XPath: How to select a node by its attribute?

Question

I have an XML that goes like this:

<?xml version="1.0" encoding="utf-8" ?>
<colors>
  <color index = "0">#FF0000</color>
  <color index = "1">#FF0200</color>
  <color index = "2">#FF0300</color>
  <color index = "3">#FF0500</color>
  [..]

I'm trying to select a node by its index:

XmlDocument ColorTable = new XmlDocument();
ColorTable.Load(HttpContext.Current.Server.MapPath("~/App_Data/ColorTable.xml"));
int percentage = 2;
string xpath = string.Format(@"//color[index={0}]", percentage.ToString());
//string xpath = string.Format(@"//color[index=""{0}""]", percentage.ToString());
//string xpath = string.Format(@"//color[index='{0}']", percentage.ToString());
var r = ColorTable.SelectSingleNode(xpath).Value;

I tried also the commented versions, but it does not return any result. Any suggestion?

Answer 1

Use //color[@index='{0}'] instead. The @ sign means "attribute".

I note that you're using a verbatim string literal by the way - the @ sign at the start of the string. There's no need in this case - you don't have any backslashes in the string, and it's not multi-line. You also don't need to explicitly call ToString on percentage - it will be converted automatically.

string xpath = string.Format("//color[@index='{0}']", percentage);

Answer 2

BTW, for those of us who doesn't speak native XPath, there are many online XPath "playgrounds" that allow you to write XML and XPath expression and see the results online.

Whenever I found myself in a "XPath hell" I usually go to those playgrounds and try various combination till I get my (needed) results, for some reason it works faster than writing C#/Python test program or even running those bloated so called XML editors.