- Updated the `scoreQuote` function to incorporate price scoring, allowing quotes to be evaluated based on selling and cost prices when the query is numeric.
- Introduced new weights for item selling and cost prices, ensuring they contribute to the overall score only when relevant.
- Enhanced the test suite to cover new price matching scenarios, ensuring accurate scoring behavior for quotes based on price inputs.
- Refactored related functions to support the new price scoring logic, maintaining consistency across the scoring process.
When the operator clicks "Mark as ready" on a buyer-tasks po_sent card
and fills in the note, the item should land in the collection pool
(`item_status: "collection_pool"` + `ready_for_collection: true`) so
a driver can be assigned. Previously the service set
`item_status: "ready_delivery"`, which was rendered as the word
"Collected" on /orders and on /orders tabs — operators reasonably
read that as "driver already picked it up", and the item never
appeared on /collection-pool where they expected to assign a driver.
Now:
- markGroupReady writes a single PB update with the same shape as
/api/v1/collection-map/items/[id]/assign (collection_pool +
ready_for_collection: true + notes)
- po_sent buyer-tasks filter excludes collection_pool and assigned,
so marked-ready items drop out of buyer-tasks and surface on
/collection-pool instead
- ITEM_STATUS_VISUAL.ready_delivery is labelled "Ready for
Collection" (kept distinct from the actual `collected` state) for
any other flow that legitimately sets this status (e.g. stock-out)
- new item-status-ui.test.ts locks down the label/tab mapping so
the regression can't return
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Co-authored-by: Orca <help@stably.ai>
