CASE Examples Quality Control

The purpose of this repository is to test sample CASE data for issues such as mis-spelled identifiers, syntax normalization, and CASE/UCO concept coverage gap.

Practices implemented in this repository are not necessarily practices agreed to by the CASE or UCO communities.

Output generated: ontology coverage

Running just make will generate ontology-concept lists:

• Used ontontology concepts - This is a list of ontological concepts used in any of the CASE-Examples illustrations or website examples. They may or may not be in the current CASE release (and associated UCO release).
• Used kindOfRelationship values - This is a list of the kindOfRelationship values used, whether they are in a vocabulary in the current CASE release (and associated UCO release) or not.
• Undefined concepts - This is the list of used concepts as above, except all ontology concepts in the current CASE release (and associated UCO release) are removed.
  • Similarly, there is a list of undefined kindofRelationship values.

Tests implemented

• Turtle data is confirmed normalizeable in form with rdf-toolkit.
  • This is confirmed by running make normalize.
• JSON-LD data is confirmed normalizeable in form with Python's json.tool module.
  • This is also confirmed by running make normalize.
• JSON-LD data is confirmed normalized in form with Python's json.tool module.
  • This is confirmed by running make check.

These tests run against the CASE and UCO ontologies at the commits tracked here as Git submodules.

Manual testing

The test suite can be run manually by running:

make normalize
make check
make

Should you need to run the tests offline, after cloning this repository, run make download to retrieve resources that require network access. Network access is only needed until make download finishes. Note that a .jar file is downloaded (rdf-toolkit.jar). Its Git-tracked hash is verified before the download reports as successful (and hence before any test runs it).

Tests can run with make -j.