DCAT-AP / mobilityDCAT-AP Test Suite

This test suite validates conformance against:

• DCAT-AP 3.0
• mobilityDCAT-AP 1.1.0

This repository contains a Python-based validation framework for testing SHACL shapes against DCAT-AP and mobilityDCAT-AP example datasets.

Validation is executed locally using pySHACL instead of SHACL Playground to ensure reliable resolution of multiple SHACL files and owl:imports. This guarantees consistent and reproducible validation behaviour across all test cases.

---

Purpose

The goal of this repository is to:

• Validate DCAT-AP and mobilityDCAT-AP example data against SHACL constraints
• Organise structured positive and negative test cases
• Detect breaking changes through regression testing
• Ensure stable constraint behaviour across specification updates

---

Project Structure

testing_SHACL/
│
├── data/
│   ├── baseline-dcat-ap/
│   ├── mobility/
│   ├── multilingual/
│   ├── ranges/
│   ├── vocabularies/
│   ├── partial-graphs/
│   └── regression/
│
├── shacl/
│   ├── dcat-ap-3.0.shacl.ttl
│   └── mobilitydcat-ap-1.1.0.shacl.ttl
│  
│
└── validate_per_data_file.py

data/

Contains all validation test cases, grouped by category.

Categories include:

• baseline-dcat-ap – Core DCAT-AP validation cases
• mobility – mobilityDCAT-AP extension cases
• ranges – Datatype and range validation tests
• multilingual – Language tag and multilingual constraints
• vocabularies – Controlled vocabulary enforcement
• partial-graphs – Structural and edge-case tests
• regression – Previously validated behaviour that must remain stable

Each category may contain:

• positive/ – Files that are expected to conform
• negative/ – Files that are expected to fail validation

---

shacl/

Contains all SHACL shape files used during validation.
All files in this folder are loaded and merged into a single validation graph before execution.

---

validate_per_data_file.py

Python script that:

1. Recursively scans the data/ directory for Turtle (.ttl) files
2. Loads and merges all SHACL files from the shacl/ directory
3. Validates each data file individually using pySHACL
4. Prints validation results directly to the console

No documents or reports are generated. Output is console-based to support automation and CI usage.

---

⚙️ How Validation Works

1. All .ttl files under data/ are discovered recursively.
2. All SHACL files under shacl/ are loaded into a single validation graph.
3. Each data file is validated using pySHACL.
4. Results are printed to the console.

Validation uses:

• RDFS inference
• Multiple SHACL files
• Console-only output (no documents generated)

---

Running the Tests

1. Install Dependencies

Install the required Python packages:

• pyshacl
• rdflib

You can install them using:

pip install pyshacl rdflib

2. Execute the Validation Script

From the root directory of the repository, run:

python validate_per_data_file.py

The script will:

Print the number of discovered test files

List the loaded SHACL files

Display PASS/FAIL results per file

Optionally print detailed validation reports (if debug mode is enabled)

3. Example Output

Passing case: PASS → baseline-dcat-ap/positive/B-P-01-minimal.ttl

Failing case: FAIL → baseline-dcat-ap/negative/B-N-01-missing-title.ttl

Regression Testing

The regression/ folder protects previously validated behaviour.

regression/positive/ → must always pass

regression/negative/ → must always fail

If a regression test changes status, this indicates:

A breaking change (previously valid data is now invalid)

A weakened constraint (previously invalid data now passes)

Regression tests safeguard specification stability.