fix: fix test in PR #5283#5381
Conversation
|
The latest updates on your projects. Learn more about Vercel for Git ↗︎ 3 Skipped Deployments
|
|
Uffizzi Preview |
matthewelwell
requested review from
khvn26 and
rolodato
and removed request for
a team and
khvn26
Docker builds report
|
| environment_cache.delete_many( | ||
| [self.initial_value("api_key"), *[eak.key for eak in self.api_keys.all()]] | ||
| ) |
There was a problem hiding this comment.
This is a temporary fix for now and I think we should perhaps asynchronise this. See further context in the conversation here.
There was a problem hiding this comment.
IMO the correct fix would be to only ever use a single cache key for environments, and not have one cache key per server-side SDK key.
To achieve this, the endpoints that accept server-side SDK keys must also receive the client-side key, which is confusing with the current terminology. For example, something like GET /environment-document/abc123 -H "Authorization: ser..." instead of deriving the environment ID from the server-side key.
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## feat/if-modified-since #5381 +/- ##
=========================================================
Coverage ? 97.62%
=========================================================
Files ? 1237
Lines ? 43006
Branches ? 0
=========================================================
Hits ? 41983
Misses ? 1023
Partials ? 0 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
2 participants
This is a PR on top of #5283 to resolve the tests.
There were 2 issues which I've resolved in this PR:
updated_atvalue wasn't actually by updated by just renaming the environment, as explained here)On top of this, I have also removed the
AuthenticatedRequestclass as it wasn't really being used correctly. Although we were annotating certain instances of therequestinstance with the class, it wasn't actually an instance of theAuthenticatedRequestclass. If we want to modify the behaviour, I think this would be a wider change outside of the scope of this PR.