The Integrating Digital Papyrology was an international project that joined together several databases of (mainly) Greek papyri; converted them to Epidoc, a fully conformant TEI P5 flavour. It formed the basis of the editorial project papyri.info.
The source databases had many differences not only in the way the data was structured, but also in the representation of ancient Greek alphabet and editorial intervention. A number of tools were created by the project team (I contributed to the development of some of the tools as well) and I wrote the Python script below to "run" all the programs according to parametrized requirements.
Some of the programs called are XSLTs, which are passed to one Saxon instance through a socket (instead of opening and closing Saxon at every transformation).