Oops. My bug. It is our example, of course, but in practise, you would never want to do this in a pipeline root, since your essentially creating a single data set pipeline root.
The more common usage would be to have subdirectories for your data.
I'll fix this for our next release, but for the moment, if you want to run a search on Example4.mzXML, you will need to create a subdirectory in your root, and move it there.