A purely-functional library for defining type-safe schemas for algebraic data types, providing free generators, SQL queries, JSON codecs, binary codecs, and migration from this schema definition.
Introduction & Highlights
SchemaZ defines a generic representation of algebraic data structures and combinators that turn such schema into a generic computation over arbitrary data. In other words, SchemaZ provides a way to express any computation that abstracts over the structure of data, such as:
- Codecs: given a serial format (binary, JSON, etc.) and the schema of a data structure (say, an ADT) we can derive a codec for that data structure and serial format.
- Data Generators: given any schema we can derive random data generators (eg. scalacheck's
Gen) that produce data satisfying that schema.
- Schema/Data Migrations: since schemas are values, we can easily verify whether two versions of a schema are forward/backward compatible and provide a generic way to upgrade/downgrade data from one version of the schema to the other.
- Diffing/Patching: given a schema we can generically compute the difference between two pieces of data satisfying that schema. In the same spirit, we have generic way to apply patches to arbitrary data structures.
- Queries: knowing a schema, we can produce SQL queries to interact with a database that holds an instance of (the SQL version of) that schema.
*: shapeless provides only the way to abstract over the structure of data, but several libraries build upon shapeless to provide the feature listed in the table.
Skeumorph is also a possible competitor, however it is focused on providing translation between different formats (Avro, Protobuf and Mu). This is achieved by using a central Schema Representation which can losslessly translate to each of the previously mentioned formats.
SchemaZ shares ideas with @nuttycom's xenomorph library. The talk below presents its design.
An Haskell port of the ideas of xenomorph has also been implemented: haskell-schema.