Skip to main content

Conformance and Validation


This section is in draft. The content and structure of the document may change until it is released.

Validation with SHACL

The TERN Ontology is defined as a SHACL (Shapes Constraint Language) Ontology. Property shapes of each class define the cardinality and value types to conform to the TERN Ontology. These shape constraints can be used with a SHACL processor to validate incoming or outgoing data.

Here is a short list of open-source SHACL processors:

A list of browser-based SHACL playgrounds:

Example usage with PySHACL

Install PySHACL:

pip install pyshacl

Save the TERN Ontology in a file named tern.ttl in the Turtle RDF format:

curl -L -H "Accept: text/turtle" >> tern.ttl

Assuming you have a file named data.ttl in the Turtle RDF format, run the following command to validate the data:

pyshacl -s tern.ttl -f human data.ttl

If there are no violations, then you should see some output similar to the following:

Validation Report
Conforms: True

What it doesn't do

The constraints defined with SHACL in the TERN Ontology do not check domain and protocol specific details. Example, it does not check if a categorical value is valid for an observation at the survey protocol level. However, if required, these constraints can be implemented by adopters of the TERN Ontology.