Skip to content

CSVW-EO Framework

CSVW-EO consists of vocabulary definitions, validation rules, and Python tooling.

Components

File Purpose
csvw-eo-vocab.ttl RDF vocabulary
csvw-eo-context.jsonld JSON-LD context
csvw-eo-constraints.ttl SHACL constraints
csvw-eo-library Python tooling

Validation Layers

CSVW-EO metadata may be validated using:

Pydantic Validation

Implemented in:

  • validate_metadata.py

Checks:

  • required fields
  • datatypes
  • structure

SHACL Validation

Implemented in:

  • validate_metadata_shacl.py

Checks:

  • RDF consistency
  • graph constraints
  • semantic rules

Python Library

The csvw-eo-library provides:

  • metadata generation
  • dummy data generation
  • validation
  • OpenDP integration
  • SmartNoise SQL conversion

Workflow Integrations

Differential Privacy Integrations

CSVW-EO can be converted to:

  • OpenDP contexts
  • SmartNoise SQL metadata

This enables automated DP pipelines (in Lomas for instance).