JSON Structure

Unit ID: M07-U04 Estimated active time: 22-30 minutes

JSON represents common structured values

JSON supports objects, arrays, strings, numbers, booleans, and null. Python's json module maps these to dictionaries, lists, strings, numbers, booleans, and None.

Write JSON

import json

records = [
    {"code": "AI-01", "hours": 8, "available": True},
]

with output_path.open("w", encoding="utf-8") as file:
    json.dump(records, file, indent=2, ensure_ascii=False)
    file.write("\n")

Indentation improves review. ensure_ascii=False preserves readable Unicode characters in UTF-8 output.

Read JSON

with output_path.open("r", encoding="utf-8") as file:
    loaded_records = json.load(file)

Inspect the top-level type and required fields. Successful parsing does not prove the data matches your task schema.

JSON is data, not executable Python

Use json.load() or json.loads(). Do not use eval() to parse JSON or other untrusted text.

JSON limitations

Tuples become arrays and load back as lists. Sets are not directly JSON serialisable. Dictionary keys in JSON are strings. Special Python objects need an explicit representation.

Malformed JSON

Missing commas, extra trailing commas, and incorrect quoting can raise json.JSONDecodeError. Preserve the file and report the error location rather than rewriting it silently.

Practice

Write a fictional accepted-record list to JSON, read it back, and assert equality. Then add an unsupported set value and inspect the TypeError without changing the original record.

Takeaway

JSON stores nested interoperable data, but parsing and schema validation are separate steps. Use the JSON module, explicit UTF-8, and reviewed representations. Next, we will handle missing, malformed, and unexpected files.