Python requests library: use of json.dumps

Question

Potential newbie question, but it has been bothering me since long time.

In a lot of Python tutorials, I see very similar snippets to post something to a REST API

import json
import requests

url = 'https://api.blabla.com/v1'
token = '1234'

headers = { 
            'Authorization: f'Bearer {token}',
            'Content-Type: 'application/json'
          }
body = {
   'name': 'Test name'
}

my_response = requests.post(url, headers=headers, data=json.dumps(body)).json()
print(my_response)

The Request documentation says that the data object should receive a dict. In the above, the body is a dict already, so why is there a need to use json.dump because that essentially makes it a string (whereas documentation says it should be a dictionary).

So why shouldn't it be:

my_response = requests.post(url, headers=headers, data=body).json()

On another note, the official Requests documentation also says we can pass the json parameter and in this example it would be (if we follow the documentation):

my_response = requests.post(url, headers=headers, json=body).json()

But the JSON parameter should be a JSON object, so if we pass the body as such, it means we are passing a dict object to the JSON parameter which is not what the documentation says.

Note:

• The official documentation is using data=json.dumps(body)
• The W3schools documentation is using data=body

Answer 1

data – the body to attach to the request. If a dictionary or list of tuples [(key, value)] is provided, form-encoding will take place.

https://requests.readthedocs.io/en/master/api/#requests.Request

_{(emphasis mine)}

If you pass a dict as data , it will be form-encoded . If you pass a dict as json , it will be JSON-serialised. The json argument is just a convenience shorthand for data=json.dumps(body) . Why many use the latter instead of the former… who knows.