chore(discover): hide Farcaster tab from Trending#7414
chore(discover): hide Farcaster tab from Trending#7414DanielSinclair merged 4 commits intodevelopfrom
Conversation
|
π§ͺ Flashlight Performance Report (AWS Device Farm) π Commit: 6c7b68a π View Artifacts
|
|
Trying to understand the intent here a bit better. The ticket and PR both describe what's changing (hide the pill, reset persisted selections) but I'm not totally sure what's driving it? Product call ahead of the discover-v2 TF, or something else? I'd assumed perf at first (BACK-1700) but AFAIU that ticket is about the per-token Also slightly confusing that this is the base of the discover-v2 stack, since #7415-#7420 don't seem to touch TrendingTokens or anything farcaster. |
ibrahimtaveras00
left a comment
ibrahimtaveras00
left a comment
There was a problem hiding this comment.
Looks good on both OS's β
|
@olerass This is the prerequisite for Discover v2; this Trending UI is fully redesigned in Phase 2 and this feature/datasource is currently deprecated and not maintained by backend. The existing Trending algorithm still remains somewhat active but there will be another PR for the new Trending UI and algorithm (TBD) later on. #7417 is where we start to implement the new UI on the Discover screen, so there is some component overlap there. |
ahh got it, that helps. Backend deprecation is a good driver, wasn't clear from the ticket/description π . One Q to check my understanding: when you say deprecated/unmaintained, does that also cover the per-token farcaster stuff ( If yes, prob worth scoping this PR a bit bigger. If no, fine as-is though 18n |
DanielSinclair
force-pushed
the
daniel/hide-farcaster
branch
from
d9d174d to
0b326f4
Compare
|
@olerass The backend tickets you had mentioned were fairly dated from the old Trending work, so closed them. We do still rely on the "farcaster friends" lookup for the current Trending list, and that will soon be deprecated/removed, but I did find a few other dead code path cleanups that I completed here for the Farcaster and Rainbow trending categories to simplify that further. |
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
π‘ Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 0b326f43b3
βΉοΈ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with π.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
| const { category } = useTrendingTokensStore.getState(); | ||
| const isStalePersistedCategory = !categories.some(availableCategory => availableCategory === category); | ||
| if (isStalePersistedCategory) { | ||
| useTrendingTokensStore.getState().setCategory(categories[0]); |
There was a problem hiding this comment.
For users whose persisted trending-tokens store still contains the removed Rainbow or Farcaster category, this cleanup runs only in a useEffect, after TrendingTokenData has already rendered and useTrendingTokensData() has constructed/fired the query with the stale category. That is especially problematic for Rainbow now that the special fetchRainbowTokens branch was removed: it gets sent as $category: TrendingCategory to the Arc GraphQL query (src/graphql/queries/arc.graphql:506), even though it was only a local string category before. Normalize during store hydration/migration, or clamp the category before building the query args, so stale persisted values never reach the query layer.
Useful? React with πΒ / π.
3 participants
APP-3655
What changed
Screen recordings / screenshots
What to test