Visual CSharp 1002005 A Developers Notebook [Electronic resources] نسخه متنی

5.7. Improve XML Manipulation with XPathDocument

.NET 2.0
provides a new, much-improved
XPathDocument class that offers a cursor-based XML
reader that is more flexible than XmlWriter and
XmlReader and is more efficient (and faster) than
XmlDocument.

With XPathDocument it is not necessary for the
entire XmlDocument to be loaded into memory.
What's more, XPathDocument
supports data binding. You can display the
XPathDocument in a control just by setting the
DataSource property to point to the
XPathDocument.

5.7.1. How do I do that?

Create a new Windows program (called
XPathDocumentTester). Add an XML document, named
BookList.xml:

<?xml version="1.0" encoding="utf-8" ?>
<Books>
<book>
<BookName>Programming C#</BookName>
<Author>Jesse Liberty</Author>
<Publisher>OReilly Media</Publisher>
</book>
<book>
<BookName>Programming ASP.NET</BookName>
<Author>Jesse Liberty</Author>
<Author>Dan Hurwitz</Author>
<Publisher>OReilly Media</Publisher>

Note: XPathDocument offers a flexible, fast, and efficient
mechanism for loading and binding XML documents.

    </book>
<book>
<BookName>Visual C# Notebook</BookName>
<Author>Jesse Liberty</Author>
<Publisher>OReilly Media</Publisher>
</book>
<book>
<BookName>Visual Basic 2005 Notebook</BookName>
<Author>Matthew MacDonald</Author>
<Publisher>OReilly Media</Publisher>
</book>
</Books>

On the form, add a label and a text box, and then add a second label
beneath the first one as well, as a button, as shown in Figure 5-28. Name the text box
txtBookName, name the second label
lblAuthor, and name the button
btnFind.

The work, of course, is done in the event handler for the Find
button, as shown in Example 5-5.

Example 5-5. Using XPathDocument and a navigator

#region Using directives
using System;
using System.Collections.Generic;
using System.ComponentModel;
using System.Data;
using System.Drawing;
using System.Windows.Forms;
using System.Xml.XPath;
#endregion
namespace XPathDocumentTester
{
partial class Form1 : Form
{
public Form1( )
{
InitializeComponent( );
}
private void btnFind_Click(object sender, EventArgs e)
{
int numAuthors = 0;
string authorName = string.Empty;
// get the file as an XPathDocument
XPathDocument document = new XPathDocument("..\\..\\BookList.xml");
// Get a navigator over the document
XPathNavigator navigator = document.CreateNavigator( );
navigator.MoveToFirstChild( );   // move to books
navigator.MoveToFirstChild( );   // move to book
do                        // look at each node
{
// move to entries under book
navigator.MoveToFirstChild( ); 
do                     // look at each sub entry under book
{
// if you match the book name, you found the book we want
if (navigator.Value =  = txtBookName.Text)
{
do               // moving through the name, author, publisher
{
// if you found the author tag, get all the authors
if (navigator.Name =  = "Author")
{
if (++numAuthors > 1)   // make a list
{
authorName += ", ";
}
authorName += navigator.Value;
}
} while (navigator.MoveToNext( ));   
}               // end if we found the book
} while (navigator.MoveToNext( ));   // go to the entry under book
navigator.MoveToParent( );         // done with this book, go up a level
} while (navigator.MoveToNext( ));      // go to next book
if (numAuthors =  = 0)
{
authorName = "Not Found";
}
lblAuthor.Text = authorName;
}
}
}

5.7.2. What just happened?

The best way to see how this code works is to step through it in the
debugger. Place a break point on this line:

XPathDocument document = new XPathDocument("..\\..\\BookList.xml");

Because the default place to look for the .xml
file is in the debug directory, you are setting
the relative path up to where the source code is held.

Run the debugger to this line. Enter the title of a book (e.g.,
Programming ASP.NET) and click Find. The
debugger stops at your break point. Press F10 and hover your mouse
cursor over the document. You'll find that you have
an object of type XPathDocument.

Press F10 to get the navigator. Open the Watch Window and add three
entries, as shown in Figure 5-29.

Your cursor is on this line of code:

navigator.MoveToFirstChild( );

and navigator.Name is blank. Press F10 and the
navigator.Name value changes to
Books. You want to go down one more level. Press
F10 and the navigator.Name value changes to
Book. Aha! Now you are examining a
Book node, which is just what you want.

Enter the do loop and execute the
MoveToFirstChild command. This sets the
navigator.Name value to
BookName and the
navigator.Value value to Programming
C#. This is not the book you are looking for, so the
if statement fails, as shown in Figure 5-30.

Press F10 to cycle through the do loop. Finally,
you've looked at all the fields for that book and
you hit this line:

navigator.MoveToParent( );

This brings you back up to the book level.
Continue to press F10 and you move to the next book and then enter
the do loop again, where you move to the first
child of that book. Now you are ready to examine the book name, and
this time you have the right book. The if
statement succeeds, so you enter the innermost do
loop to iterate through the entries for this book.

When the navigator.Name field is equal to
Author you have the right author for the right
book, so you can add to the authorName string (the
final if statement just puts commas between names
for multiple authors).

When you exit all of this code, you've grabbed the
names for the authors of the chosen book and you've
displayed them in the label, as shown in Figure 5-31.

5.7.3. What about...

...selecting a portion of the XML document rather than searching the
entire document? Can I do that?

Because you know the structure of your document, you can use
XPath to narrow the selection. This is shown in
the next lab.

5.7.4. Where can I learn more?

Learn more about
XPathDocument on the MSDN site in the article
titled "XPathDocumentClass." Also,
a good weblog article is on DotNetJunkies at http://dotnetjunkies.com/WebLog/jcmag/archive/2004/02/09/6862.aspx.