Python JSON

Python JSON: Data Serialization & Custom Encoders

JSON (JavaScript Object Notation) is the standard lightweight data-interchange format. Python's json module translates JSON strings to Python primitives (and vice-versa).

Type Translation Map

flowchart LR
    subgraph JSON Primitive
        j_obj["object {}"]
        j_arr["array []"]
        j_str["string"]
        j_num["number"]
        j_bool["true/false"]
        j_null["null"]
    end
    subgraph Python Type
        p_dict["dict"]
        p_list["list"]
        p_str["str"]
        p_num["int / float"]
        p_bool["bool (True/False)"]
        p_none["None"]
    end
    j_obj --> p_dict
    j_arr --> p_list
    j_str --> p_str
    j_num --> p_num
    j_bool --> p_bool
    j_null --> p_none

Core API Functions

• json.dumps(obj): Serialize Python object to a JSON formatted str.
• json.loads(json_str): Deserialize JSON str to a Python object.
• json.dump(obj, file_obj): Serialize object directly to a file stream.
• json.load(file_obj): Deserialize object directly from a file stream.

Practical Code Example

import json
from datetime import datetime
from typing import Any

class DatetimeEncoder(json.JSONEncoder):
    """Custom JSON Encoder supporting datetime objects."""
    def default(self, obj: Any) -> Any:
        if isinstance(obj, datetime):
            return obj.isoformat()
        return super().default(obj)

def json_serialization_demo() -> None:
    data = {
        "user_id": 1024,
        "username": "coder_dev",
        "roles": ["developer", "admin"],
        "active": True,
        "created_at": datetime.now()
    }

    # Serialize using Custom Encoder
    json_string = json.dumps(data, cls=DatetimeEncoder, indent=2)
    print("Serialized JSON:
", json_string)

    # Deserialize back
    parsed = json.loads(json_string)
    print(f"
Deserialized User: {parsed['username']} (Roles: {parsed['roles']})")

if __name__ == "__main__":
    json_serialization_demo()

Best Practices & Gotchas

• Tuples Become Arrays: When serializing a tuple to JSON (json.dumps((1, 2))), it converts to a JSON array [1, 2]. Deserializing it returns a Python list, not a tuple.
• Keys Must Be Strings: JSON dictionary keys must be strings. If a Python dict has integer keys ({1: "a"}), json.dumps() converts them to string keys ({"1": "a"}).
• Use indent=2 for Human Readability: Use indent=2 during debugging or config file generation.

Self-Check Challenge

Write a function save_json(filepath: str, data: dict) that safely writes a dictionary to a JSON file using json.dump() with 4-space indentation.