GHGA JSON Subschema

Note: This is a fork of IBM/jsonsubschema maintained by the German Human Genome-Phenome Archive (GHGA). It was created to bring in necessary fixes, updates, and functionality required by GHGA-related projects.

ghga-jsonsubschema checks if one JSON schema is a subschema (subtype) of another.

For any two JSON schemas s1 and s2, s1 <: s2 (reads s1 is subschema/subtype of s2) if every JSON document instance that validates against s1 also validates against s2.

jsonsubschema is very useful in analysing schema evolution and ensuring that newer schema versions are backward compatible. jsonsubschema also enables static type checking on different components of a system that uses JSON schema to describe data interfaces among the system's different components.

For a practical overview of the architecture, purpose, and usage of this library, please see DETAILS.md. For the formal foundations and deep technical details, please refer to the ISSTA 2021 paper by Andrew Habib, Avraham Shinnar, Martin Hirzel, and Michael Pradel, the original authors of this library.

Installation

Requirements

• Python 3.13+

Install from PyPI

pip install ghga-jsonsubschema

Install from source

git clone https://github.com/ghga-de/ghga-jsonsubschema.git
cd ghga-jsonsubschema
uv sync

Running subschema

JSON subschema provides two usage interfaces:

CLI interface

First, create two JSON schema examples by executing the following:

echo '{"type": ["null", "string"]}' > s1.json
echo '{"type": ["string", "null"], "not": {"enum": [""]}}' > s2.json

Then, invoke the CLI by executing:

python -m jsonsubschema s2.json s1.json

Python API

from jsonsubschema import is_subschema

def main():
    s1 = {'type': "integer"}
    s2 = {'type': ["integer", "string"]}

    print(f'LHS <: RHS {is_subschema(s1, s2)}')

if __name__ == "__main__":
    main()

Development

Set up a local development environment:

uv sync --extra dev
uv run pre-commit install

Run the test suite:

uv run pytest tests/

Run the test suite with coverage:

uv run pytest --cov tests/

Changes made by GHGA

This fork is based on version 0.0.8 of IBM/jsonsubschema and introduces additional changes:

• Public API names have been changed to align with PEP 8.
• The minimum required Python version is now 3.13.
• Packaging uses more modern conventions.
• Tests have been converted from unittest to pytest.
• An empty enum is now treated as an uninhabited schema.
• Bugs inherited from upstream have been fixed: negating a numeric schema now respects exclusiveMinimum/exclusiveMaximum, intersecting numeric schemas no longer drops exclusive bounds, nested anyOf unions are now fully flattened (previously, adjacent nested unions could make two equivalent schemas compare as unrelated), and arrays with at most one item are now recognized as satisfying uniqueItems.
• The dependencies keyword (which upstream silently ignores) now raises exceptions.UnsupportedDependencies instead of potentially returning unsound verdicts.
• Negating an integer schema (e.g. {"not": {"type": "integer", "minimum": 10, "maximum": 20}}) now yields the exact complement — including the non-integer numbers, represented internally as {"type": "number", "not": {"multipleOf": 1}} — where upstream silently computes a too-small complement that can yield unsound verdicts. Only negating a numeric schema with a non-trivial multipleOf (whose complement would contain the non-multiples) raises exceptions.UnsupportedNegatedNumeric instead of returning potentially wrong results.
• Uninhabited numeric schemas whose multipleOf has no multiple within the schema's bounds are now recognized as such, and subtype checks of numeric schemas admitting a single value are now exact (e.g. {"type": "integer"} is now a subschema of {"type": "number", "multipleOf": 0.5}).

License

This repository is distributed under the terms of the Apache 2.0 License, see LICENSE.txt.