Working with JSON in Python

JSON (JavaScript Object Notation) is a lightweight, text-based data format widely used for storing configuration data and exchanging information between programs — especially over the web via APIs. Python's built-in json module provides a direct, well-defined mapping between JSON text and native Python data structures.

Because JSON's structure (objects, arrays, strings, numbers, booleans, and null) maps almost one-to-one onto Python's own dictionaries, lists, strings, numbers, booleans, and None, converting between the two formats is usually simple and predictable.

Why it matters

JSON is the format in which web APIs send and receive data, and it is also widely used for configuration files. Every time a Python program talks to a web service, it converts dictionaries to JSON text and back. The mapping between the two is close, but the few differences are worth knowing before they cause a bug.

Converting Python to JSON: json.dumps()

json.dumps(obj) serializes a Python object into a JSON-formatted string. It accepts useful keyword arguments such as indent=4 for pretty-printed, human-readable output, and sort_keys=True to order dictionary keys alphabetically. To write JSON directly to a file instead of returning a string, use json.dump(obj, file).

Converting JSON to Python: json.loads()

json.loads(text) parses a JSON-formatted string and returns the equivalent Python object. To read JSON directly from an already-open file object instead of a string, use json.load(file). If the text is not valid JSON, both functions raise a json.JSONDecodeError.

The Type Mapping Between JSON and Python

Understanding exactly how types translate in each direction avoids confusion when working with parsed data.

• JSON object {} ↔ Python dict
• JSON array [] ↔ Python list
• JSON string "..." ↔ Python str
• JSON number (integer or decimal) ↔ Python int or float
• JSON true / false ↔ Python True / False
• JSON null ↔ Python None

Examples

import json

person = {
    "name": "Elena",
    "age": 31,
    "is_member": True,
    "languages": ["Python", "SQL"],
    "manager": None,
}

json_text = json.dumps(person, indent=2)
print(json_text)

{
  "name": "Elena",
  "age": 31,
  "is_member": true,
  "languages": [
    "Python",
    "SQL"
  ],
  "manager": null
}

import json

data = '{"name": "Elena", "age": 31, "languages": ["Python", "SQL"]}'
parsed = json.loads(data)

print(type(parsed))
print(parsed["name"])
print(parsed["languages"][0])
print(parsed["age"] + 1)

<class 'dict'>
Elena
Python
32

Exercise

Convert the dictionary to a JSON string and print it. Then parse the JSON text given below and print the Python type of its price and the value of its tags.

Expected output: {"name": "Asha", "skills": ["python", "sql"], "active": true} then float None

import json

user = {"name": "Asha", "skills": ["python", "sql"], "active": True}
# TODO: print the dictionary as a JSON string

text = '{"price": 12.5, "tags": null}'
# TODO: parse the text, then print the type name of price and the value of tags

import json

user = {"name": "Asha", "skills": ["python", "sql"], "active": True}
print(json.dumps(user))

text = '{"price": 12.5, "tags": null}'
data = json.loads(text)
print(type(data["price"]).__name__, data["tags"])

Quick quiz

Interview questions