Stamp PlateRecord onto inventory plate

Stamp a plate record’s (or plate design’s) snapshot onto an existing inventory plate. Every layer the source plate carries is applied; a plate record or plate design only carries a subset of layers, and layer types it doesn’t have are left untouched on the target plate. Content and volumes from the plate’s content layer become the plate’s authoritative state for that dimension: they’re written onto the target plate’s wells, and any well the layer doesn’t cover is cleared to match. A plate with no content layer at all doesn’t capture that dimension, so the target plate’s existing contents and volumes are left as they are. A label layer sets the target plate’s well experimental roles, schema tags, and volumes. Column mappings map columns from any plate layer onto the target plate’s existing container-schema fields.

The scratch file must be a JSONL file where each line is a JSON object with the following structure:

{“request_id”: “unique-id-per-row”, “input_object”: {}}

The input_object fields (all camelCase) are:

• plateRecordId (required): ID of the plate record whose snapshot should be stamped onto the target plate.
• inventoryPlateId or inventoryPlateBarcode (exactly one required): ID or barcode of the existing inventory plate to stamp. The plate must be writable, must be a well plate (not a matrix plate), and its dimensions must exactly match the source plate record; otherwise the row returns a 400. Each inventory plate can be stamped at most once per request.
• columnMappings: List of {layerName, columnName, schemaFieldName} objects mapping plate record layer columns onto the target plate’s container-schema fields.