JSON Structure Python SDK

Python validators for JSON Structure schemas and instances.

JSON Structure is a type-oriented schema language for JSON, designed for defining data structures that can be validated and mapped to programming language types.

Installation

pip install json-structure

Quick Start

Validate a Schema

from json_structure import SchemaValidator

schema = {
    "$schema": "https://json-structure.org/meta/core/v0/#",
    "$id": "https://example.com/person",
    "name": "Person",
    "type": "object",
    "properties": {
        "name": {"type": "string"},
        "age": {"type": "int32"},
        "email": {"type": "string"}
    },
    "required": ["name"]
}

validator = SchemaValidator()
errors = validator.validate(schema)

if errors:
    print("Schema is invalid:")
    for error in errors:
        print(f"  - {error}")
else:
    print("Schema is valid!")

Validate an Instance

from json_structure import InstanceValidator

schema = {
    "$schema": "https://json-structure.org/meta/core/v0/#",
    "$id": "https://example.com/person",
    "name": "Person",
    "type": "object",
    "properties": {
        "name": {"type": "string"},
        "age": {"type": "int32"}
    },
    "required": ["name"]
}

instance = {
    "name": "Alice",
    "age": 30
}

validator = InstanceValidator(schema)
errors = validator.validate_instance(instance)

if errors:
    print("Instance is invalid:")
    for error in errors:
        print(f"  - {error}")
else:
    print("Instance is valid!")

Features

Supported Types

All 34 types from JSON Structure Core v0 are supported:

Primitive Types:

• string, number, integer, boolean, null
• int8, uint8, int16, uint16, int32, uint32
• int64, uint64, int128, uint128 (string-encoded)
• float8, float, double, decimal
• date, datetime, time, duration
• uuid, uri, binary, jsonpointer

Compound Types:

• object, array, set, map, tuple, choice, any

Extensions

• Conditional Composition: allOf, anyOf, oneOf, not, if/then/else
• Validation Addins: minimum, maximum, minLength, maxLength, pattern, etc.
• Import Extension: $import, $importdefs for schema composition

Command Line Tools

# Validate a schema file
json-structure-check schema.json

# Validate an instance against a schema
json-structure-validate instance.json schema.json

API Reference

SchemaValidator

from json_structure import SchemaValidator

validator = SchemaValidator(
    allow_dollar=False,      # Allow '$' in property names
    allow_import=False,      # Enable $import/$importdefs
    import_map=None,         # Dict mapping URIs to local files
    extended=False,          # Enable extended validation features
    external_schemas=None    # List of schema dicts to sideload (matched by $id)
)

errors = validator.validate(schema_dict, source_text=None)

InstanceValidator

from json_structure import InstanceValidator

validator = InstanceValidator(
    root_schema,             # The JSON Structure schema dict
    allow_import=False,      # Enable $import/$importdefs
    import_map=None,         # Dict mapping URIs to local files
    extended=False,          # Enable extended validation features
    external_schemas=None    # List of schema dicts to sideload (matched by $id)
)

errors = validator.validate_instance(instance)

Sideloading External Schemas

When using $import to reference external schemas, you can provide those schemas directly instead of fetching them from URIs:

from json_structure import InstanceValidator

# External schema that would normally be fetched from https://example.com/address.json
address_schema = {
    "$schema": "https://json-structure.org/meta/core/v0/#",
    "$id": "https://example.com/address.json",
    "name": "Address",
    "type": "object",
    "properties": {
        "street": {"type": "string"},
        "city": {"type": "string"}
    }
}

# Main schema that imports the address schema
main_schema = {
    "$schema": "https://json-structure.org/meta/core/v0/#",
    "$id": "https://example.com/person",
    "name": "Person",
    "type": "object",
    "properties": {
        "name": {"type": "string"},
        "address": {"type": {"$ref": "#/definitions/Imported/Address"}}
    },
    "definitions": {
        "Imported": {
            "$import": "https://example.com/address.json"
        }
    }
}

# Sideload the address schema - matched by $id
validator = InstanceValidator(
    main_schema,
    allow_import=True,
    external_schemas=[address_schema]
)

instance = {
    "name": "Alice",
    "address": {"street": "123 Main St", "city": "Seattle"}
}

errors = validator.validate_instance(instance)

You can supply multiple schemas to satisfy multiple imports. The schemas are matched by their $id field against the import URIs.

Development

# Clone the repository
git clone https://github.com/json-structure/sdk.git
cd sdk/python

# Install development dependencies
pip install -e ".[dev]"

# Run tests
pytest

# Run tests with coverage
pytest --cov=json_structure --cov-report=term-missing

License

MIT License - see LICENSE for details.

Links

• JSON Structure Specification
• JSON Structure Primer
• SDK Repository