How to access the response headers?

[luzfcb]

Hi, I'm maintaining an existing application from a customer and there is an API client implementation of an external service that is implemented using tapioca-wrapper (v1.4.2). The external service now includes some headers in the response of each request and I need to perform different actions depending on the value that specific headers have.

Is there an official way to access the raw response object (not only the response body) from the TapiocaClient or TapiocaClientExecutor?

From the example below, how to access the response HEADERS from the tapioca_client instance, tapioca_client_executor instance, or from the result object of the get() method call?

from tapioca import TapiocaAdapter, generate_wrapper_from_adapter, JSONAdapterMixin
from requests.auth import HTTPBasicAuth

from .mapping import RESOURCE_MAPPING


class FooBarClientAdapter(JSONAdapterMixin, TapiocaAdapter):
    api_root = 'https://api.foobar.com/'
    resource_mapping = RESOURCE_MAPPING


FooBarClientClass = generate_wrapper_from_adapter(FooBarClientAdapter)

tapioca_client = FooBarClientClass()

url = "https://api.foobar.com/resource/123/?x=1&y=2"

tapioca_client_executor = tapioca_client._wrap_in_tapioca_executor(url)

page = 1

data = tapioca_client_executor.get(params={'page': page})

# The .response raises an exception
# tapioca_client_executor.response

[fjsj]

Have you tried accessing the response from the executor? We have a property there:

tapioca-wrapper/tapioca/tapioca.py

Lines 209 to 212 in edcbe11

	def response(self):
	if self._response is None:
	raise Exception("This instance has no response object")
	return self._response

So data().response is probably the right answer here.

To get the headers: data().response.headers.
Just tested with tapioca-github on gists endpoint and it worked fine.

[luzfcb]

@fjsj Thanks a lot. This is a good candidate for a new Cookbook section in the documentation. The current documentation is neither nearly clear nor straightforward on that particular subject.

[luzfcb]

Anyone who has the same doubts, this is a 100% working example.
Note: I used https://designer.mocky.io/design to create a mock response that includes a header.

from tapioca import TapiocaAdapter, generate_wrapper_from_adapter, JSONAdapterMixin


RESOURCE_MAPPING = {}

class FooBarClientAdapter(JSONAdapterMixin, TapiocaAdapter):
    api_root = 'https://run.mocky.io'
    resource_mapping = RESOURCE_MAPPING


FooBarClientClass = generate_wrapper_from_adapter(FooBarClientAdapter)

tapioca_client = FooBarClientClass()

url = "https://run.mocky.io/v3/927b0e28-7ec2-44fc-ab3a-d4bd386ad130"

tapioca_client_executor = tapioca_client._wrap_in_tapioca_executor(url)

new_tapioca_client_class = tapioca_client_executor.get(params={'page': 1})
new_tapioca_client_executor = new_tapioca_client_class()
print(new_tapioca_client_executor.response.headers)

[fjsj]

This is a good candidate for a new Cookbook section in the documentation. The current documentation is neither nearly clear nor straightforward on that particular subject.

Agreed @luzfcb , I'll leave the issue open due to that. Please feel free to open a PR for this additional documentation.