validate.py

import sys
from jsonschema import validate, RefResolver, Draft202012Validator
import json
from pathlib import Path
import logging
import argparse


def handle_uri(u):
    print(u)

def load_json(path : str):
    with open(path) as f:
        return json.load(f)

if __name__ == "__main__":
    parser = argparse.ArgumentParser()

    parser.add_argument('-s','--schemadir', action="append", help="Directory containing supporting schema files to use for validation")
    parser.add_argument('-v', '--verbose', action="count", help="Verbose logging (can be specified multiple times)")
    parser.add_argument('-i', '--input', type=argparse.FileType('r'), default=sys.stdin, help="Path to input file (if absent, stdin is used)")
    parser.add_argument('schema', help="Primary schema to validate against")    

    args = parser.parse_args()

    match args.verbose:
        case v if v and v >= 2:
            logging.basicConfig(level=logging.DEBUG)
        case 1:
            logging.basicConfig(level=logging.INFO)
        case _:
            logging.basicConfig(level=logging.WARNING)

    logging.debug(f"Arguments: {args}")

    instance_doc = json.loads(args.input.read())
    args.input.close()
    main_schema = load_json(args.schema)    
    schema_dict = { main_schema['$id'] : main_schema }

    if args.schemadir:
        schema_paths = []    
        for d in args.schemadir:
            schema_paths += [f for f in Path(d).rglob("*.schema.json")]
        logging.info(f"Schema files loaded: {schema_paths}")

        schemas_json = [json.load(p.open()) for p in schema_paths]
        schema_dict = schema_dict | { s['$id'] : s for s in schemas_json }

    logging.info(f"Schema IDs loaded: {[k for k in schema_dict.keys()]}")

    logging.debug (f"Instance doc: {instance_doc}")
    logging.debug (f"Main schema: {main_schema}")

    resolver = RefResolver(None, 
                           referrer=None, 
                           store=schema_dict)
    
    v = Draft202012Validator(main_schema, resolver=resolver)

    v.validate(instance_doc)

    logging.info("Done")