Skip to content

Latest commit

 

History

History
42 lines (34 loc) · 1.71 KB

intro.md

File metadata and controls

42 lines (34 loc) · 1.71 KB

Introduction

We want to use APIs from web UIs that require login. What we want to do is to set up a session, login and then use the API of the page. The reason we want to do this is to use existing APIs. For instance, the user group management service at KTH, see 1{reference-type="ref" reference="UGEditor"}. We can then use the service, track the requests in the browser's developer tools. Then we can simply make the same requests from Python.

Screenshot of the KTH UG Editor with Firefox's Developer Tools open, showing network requests made.

For instance, we can redo the request in Figure 1 (above) like this:

import weblogin
import weblogin.kth
import os

ug = weblogin.AutologinSession([
                weblogin.kth.UGlogin(os.environ["KTH_LOGIN"],
                                     os.environ["KTH_PASSWD"],
                                     "https://app.kth.se/ug-gruppeditor/")
            ])

response = ug.get("https://app.kth.se/ug-gruppeditor/api/ug/users"
                  "?filter=memberOf eq 'u26yk1i3'")

The code above will access the API used by the KTH UG group editor service. It will automatically sign in when needed. The API URLs don't trigger a redirect to log in, they just give a 401 unauthorized error. However, we can use the main URL to the UI to trigger such an event, log in and then access the API. All this happens automatically in the background.

The way we do this is to subclass the requests.Session class to intercept all requests of a session to check for signs indicating that we must log in. When we detect such sign, we log in and resume as if nothing ever happened.