XPath以选择名称开头的节点

Question

I have XML that looks like this:

<detail>
  <address>
    <line1/>
    <line2/>
    <line3/>
    <postcode/>
  </address>
</detail/>

There could be any number of <line*> nodes which I want to select, and other nodes which I don't want to select.

I have tried this, which doesn't seem to work (in C# anyway):

/detail/address/[substring(name(),4) = 'line']

Al, help appreciated!

Answer 1

尝试

/detail/address/*[starts-with(name(), 'line')]

Answer 2

From memory, I think the index is 1 based not 0 based. Perhaps this will work:

/detail/address/[substring(name(),1,3) = 'line']

Alternatively, do you have control over the format of the XML? If so, a better approach might be to structure it like so:

<detail>
  <address>
    <line number='1'/>
    <line number='2'/>
    <line number='3'/>
    <postcode/>
  </address>
</detail/>

Then you can use the following to retrieve the lines:

/detail/address/line

Answer 3

Consider using XDocuments.

        XDocument doc = XDocument.Parse(xmlString);
        foreach (XElement element in doc.Descendants())
        {

            if(element.Name.LocalName.StartsWith("line"))
               //DoStuffWithValueOfThatElement(element.Value)...
        }

Answer 4

You can try with contains :

/detail/address/*[contains(name(),'line')]

If you want to use substring you've to know that the index starts with 1 and you're also missing the wildcard:

/detail/address/*[substring(name(),1,4) = 'line']