How to read and write a Pydantic class to/from a JSON file.

How to read and write a Pydantic class to/from a JSON file.

Pydantic classes are (de)serializable, although deserializing can be a little tricky.

Here is some example code, with functions to load and save data for an example Pydantic class 'MySerializableType':

import logging

from cornsnake import util_file, util_json

from pydantic import BaseModel, TypeAdapter

logger = logging.getLogger(__file__)

class CustomBaseModel(BaseModel):

    class Config:

        populate_by_name = True

        extra = "forbid"

class MySerializableType(CustomBaseModel):

    favourite_color: str = "blue"

    file_schema_version: str = "0.2"

def read_from_file(path_to_file: str) -> MySerializableType:

    try:

        json_data = util_json.read_from_json_file(path_to_file)

        deserialized =  TypeAdapter(MySerializableType).validate_python(json_data)

        return deserialized

    except Exception as e:

        logger.exception(e)

def save_to_file(path_to_file: str, data: MySerializableType) -> None:

    json_data = data.model_dump_json(by_alias=True)

    util_file.write_text_to_file(json_data, path_to_file)

notes:

- read_from_file(): TypeAdapter is used to validate and map the JSON back to the expected Pydantic class

- save_to_file(): model_dump_json() is used to serialize the Pydantic object to JSON. The parameter by_alias=True ensures that the BaseModel config options are followed (in particular, if camelCasing is specified as opposed to snake_case).