Selenium clicking on javascript link

Question

I'm having an issue with writing a Selenium test using Java and Spring.

I need Selenium to click on a delete button on a page that contains multiple delete buttons (a list of files). Using the Selenium IDE generates the following code

selenium.click("link=Delete");

which is basically useless. I haven't been able to figure out how to target the specific element contained in a table. Here's the source:

<tr onmouseover="mouseOver(this)" onmouseout="mouseOut(this)">
  <td class="thumbnail" align="center"><img src="/services/images/nav/resources.gif" /></td>
  <td colspan="3" onClick="nav('FileName'); return false">
    <a href="javascript:nav('FileName')">Basics</a></td>
 <td>
   <a class="actionButton" href="javascript:del('FileName')">Delete</a></td>
<td>&nbsp;</td>
</tr>   

I need to either a) find a way to return the xpath of the correct delete action or b) send the javascript command itself through the java code. I haven't been able to figure out how to to either, can anyone point me in the right direction?

Answer 1

Is the "Filename" part unique?

If so, the appropriate XPath would be:

selenium.click("//a[@href=\"javascript:del('FileName')\"")

Also, did you click the first Delete link from Selenium IDE? If so, try clicking on one of the subsequent ones instead and see what it comes up with.

Answer 2

There is this firefox extension called XPath Checker . You can open the html source in firefox, and then get the XPath for what you want. However, I remember having problems with some XPath expressions that didn't work properly in Selenium.

Answer 3

If it's a tool to find out the xpath of your button you're after, then there are various options:

• Use FireBug (Firefox addon): if you right-click an element in the HTML tab of firebug, you can copy its Xpath expression to the clipboard.
• Use Xpath (Firefox addon): an addon which allows you to generate, edit and inspect xpath expressions

Answer 4

I am partial towards using CSS for Selenium identifiers for 2 reasons -

1. Readability
2. Browser Independence (XPath recognition is slower on IE)

That being said, the limitation is that multiple elements having the same name, should have a distinguishable attribute.

In your source, the Selenium command with a CSS identifier would read:

selenium.click("css=a.actionButton[href=javascript:del('FileName')]");

if indeed the href attribute provides distinction between the other Delete buttons