python requests handle error 302?

Question

I am trying to make a http request using requests library to the redirect url (in response headers-Location). When using Chrome inspection, I can see the response status is 302.

However, in python, requests always returns a 200 status. I added the allow_redirects=False, but the status is still always 200.

• The url is https://api.weibo.com/oauth2/authorize?redirect_uri=http%3A//oauth.weico.cc&response_type=code&client_id=211160679
• the first line entered the test account: [email protected]
• the second line entered the password: 112358

and then click the first button to login.

My Python code:

import requests

user_agent = 'Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/42.0.2311.152 Safari/537.36'
session = requests.session()
session.headers['User-Agent'] = user_agent
session.headers['Host'] = 'api.weibo.com'
session.headers['Origin']='https://api.weibo.com'
session.headers['Referer'] ='https://api.weibo.com/oauth2/authorize?redirect_uri=http%3A//oauth.weico.cc&response_type=code&client_id=211160679'
session.headers['Connection']='keep-alive'

data = {
    'client_id': api_key,
    'redirect_uri': callback_url,
    'userId':'[email protected]',
    'passwd': '112358',
    'switchLogin': '0',
    'action': 'login',
    'response_type': 'code',
    'quick_auth': 'null'
}

resp = session.post(
    url='https://api.weibo.com/oauth2/authorize',
    data=data,
    allow_redirects=False
)
code = resp.url[-32:]
print code

Solution

You are probably getting an API error message. Use print resp.text to see what the server tells you is wrong here.

Note that you can always inspect resp.history to see if there were any redirects; if there were any you'll find a list of response objects.

Do not set the Host or Connection headers; leave those to requests to handle. I doubt the Origin or Referer headers here needed either. Since this is an API, the User-Agent header is probably also overkill.

Comments

[Guixing Wei]

thanks for your response. I checked the resp.history, it's empty. and the content of resp.text showed that it still stay in the same page. But when I used browser to login this url, the redirect will happen.

[Martijn Pieters]

@GuixingWei: so you are missing something else. Perhaps the page uses CSRF tokens? Perhaps you need to GET the login page first to get the session cookie?

[Guixing Wei]

appreciate a lot for your suggestion. I fixed this problem. Yes,your idea is right. I need to get the cookie from login page, and then use the cookie to get the authorization page. Thanks so much for this idea!!