multipart response preserve carriage return linefeed pairs

[rimmartin]

Hi,
Trying to process the response that is a multipart/form-data.

    multipart_data = MultipartDecoder.from_response(response)
    for part in multipart_data.parts:
        print(part.content)
        print(part.headers)

which fails with requests_toolbelt.multipart.decoder.ImproperBodyPartContentException: content does not contain CR-LF-CR-LF
And when I try to get to the body of the response(runtime version 2.32.3) any other way the "\r\n"'s don't exist.
response headers are

HTTPHeaderDict({'Content-Length': '120640', 'Connection': 'Keep-Alive', 'Content-Type': 'multipart/form-data; boundary=e9bb939421e7436ebdbec45d4142245e'})
status=200

Started with conda env yesterday

conda create -n pytest_env python
conda activate pytest_env
conda install -c conda-forge requests-toolbelt pytest requests -y

        response = requests.post(
            endpoint,
            data=mpb,
            headers=headers,
            verify=False,  # Disable SSL verification for HTTPS
            timeout=50,  # Set timeout to 50 seconds
            stream=True
        )

is the POST request and it sends over a multpart via toolbelt's MultipartEncoder.

After response returns tried sending content to binary file out and see no "\r\n"'s there either.

    with open("debug.txt", "wb") as debug_file:
        for chunk in response:
            debug_file.write(chunk)

I've written nodejs & Java clients that leave the response "\r\n"'s alone and I have choices of how the response body gets processed.

Is there a setting for requests that corrects this behavior? Many suggestions out on the wild web but none successful so far

[github-actions]

As described in the template, we won't be able to answer questions on this issue tracker. Please use Stack Overflow