You might be able to reverse engineer an http search, but it would take longer than using CURL on their API.
Have you had ago at setting it up? I think you will be surprised at how easy it is, and you can then search and pull data in to bubble as if it was in your data base.