Concepts

This part explains the ideas you need to use FerroEHR effectively. Four chapters; the first two read in either order, the last two go deeper:

• openEHR primer: the standard itself: the Reference Model, archetypes and templates, compositions, versioning, and AQL, with no prior openEHR knowledge assumed. Read this first if the words “archetype” and “composition” are new to you.
• System architecture: how this server is put together and where your data actually lives, so the behaviour you see through the API makes sense: why the specification layer is generated, how storage and versioning work on PostgreSQL, and how an AQL query becomes SQL.
• Storage architecture: the deep-dive on the tables: the temporal version table, the decomposed node model with its nested-set index, the one-transaction write path, and the cold archival tier, each with a diagram.
• The AQL engine: the query pipeline from lexer to RESULT_SET, what each stage refuses, and the design reasons queries are fast.

None of them is a prerequisite for Getting started; you can commit a composition and run a query without them. Come back when you want to know why the API behaves as it does, or before you make one of the decisions you then live with: a specification generation, a template design, a query shape.