Abstract
Document databases use JSON (JavaScript Object Notation) for data representation. In the category of NoSQL databases they support an ability to handle large volumes of data at the absence of an explicit data schema.
On the other hand, schema information is sometimes essential for applications during data retrieval. Consequently, there are approaches to schema construction in the JSON community.
We will suppose JSON collections equipped by a schema compatible in its expressivity to the JSON Schema recommendation. We use a formal approach based on typed functional data objects.
We accommodate a classical approach to functional databases based on a typed λ-calculus to obtain a powerful non-procedural query language over JSON data with sound semantics.