JSON is a human-readable text-based data format. It is language independent and used for data interchange between applications.

In this article, we’ll explain how to parse JSON data in Python.

Python JSON #

The json module that allows you to encode and decode JSON data is a part of the Python standard library.

JSON is a string that represents data. Encoding or serialization means transforming a Python object into a JSON string that can be stored in a file or transmitted over the network. Decoding or de-serialization the reverse process of encoding where a JSON string is transformed into Python object.

Below is a table showing Python objects and their equivalent JSON representation:

To work with JSON simply import the module at the top of your file:

import json

Encoding JSON in Python #

The json module has two methods for encoding Python objects into JSON formatted strings: dump() and dumps().

The dump() method sends the output to a file-like object. It takes two positional arguments: the object to be encoded and the file-like object. Here is an example:

data = {
    "country": "Germany",
    "vehicle": {
        "name": "Volkswagen",
        "model": "T-Roc"
    }
}

with open("file.json", "w") as file:
    json.dump(data, file)

If you run the script it will create a file named file.json:

{"country": "Germany", "vehicle": {"name": "Volkswagen", "model": "T-Roc"}}

The dumps() method works same as dump() but instead of sending the output to a file-like object, it returns a string:

data = {
    "country": "Germany",
    "vehicle": {
        "name": "Volkswagen",
        "model": "T-Roc"
    }
}

json.dumps(data)

'{"country": "Germany", "vehicle": {"name": "Volkswagen", "model": "T-Roc"}}'

Both methods accept same keyword arguments. For example if you are analyzing or debugging the JSON dat you may want to specify the indentation level:

data = {
    "country": "Germany",
    "vehicle": {
        "name": "Volkswagen",
        "model": "T-Roc"
    }
}

print(json.dumps(data, indent=2))

{
  "country": "Germany",
  "vehicle": {
    "name": "Volkswagen",
    "model": "T-Roc"
  }
}

Decoding JSON in Python #

To transform JSON encoded data into Python objects, use the load() and loads() methods.

The load() method reads JSON structure from a file-like object and transforms it into a Python object.

Let’s say we have the following JSON file:

[
  {
    "userId": 1,
    "id": 1,
    "title": "Meet with Lisa",
    "completed": true
  },
  {
    "userId": 1,
    "id": 2,
    "title": "Design a prototype",
    "completed": false
  }
]

To transform the JSON data to a Python representation, you would use something like this:

import json

with open('file.json') as f:
  data = json.load(f)

type(data)

The JSON is transformed into a Python list, that you can use in your code:

<class 'list'>

The loads() method converts a string containing a JSON document to a Python object:

import json

json_str= '{"userId": "1", "id": "1", "title": "Meet with Lisa", "completed": "True"}'

print(json.loads(json_str))

The string is transformed into a Python dictionary:

{'userId': '1', 'id': '1', 'title': 'Meet with Lisa', 'completed': 'True'}

Here is a more advanced example that shows how to make an api request and decode the JSON data:

import json
import requests

response = requests.get("https://jsonplaceholder.typicode.com/users")
users = json.loads(response.text)

print(users)

Conclusion #

We’ve hows you how to encode and decode JSON data in Python.

If you have any questions or feedback, feel free to leave a comment.