How to find text on a webpage and get its XPath or CSS with Selenium and Java

Question

There is a page that lists items on a dynamic table. The order and the number of items on the table changes randomly.

I would like to find one of the items "Itemx" based on the text "Itemx" and then get and use its XPath or CSS in another part of the code.

I can find the element based on the text "Itemx"

driver.findElement(By.cssSelector("BODY")).getText().matches("^[\\s\\S]*Itemx[\\s\\S]*$");

But how can I get its XPath or CSS?

I can't seem to find anything on the web. Everything out there shows you how to verify or find the text using XPath. For me it's the opposite. I have the text and I want to find its XPath.

Chris Morgan · Accepted Answer · 2013-02-15 19:43:12Z

3

Both XPath and CSS selectors are arbitrary. There is no good answer for this in the general case.

Selecting the first c in <a><b></b><c></c><c></c></a> with a CSS selector might be c:first-of-type, or a > c:nth-child(2), or a c:first-of-type, or a > b + c, or a number of other things. Similarly as an XPath there are many ways of representing it. If you change the markup, what should happen?

It all depends upon the context, the stability required, whether false positives or false negatives are worse (you can make it very strict so that if the markup changes it will probably break, or you can make it very loose so that if the markup changes you may well select the wrong thing) and the whim of the person doing it. (Which of the ways above will I select? Eeny, meeny, miny, mo...)

If, however, you only care about the current state of a document and have no need for resilience, first of all consider if you actually need to have such a selector. You can probably pass the DOM element around instead. If you want to uniquely identify an element, though, following through the DOM tree is going to be a fairly straightforward way, travelling down level by level, observing which child the node is. You'll end up with something like this:

:root > :nth-child(3) > :nth-child(1) > :nth-child(4)

Ugly, but it may be what you want.

answered Feb 15, 2013 at 19:43

Chris Morgan

91.6k28 gold badges217 silver badges220 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

Ross Patterson Over a year ago

+1 for a great description. I usually tell people that "XPath and CSS locators are like driving directions, not addresses. Both techniques get you to the same place, but only the latter identifies the place uniquely."

Ross Patterson · Accepted Answer · 2013-02-15 22:55:32Z

1

If you can identify the item by its text, you already have an XPath for it: //*[contains(., 'Itemx').

answered Feb 15, 2013 at 22:55

Ross Patterson

9,59936 silver badges51 bronze badges

Comments

ntk4 · Accepted Answer · 2017-09-14 23:01:35Z

0

There's an add-on for Firefox's firebug called Firepath. Basically with it installed, you right click on any element on a page and select: inspect in FirePath. You can also run your tests in here too, if you're trying to modify the XPATH or CSS SELECTOR path shorter, etc.

answered Sep 14, 2017 at 23:01

ntk4

1,3071 gold badge13 silver badges19 bronze badges

Collectives™ on Stack Overflow

How to find text on a webpage and get its XPath or CSS with Selenium and Java

3 Answers 3

1 Comment

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

3 Answers 3

1 Comment

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related