The app gets the tld file names from web.xml, and loads the files. When a tag needs to be rendered, the server looks up the tld file to validate the tag, and if valid it gets the UIComponentTag subclass for this tag. This files in turn tells the server the name of the JSF component. This name is mapped to an actual class in faces-config.xml. That's how the server loads the component class (some subclass of UIComponent).

Anyways, so far my problem was an improperly named tld file in web.xml.
Fixed the problem, and got the simple table exercise to run.

This seems to be the order in which various methods of the Tag class, Component class, and Backing Bean are invoked.
  1. The Tag class is instantiated
  2. Properties specified in the JSP page for that tag are set in the Tag class
  3. The runtime tries to get the type of the component by invoking the getComponentType() method of the Tag class
  4. The runtime looks in faces-config.xml (it must be looking elsewhere also because core components are not defined in faces-config.xml) to see if a component has been defined with that type (nexted tag, in the tag). It gets the fully qualified class name of the component
  5. The component is instantiated
  6. The Tag class' setProperties() method is invoked. This method is responsible for setting properties from the Tag class to the component class. Not all properties in the Tag class will have corresponding properties in the component class. Some may in which case, they will be set directly (or after some conversions). Some properties may be set in the attributes Map of the component. These properties are usually those that will not be needed directly by the component but will be used by other classes such as the Renderer. Some properties may be bound to the backing bean. These properties are saved in the component using ValueBindings (The ValueBinding class interprets the JSF-EL and figures out the property of the backing bean we want to bind to).
  7. The component is Rendered by either invoking encodeBegin() or by invoking a custom Renderer. Here we may use properties from the backing bean. We can retrieve them by using the ValueBinding object.
  8. The save of the component is saved after rendering is complete.


Popular posts from this blog

Planning a User Guide - Part 3/5 - Co-ordinate the Team

Photo by  Helloquence  on  Unsplash This is the third post in a series of five posts on how to plan a user guide. In the first post , I wrote about how to conduct an audience analysis and the second post discussed how to define the overall scope of the manual. Once the overall scope of the user guide is defined, the next step is to coordinate the team that will work on creating the manual. A typical team will consist of the following roles. Many of these roles will be fulfilled by freelancers since they are one-off or intermittent work engagements. At the end of the article, I have provided a list of websites where you can find good freelancers. Creative Artist You'll need to work with a creative artist to design the cover page and any other images for the user guide. Most small to mid-sized companies don't have a dedicated creative artist on their rolls. But that's not a problem. There are several freelancing websites where you can work with great creative ar

Inheritance vs. composition depending on how much is same and how much differs

I am reading the excellent Django book right now. In the 4th chapter on Django templates , there is an example of includes and inheritance in Django templates. Without going into details about Django templates, the include is very similar to composition where we can include the text of another template for evaluation. Inheritance in Django templates works in a way similar to object inheritance. Django templates can specify certain blocks which can be redefined in subtemplates. The subtemplates use the rest of the parent template as is. Now we have all learned that inheritance is used when we have a is-a relationship between classes, and composition is used when we have a contains-a relationship. This is absolutely right, but while reading about Django templates, I just realized another pattern in these relationships. This is really simple and perhaps many of you may have already have had this insight... We use inheritance when we want to allow reuse of the bulk of one object in other

Running your own one person company

Recently there was a post on PuneTech on mom's re-entering the IT work force after a break. Two of the biggest concerns mentioned were : Coping with vast advances (changes) in the IT landscape Balancing work and family responsibilities Since I have been running a one person company for a good amount of time, I suggested that as an option. In this post I will discuss various aspects of running a one person company. Advantages: You have full control of your time. You can choose to spend as much or as little time as you would like. There is also a good chance that you will be able to decide when you want to spend that time. You get to work on something that you enjoy doing. Tremendous work satisfaction. You have the option of working from home. Disadvantages: It can take a little while for the work to get set, so you may not be able to see revenues for some time. It takes a huge amount of discipline to work without a boss, and without deadlines. You will not get the benefits (insuranc