The code that parses Overdrive representations needs to be refactored to work more like other clients, in particular the Bibliotheca and Axis 360 clients. This means:

• Integrated error handling – error messages should detected centrally and raised as exceptions.
• Use of the metadata layer throughout (Metadata, CirculationData, and FulfillmentData objects) rather than ad-hoc Python dictionaries.
• Making full use of version 2 of the availability API, which supports bulk lookups of up to 25 titles at a time. Taking advantage of this can dramatically speed up some of our Overdrive scripts – see the Bibliotheca client for an example.

While we do this, we can add support for simultaneous use titles and cost-per-circ titles, two licensing regimes that Overdrive didn't offer when this code was first written in 2014.

We may also be able to perform autodetection of Overdrive Advantage accounts, reducing or eliminating the need to have admins manually configure them.