BQ doesn't have primary keys. Perhaps you are thinking of the id that can be supplied with the streaming insert? This has very loose guarantees on what is de-duplicated (~5m iirc)
yea I think within the context of BigQuery the most sensible thing would be to do an aggregate per the column that would be considered a primary key. For example [0]. That said, Streaming API de-dupe window is very nice in practice.
I mentioned elsewhere on Google Cloud the most elegant way of doing this is with Google Cloud Dataflow [1]
