Microsoft SharePoint 2007 Technologies

354 13.5 Coding Web Parts

ever, when developing more complex Web Parts, the underlying develop-

ment steps that you need to follow to create and deploy the Web Part remain

the same. Now that we understand the basics of Web Part development, let’s

move on to building slightly more complex Web Parts.

13.5 Coding Web Parts

When working with Web Parts, it helps to understand how a Web Part is

actually rendered. The ASP.NET 2.0 Web Part framework, and in particular

the

SPWebPartManager control, is at the heart of SharePoint’s rendering

process. The

SPWebPartManager control overrides the standard ASP.NET

2.0

WebPartManager control and allows data to be stored in the SharePoint

content database instead of the ASP.NET services database. When a request

is made for a Web Part page, SharePoint retrieves the page source. If the

code has not been customized using for example, SharePoint Designer

2007, it is considered an uncustomized (ghosted) page, and the source is

initially retrieved directly from the SharePoint directory on the Web server.

If the page has been customized, the source is retrieved from the content

database.

13.5.1 Web Part Rendering Process

The rendering process for an individual Web Part can be broken down by

examining the sequence of ASP.NET events executed throughout the pro-

cess (Figure 13.45). When a Web Part page is requested, the

OnInit event is

triggered. The

OnInit event launches the initialization phase for the Web

Part. During the initialization phase, the Web Part framework retrieves the

property values associated with the Web Part from the database, such as the

title of the Web Part and its height and width. If your custom Web Part

needs to perform any additional initialization steps, you can override the

OnInit event. If this is the first time the Web Part is loaded, it proceeds to

the

OnLoad event, which sets the Web Part properties with the values

retrieved from the initialization phase. If the Web Part has been loaded pre-

viously, the

LoadViewState method is invoked to load the state informa-

tion previously serialized into the Web Part, after which the

Onload event

proceeds and updates the properties. You can leverage the

OnLoad event to

retrieve information from various data sources, such as a database, that you

want to display in the Web Part. If your Web Part uses any ASP.NET 2.0

controls, the

EnsureChildControls and CreateChildControls events are

launched to ensure that all the appropriate server controls are initialized and

created in the control tree. Controls added by overriding the

CreateChild-

Controls

method are automatically rendered by the framework and do not

require further processing in the subsequent rendering events.

13.5 Coding Web Parts 355

Chapter 13

Once the Web Part has finished creating the required server controls and

has handled user-generated events, it is ready to render the output to the

browser. The rendering phase commences with the

OnPreRender event.

Developers can use the

OnPreRender event to change Web Part properties

before the output is rendered to the browser. On completion of the

OnPreRender event, the view state information of the Web Part is serialized

and saved to a hidden variable on the Web page for future load requests. At

this stage of the rendering process, the output can actually be rendered to the

browser. Two rendering methods, or events, are responsible for producing the

final output: the

Render event, and the RenderContents event. The Render

event is responsible for rendering the entire Web Part, including the chrome,

whereas the

RenderContents event is only responsible for rendering the

contents of the Web Part.

13.5.2 Web Part Properties and Methods

So far, we know that a Web Part is an ASP.NET control that inherits its prop-

erties and methods from the ASP.NET 2.0

WebPart base class. The proper-

ties associated with a Web Part determine the appearance and behavior of the

Web Part, and each property typically has a get and set accessor method

defined. The get accessor enables you to retrieve the value of a property, and

the set accessor enables you to set the value of a property. Figure 13.46 shows

sample C# code to define a custom Web Part property. If you have experi-

ence developing Web Parts for Windows SharePoint Services 2.0 and Share-

Point Portal Server 2003, you will notice some similarities in the process of

defining custom properties; you will also notice some differences in the

attribute terminology. For example, the

WebBrowsable attribute determines

whether the Web Part property appears as a property within the Web Part

tool pane; the same attribute was named

Browsable in the previous version

of SharePoint.

Figure 13.45

Web Part

rendering process.