CODEWITHSUNDEEP
xmlxpath
dmonopoly
dmonopoly

Reputation: 3331

XPath following-sibling not working

Given this:

<details>
  <summary>1</summary>
  <from>2</from>
  <heading>WHAT YOU WANT</heading>
  <body>No not this</body>
</details>

I'm trying to grab WHAT YOU WANT strictly by first finding a match of "1" in the summary, then finding "2" in the from, and lastly grabbing the node just after the one with "2". (Just take the conditions of requiring checking 1 and 2 for granted - this is a simplified example of a larger problem.)

I tried this but it only matches "Node from 2":

//details[summary[text()="1"]]//following-sibling::*[contains(text(), "2")]

Even though my understanding from https://www.w3schools.com/xml/xpath_axes.asp tells me this should work?

Upvotes: 1

Views: 1669

Answers (1)

kjhughes
kjhughes

Reputation: 111541

This XPath,

//details[1]/summary[.='1']
            /following-sibling::from[.='2']
            /following-sibling::heading

will select the heading element following the from element whose string value equals 2 following the summary element whose string value equals 1, all contained within the first details element in the document.

Responding to follow-up questions in the comments:

From w3schools, the / indicates that we're looking for children of the previous node, so a / after summary, to me, means that we're looking at children of summary, for which there is nothing... Why?

Oh I see - if you use an axis right after the /, the meaning is not necessarily child anymore; it's whatever the axis specifies, like following-sibling.

Right, your original confusion stemmed from the meaning of / in XPath.

/ separates location steps. A location step consists of an axis, a node test, and zero or more predicates. The default axis is child, but others are possible. In the above XPath, the following-sibling axis is used.

Upvotes: 1

Related Questions