urllib: TypeError: expected string or bytes-like object

March 21, 2023

I am trying to make an API call where I pass a file in the header

import urllib.request

headers = {'files[]': open('File.sql','rb')}
datagen ={}

request = urllib.request.Request('https://www.rebasedata.com/api/v1/convert', datagen, headers)

response1 = urllib.request.urlopen(request) 
# Error : TypeError: expected string or bytes-like object

response2 = urllib.request.urlopen('https://www.rebasedata.com/api/v1/convert', datagen, headers)
#Error: 'dict' object cannot be interpreted as an integer

The error I am retrieving is:

TypeError: expected string or bytes-like object

MEDevel.com: Open-source for Healthcare and Education

Collecting and validating open-source software for healthcare, education, enterprise, development, medical imaging, medical records, and digital pathology.
Visit Medevel

Any help will be appreciated

>Solution :

The example you’ve based your code on is antiquated, and in any case, you’ve translated into something else – you can’t read a file into a header; the example’s poster library returns a body stream and the associated headers, which is not what you’re doing.

Here’s how you’d do the equivalent POST with the popular requests library:

import requests

resp = requests.post(
    'https://www.rebasedata.com/api/v1/convert',
    files={'files[]': open('file.sql', 'rb')},
)
resp.raise_for_status()
with open('/tmp/output.zip', 'wb') as output_file:
    output_file.write(resp.content)

(If you’re expecting a large output you can’t buffer into memory, look into requests‘s stream=True mode.)

Or, alternately (and maybe preferably, these days), you could use httpx:

import httpx

with httpx.Client() as c:
    resp = c.post(
        "https://www.rebasedata.com/api/v1/convert",
        files={"files[]": open("file.sql", "rb")},
    )
    resp.raise_for_status()
    with open("/tmp/output.zip", "wb") as f:
        f.write(resp.content)