Closed emilia-friedberg closed 6 days ago
Normally, a large diff is mainly snapshot test or package.lock-json inflation, but this is a lot of code! A significant undertaking.
I think this looks PR is amazing. There's one change I'm going to recommend, but I wanted to make sure it was straightforward so I just wrote it up in a branch https://github.com/empirical-org/Empirical-Core/commit/1811426910b80287e13dbf5294f16919eb1753f0
Basically, I think you can simplify the controller create/update with nested attributes.
WHAT
Build the individual dataset page in the Evidence CMS, which displays data subsets, trials, guidelines, and related texts, and includes modal functionality for creating and updating the latter two record types.
WHY
This is the second part of the main project to pull Brendan's work into the main Evidence CMS.
HOW
Build out requisite frontend components and controller actions for all flows on this page. Bonus:
GenAIRecordBuilder
andDatasetImporter
to find or createRelevantText
records when GenAI activities are created or updated, and make sure the right ones are instantiated as the default when a newDataset
is created.RadioButton
andCheckbox
for both accessibility and ease of jest testing.I know this PR is too big. A lot of it is snapshot tests and code reorganization, but I will do a better job in future PRs for this project of splitting out work along logical divisions.
Screenshots
Notion Card Links
https://www.notion.so/quill/Product-Board-30f97e4eb01246dbb8264253fb385073?p=129d42e6f9418054bb1bd8824be496f1&pm=c
What have you done to QA this feature?
Tested all flows on staging, having Jamie QA as well.