mirror of
https://github.com/AmanTahiliani/box-box.git
synced 2026-08-07 19:56:18 -04:00
79 lines
2.3 KiB
Markdown
79 lines
2.3 KiB
Markdown
|
# Prompt For Cursor: Phase 11 Weekend Ingestion
|
||
|
|
|
||
|
|
You are working in the `box-box` repository on Phase 11. The Web UI can now
|
||
|
|
browse local years, meetings, and sessions. Your task is to make the CLI able
|
||
|
|
to populate a whole race weekend/meeting in one backend ingestion flow.
|
||
|
|
|
||
|
|
## Read First
|
||
|
|
|
||
|
|
Open these files first:
|
||
|
|
|
||
|
|
- `documentations/refactor/19-phase-11-weekend-ingestion.md`
|
||
|
|
- `cmd/main.go`
|
||
|
|
- `internal/ingest/ingest.go`
|
||
|
|
- `internal/ingest/ingest_test.go`
|
||
|
|
- `internal/ingest/openf1.go`
|
||
|
|
- `internal/store/meetings.go`
|
||
|
|
- `internal/query/navigation.go`
|
||
|
|
|
||
|
|
Only open older docs if you are blocked.
|
||
|
|
|
||
|
|
## Goal
|
||
|
|
|
||
|
|
Make meeting/weekend ingestion useful for the local-first Web UI. A user should
|
||
|
|
be able to ingest a meeting and have all sessions for that meeting populated
|
||
|
|
with Race Hub datasets where available.
|
||
|
|
|
||
|
|
## Current Shape
|
||
|
|
|
||
|
|
The project already has:
|
||
|
|
|
||
|
|
- `--ingest-year`
|
||
|
|
- `--ingest-meeting`
|
||
|
|
- `--ingest-session`
|
||
|
|
- session-level Race Hub ingestion datasets;
|
||
|
|
- local navigation APIs and UI that depend on ingested meeting/session data.
|
||
|
|
|
||
|
|
Confirm the exact current behavior before editing. If `--ingest-meeting`
|
||
|
|
currently only stores meeting/session metadata, extend it or add a clearly named
|
||
|
|
flag. Prefer the least surprising CLI behavior.
|
||
|
|
|
||
|
|
## Work To Do
|
||
|
|
|
||
|
|
1. Add meeting/weekend orchestration that fetches sessions for a meeting and
|
||
|
|
ingests Race Hub datasets for each session.
|
||
|
|
2. Preserve single-session ingestion behavior.
|
||
|
|
3. Return/report per-session summaries clearly.
|
||
|
|
4. Keep raw payload provenance for all fetched endpoints.
|
||
|
|
5. Make partial failures visible without erasing successful session data.
|
||
|
|
6. Add focused offline tests with fake sources.
|
||
|
|
|
||
|
|
## Guardrails
|
||
|
|
|
||
|
|
- Do not fetch OpenF1 from React.
|
||
|
|
- Do not introduce background ingestion from normal page views.
|
||
|
|
- Do not persist high-volume car telemetry in this phase.
|
||
|
|
- Do not break existing e2e seed behavior.
|
||
|
|
- Keep completed historical sessions as the default mental model.
|
||
|
|
|
||
|
|
## Verification
|
||
|
|
|
||
|
|
Run:
|
||
|
|
|
||
|
|
```bash
|
||
|
|
go test ./internal/ingest/... ./internal/store/... ./internal/query/... ./internal/web/...
|
||
|
|
go build -o /private/tmp/box-box ./cmd/main.go
|
||
|
|
cd frontend && npm test -- --run
|
||
|
|
cd frontend && npm run build
|
||
|
|
npm run test:e2e
|
||
|
|
```
|
||
|
|
|
||
|
|
## Report Back
|
||
|
|
|
||
|
|
Summarize:
|
||
|
|
|
||
|
|
- files changed;
|
||
|
|
- CLI behavior added or changed;
|
||
|
|
- tests run and results;
|
||
|
|
- follow-up risks, especially around OpenF1 rate limits or partial sessions.
|