Xpath to get the 2nd url with the matching text in the href tag

Question

A html page has paging links, 1 set at the top of the page and another on the bottom of the page.

Using HtmlUnit, I am currently getting the HtmlAnchor on the page using getByAnchorText("1");

There is a problem in some of the links on the top, so I want to reference the bottom links using XPath.

nextPageAnchor = (HtmlAnchor) page.getByXPath("");

How can I reference the 2nd link on the page, with using xpath?

I need to reference the link using the AnchorText, so a link like:

<a href="....">33</a>

The href has random text, and is a javascript function so I have no idea what it will be.

Is this possible with xpath?

Answer 1

To select the second a element anywhere in the document:

(//a)[2]

To select the second a element with a particular text in the href attribute:

(//a[@href='...'])[2]

Note that the parantheses are required, and that the expression //a[2] will not do what you intend: it will select all a elements that are the second a element of any parent. If your input is

<p>Link <a href="one.html">One</a></p>
<p>Link <a href="two.html">Two</a> and <a href="three.html">Three</a>.</p>
<p>Link <a href="four.html">Four</a> and <a href="five.html">Five</a>.</p>

(//a)[2] will return the second link (two.html), while //a[2] will return the third and fifth link (three.html and five.html), since these both are the second a child of their parent.

Answer 2

It's pretty simple:

 (//a)[2]

the //a gets all anchors on the page and the [2] gets the second one (it's one-indexed not zero-indexed, so 2, is actually the 2nd, not the 3rd as you would expect with an array, for example)

If you want to get a link with the text of 33 then you can use:

 //a[./text() = "33"]

See http://www.w3.org/TR/xpath/ for the full xpath definition.

EDIT

To address Alexandre's comment, you could use

 (//a[./text() = "33"])[2]

This will first select all <a> tags with a text of 33, and then it will select the second of those.

EDIT 2

NOTE: The location path //para[1] does not mean the same as the location path /descendant::para[1]. The latter selects the first descendant para element; the former selects all descendant para elements that are the first para children of their parents.

Markusk is indeed correct. The quote above is from the xPath definition referenced above.