Integration Guide
This guide shows how to integrate OVS with applications, APIs, and databases.
Storage
- Store canonical OVS records as-is alongside your domain data
- Reference records by OVS ID
- Avoid denormalizing unless necessary; re-derive views at query time
APIs
- Expose OVS IDs in your responses
- Accept OVS IDs for lookups
- Document version compatibility (e.g., supports OVS v1)
Tooling (from the repo)
Validation:
python -m ovs_tools.validate --dataset data/reference
Convert JSON↔CSV:
python -m ovs_tools.convert json-to-csv --dataset data/reference --out data/csv
Query YMM and IDs:
python -m ovs_tools.query --dataset data/reference ymm --year 2014 --make-id OVS-MK-HONDA --model-id OVS-MD-HONDA-CIVIC
Repo structure highlights:
- Schemas:
/schemas(e.g.,vehicle.json,engine.json,transmission.json,body.json) - Reference dataset:
/data/reference(current),/data/YYYY(release snapshots) - Latest copy:
/data/latest
Example: Python
record = {
"year": 2024,
"make": "Example",
"model": "Alpha",
"trim": "Base"
}
ovs_id = "OVS-V-2024-EXAMPLE-ALPHA-BASE"
db.save("vehicles", ovs_id, record)
Do not ingest proprietary sources without respecting their licenses.
Note: Examples on this site are illustrative.