Sams Teach Yourself XML in 21 Days

Sams Teach Yourself XML in 21 Days

By Steven Holzner

Part V. In Review

In Part V we took a look at working with databases and XML. This is a natural connection because XML is all about data. We took two main approaches here: treating XML documents as standard databases and using XQuery for the native XML way of handling XML documents as databases.

We started by examining the way you can bind data to Web pages in Internet Explorer. We talked about the various DSOs that you can use to connect to data sources and make XML data accessible to HTML controls and scripting code.

There are two XML DSOs in Internet Explorer: XML data islands and the XML DSO applet. You create an XML data island by using the <XML> element and the SRC attribute to connect it to an XML document. And you can connect a XML DSO to an XML document by using a <PARAM> element in the <APPLET> element.

When you want to bind data from a DSO to HTML controls, you use element attributes such as DATASRC and DATAFLD. You assign the DATASRC attribute the name of the DSO, and you assign the DATAFLD attribute the name of the field in the current record you want to work with. You can also use methods such as moveNext and moveLast to change the current record, and any such change is reflected in all the controls that are bound to that DSO.

Let's look at an example that shows some of the work we did in Day 19. In this case, we're going to search a bound XML document for some data, and we'll bind the results to HTML controls, combining our searching and navigating examples. Here's the XML document we'll use, projects.xml, which holds data about various programming projects you are doing for a set of clients:

<?xml version = "1.0" standalone="yes"?>
<document>
    <client>
            <lastname>Kirk</lastname>
            <firstname>James</firstname>
        <contractDate>September 5, 2092</contractDate>
        <contracts>
            <contract>
                <app>Comm</app>
                <id>111</id>
                <fee>$111.00</fee>
            </contract>
            <contract>
                <app>Accounting</app>
                <id>222</id>
                <fee>$989.00</fee>
            </contract>
        </contracts>
    </client>
    <client>
            <lastname>McCoy</lastname>
            <firstname>Leonard</firstname>
        <contractDate>September 7, 2092</contractDate>
        <contracts>
            <contract>
                <app>Stocker</app>
                <id>333</id>
                <fee>$2995.00</fee>
            </contract>
            <contract>
                <app>Dialer</app>
                <id>444</id>
                <fee>$200.00</fee>
            </contract>
        </contracts>
    </client>
    <client>
            <lastname>Spock</lastname>
            <firstname>Mr.</firstname>
        <contractDate>September 9, 2092</contractDate>
        <contracts>
            <contract>
                <app>WinHook</app>
                <id>555</id>
                <fee>$129.00</fee>
            </contract>
            <contract>
                <app>MouseApp</app>
                <id>666</id>
                <fee>$25.00</fee>
            </contract>
        </contracts>
    </client>
</document>

This example lets the user enter the last name of the client he or she wants to search for and click a button; the code will then jump to that record, which means the fields of that client's record will appear in the Web page.

To find the record the user is searching for, we begin by creating a new function, search. This function will read the last name the user has entered and find the matching record (in production code, you would add code here to handle problems if there was no match to the last name the user entered):

function search()
{
    var findMe = form1.text1.value.toLowerCase()

    while (!clients.recordset.EOF) {
        var clientLastName = new String(clients.recordset("lastname"))
        clientLastName = clientLastName.toLowerCase()
        if (clientLastName.indexOf(findMe) >= 0) {
            return
        }
        else {
            clients.recordset.moveNext()
        }
    }
}

After the match is made, the current record will be the record the user was searching for. To display that record's data, all we have to do is bind our DSO to various HTML controls. In this example, we'll just use two <SPAN> elements to display the client's first and last names, like this:

<APPLET CODE="com.ms.xml.dso.XMLDSO.class"
    ID="clients"
    WIDTH="0" HEIGHT="0"
    MAYSCRIPT="true">
    <PARAM NAME="URL" VALUE="projects.xml">
</APPLET>
        .
        .
        .
<FORM ID="form1">
    Last name to find: <INPUT TYPE="TEXT" NAME="text1">
    <BR>
    <BR>
    <INPUT TYPE="BUTTON" VALUE="Find client"
        ONCLICK="search()">
    <BR>
    <BR>
    First Name: <INPUT TYPE="TEXT" DATASRC="#clients"
        DATAFLD="firstname" SIZE="10">

    <BR><BR>
    Last Name: <SPAN DATASRC="#clients"
        DATAFLD="lastname"></SPAN>

</FORM>

Now when the user enters the last name of a client and clicks the Find Client button, the code moves to the client's record and display the client's first and last names automatically. Here's what the whole Web page looks like:

<HTML>
    <HEAD>
        <TITLE>
            Finding a client
        </TITLE>

        <SCRIPT LANGUAGE="JavaScript">
            function search()
            {
                var findMe = form1.text1.value.toLowerCase()

                while (!clients.recordset.EOF) {
                    var clientLastName = new
                        String(clients.recordset("lastname"))
                    clientLastName = clientLastName.toLowerCase()
                    if (clientLastName.indexOf(findMe) >= 0) {
                        return
                    }
                    else {
                        clients.recordset.moveNext()
                    }
                }
            }
        </SCRIPT>
    </HEAD>

    <BODY>
        <H1>
            Finding a client
        </H1>

        <APPLET CODE="com.ms.xml.dso.XMLDSO.class"
            ID="clients"
            WIDTH="0" HEIGHT="0"
            MAYSCRIPT="true">
            <PARAM NAME="URL" VALUE="projects.xml">
        </APPLET>

        <BR>
        <FORM ID="form1">
            Last name to find: <INPUT TYPE="TEXT" NAME="text1">
            <BR>
            <BR>
            <INPUT TYPE="BUTTON" VALUE="Find client"
                ONCLICK="search()">
            <BR>
            <BR>
            First Name: <INPUT TYPE="TEXT" DATASRC="#clients"
                DATAFLD="firstname" SIZE="10">

            <BR><BR>
            Last Name: <SPAN DATASRC="#clients"
                DATAFLD="lastname"></SPAN>

        </FORM>
    </BODY>
</HTML>

The results are shown in the following figure:

In Part V we also took a look at handling XML documents as true databases. To start, we were able to use an ASP page as an intermediary, letting us connect to a table in a database and getting back an XML document in a Web browser.

There's also a large amount of XML-enabled database support in .NET programming, and we took a look at it in overview in Day 20. For example, you were able to use the WriteXml method to write out a database table, the WriteXmlSchema method to write out an XML schema for the table, and the ReadXml method to read the table back in.

You were also able to use the .NET XmlDataDocument class to target the data in a database by using XPath. That can be useful, although XPath as currently supported often can't get the job done when you need database access. That's why XQuery was introduced—to provide a native XML way of working with XML documents and handling them as if they were databases. XQuery is still in the working draft stage, but we were able to use it along with the Galax XQuery engine in Part V.

You can use XQuery to create variables, functions, and templates, and you can then use XQuery to query XML documents to extract or manipulate the data you want. The idea is that XQuery will ultimately provide a true XML way of querying XML documents, much like SQL does for databases.

In addition, there's a great deal of XML support in .NET programming languages, and we got a look at it in overview in Part V. For example, you can create XML documents by using XML designers that are built into Visual Studio .NET, and you can create XML schemas either automatically or by dragging items into other items in an XML schema designer. You did both in Part V.

The .NET XmlTextWriter and XmlTextReader classes let you write XML documents and read them in. And XML Web controls let you display formatted XML in Web applications. To use an XML control, you use an XSLT style sheet, assigning its location to the control's TransformSource property, and you assign the XML document's location to the XML control's DocumentSource property.

The final .NET programming topic in Part V involved how to create and work with XML Web services. XML Web services let applications access code on the Web, and they use XML to communicate behind the scenes. Web services are designed to expose methods that you can call from other applications, and they typically involve database access on a Web server. You saw how that works in Part V, where you created an XML Web service that returned an entire database table by using XML on the Web.

Share ThisShare This

Informit Network