docs-scraper
docs-scraper copied to clipboard
Keycloak support is broken due to unmaintained library
Description
TL;DR: This scraper used to support Keycloak SSO, but now it no longer does (unless you're running a very old Keycloak).
Support for Keycloak SSO was added in #73 by @spahrson and it is a great addition that we're super appreciative of 🎉
Unfortunately, the python-keycloak-client library used to add this functionality is de-facto unmaintained, which is a problem because Keycloak introduced a breaking change in Keycloak v17 which changed the URLs under which to find the OIDC endpoints.
Given that:
- The Keycloak 17.0 release dates from February 2022
- Releases prior to 17 (aka pre-Quarkus releases) have been deprecated ever since
- The latest Keycloak release is 25.0.1
The end result is that this no longer supports keycloak SSO in a meaningful way.
I was hoping we could get that support back. Ways to do so include:
- Migrate to a maintained Keycloak library
- Monkey-patch the existing library to strip the leading
auth/prefix from the URLs
Other I'm not familiar with the Python ecosystem, so I cannot really offer to do it myself. But I am happy to help testing, or update docs, or help get this done somehow.