Support multiple redirect_uris when creating OAuth 2.0 Applications - #29192
Conversation
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## main #29192 +/- ##
==========================================
- Coverage 85.01% 85.01% -0.01%
==========================================
Files 1059 1060 +1
Lines 28277 28288 +11
Branches 4538 4537 -1
==========================================
+ Hits 24040 24049 +9
- Misses 3074 3075 +1
- Partials 1163 1164 +1 ☔ View full report in Codecov by Sentry. |
c43497a to
69d1781
Compare
69d1781 to
d8cc3f3
Compare
|
The following issue was discovered whilst working on the tests: #30152 |
7de288b to
e35590f
Compare
|
Working on a few additional changes after noticing some gaps when authoring the documentation for this PR. |
|
@ClearlyClaire okay, we're good — I've added:
|
This brings us more inline with the dynamic client registration specification
…ializer for /api/v1/apps and ApplicationSerializer for /api/v1/apps/verify_credentials
0f6ffae to
1b04819
Compare
ClearlyClaire
left a comment
There was a problem hiding this comment.
Looks good to me although I'm confused why this is going back on client_id being introspectable.
This brings us more inline with the dynamic client registration specification, where you can create a single application with multiple redirect URIs, see the example request at: https://datatracker.ietf.org/doc/html/rfc7591#section-3.1
This works around some weirdness in doorkeeper, where it only has a
redirect_urifield on an Application, but it does some weird "join multiple values with a \n for the database" thing.Arguably, this could be considered a breaking change because it introduces some kinda weird behavior, but it is technically the "correct" behavior with regards to OAuth 2.0...
This is currently a draft as it requires test coverage to be written, and I have had the time to write said tests yet.