Question

What I am trying to achieve is to extract all links with a href attribute that starts with http://, https:// or /. These links lie within a table (tbody > tr > td etc) with a certain class. I thought I could specify just the the a element without the whole path to it but it does not seem to work. I get a NullReferenceException at the line that selects the links:

var table = doc.DocumentNode.SelectSingleNode("//table[@class='containerTable']");
if (table != null)
{
    foreach (HtmlNode item in table.SelectNodes("a[starts-with(@href, 'https://')]"))
    {
        //not working

I don’t know about any recommendations or best practices when it comes to XPath. Do I create overhead when I query the document two times?

Answer 1

Use:

 //tbody/descendant::a[starts-with(@href,'https://')
                     or
                       starts-with(@href,'http://')
                     or
                       starts-with(@href,'./') 
                      ]

You will still have a problem, unless you correct your code to reflect the fact that the the XmlNode.SelectNodes() instance method has a return type of XmlNodeList, not HtmlNode.