Skip to main content

Technical Overview

Java Content Repository (JCR, JSR 170) is an emerging specification that defines a common programmatic interface to content repositories. Standard relational databases are problematic because they only know how to work with structured information. JCR enables the combined use of structured and unstructured information.

Users are not required to understand JCR, and simply need to define their specific domain (e.g., as defined by HL7).

How it works

  1. The domain expert creates a UML model that defines the data items belonging to this domain, as well as the association between the various data items. Many standards such as OMG MAGE provide these UML models. HL7 CDA uses a similar type of model, referred to as RMIM.
  2. The JCR Repository Agent uses the UML model as input and generates a 'schema' for the JCR-compliant repository.
  3. The JCR Repository Agent receives the XML documents that are compliant with the defined UML model as input and stores them in the JCR-compliant repository.
  4. Users can now perform queries, search the repository data, or even use the JCR API to easily build applications on top of the repository.