It failed if the default selector was css, and failed with selenium with xpath. Calling find('.') or find('/') in selenium returns a element that causes errors. It now calls find(:xpath, '/html') which seems to work everywhere.
We have too many top level files, we have multiple
classes in the same file. This allows us to solve
both problems, while also providing a good place
for the new Capybara::Node::Simple (formerly
Capybara::StringNode)
Jonas Nicklas