Server should not longer crash. Major issues with front end are also handled.
In details:
Asynchronous API calls in the front end has dependencies (analytics functions require the data to be present in the database, which is not recorded until the contributor hits submit). Big cause of crashing in prod. This was changed so that Analytics functions only run once the submission has gone through.
Removed clustering on start. This was the main cause of prod crashing. Instead, a cron job should be used. #88
The experience API endpoint was not called with real data, causing the test data to be used instead. It now uses actual submission data.
The experience API endpoint was not implemented to spec. It now is.
Experience did not count the number of submissions in the cluster properly. This is now fixed.
5 experience clusters was not enough to represent the seed data. It is now 7. The objective function is still a little too high, so more data would be awesome here.
Stuff I caught but out of scope:
Experience pie chart in the Complete screen does not work properly. Specifically, the colors are completely wrong
What does this PR do?
Server should not longer crash. Major issues with front end are also handled. In details:
Stuff I caught but out of scope:
Github Issue Number
Resolves: #88 (everything else I caught on the fly, so no issues were made)
Relates to:
Author checklist
If any of these points have yet to be satisfied, make sure that you set the title to the format DRAFT #issue-num Title
Reviewer checklist
docker compose up --build
)