Cursor-style iteration pattern
Note: this entry has moved.
The usual enumeration pattern in .NET is either of these:
<![CDATA[foreach (string value in values)
{
// do something.
}
// or
for (int i = 0; i < values.Length; i++)
{
string value = values[i];
// do something.
}]]>
When working with cursor-style APIs like the XmlReader or
XPathNavigator, the pattern usual becomes something like this (note this
is the same code for both APIs):
<![CDATA[if (nav.MoveToFirstAttribute())
{
do
{
// Do something with nav.Current.
} while (nav.MoveToNextAttribute();
}]]>
This is not as readable or compact as the usual approach. One solution to get back the for style
of iteration, which I use all the time, is the following:
<![CDATA[for (bool go = nav.MoveToFirstAttribute(); go; go = nav.MoveToNextAttribute())
{
// Do something with nav.Current.
}]]>
With this approach, you can the usual iteration pattern again, and have more compact and readable code. It can
be used with all XPathNavigator methods that use MoveToFirst/MoveToNext:
<![CDATA[// To process namespaces
for (bool go = nav.MoveToFirstNamespace(XPathNamespaceScope.Local); go;
go = nav.MoveToNextNamespace(XPathNamespaceScope.Local))
{
}
// To process all child elements of current
for (bool go = nav.MoveToFirstChild(); go; go = nav.MoveToNext())
{
}]]>