Blog / XPath and escaping a comma

For those of you looking for the answer to how to escape a comma in an XPath text match, I’m sorry but I just don’t know it. But if you are using Capybara, have no fear, there is a way round.

It was Friday afternoon and I was cleaning up some tests, and had the XPath:

//ul[@id='stories_page']/li[@class='story_headline'][1]//p[2][text()='Posted by Clint Eastwood on Wednesday, 9 September 2009']

The comma appearing after ‘Wednesday’ was causing all manner of havoc. Using the XPath Checker Firefox add-on, I tried escaping it with backslashes, ampersands, a forty-four magnum (I was desperate) but no go. It was only when I can across this capybara spec test for has_xpath that the solution became apparent:

have_xpath("//ul[@id='stories_page']/li[@class='story_headline'][1]//p[2]", :text => 'Posted by Clint Eastwood on Wednesday, 9 September 2009')

Although there was the same solution much closer to home. If anybody knows the answer to escaping a comma without Capybara, then please let us know. For once, Google failed me.