The binding name and nothing else, because a pipeline has no send-time
options to default. Everything that shapes what happens to a record —
batching, the destination bucket, file format, partitioning — is configured
on the pipeline itself and applies no matter which Worker sends to it.
That is most of the reason to send through a pipeline rather than writing to
R2 directly: the buffering policy stops being something each producer decides
and starts being a property of the destination.
Settings for a Pipelines binding.
Remarks
The binding name and nothing else, because a pipeline has no send-time options to default. Everything that shapes what happens to a record — batching, the destination bucket, file format, partitioning — is configured on the pipeline itself and applies no matter which Worker sends to it.
That is most of the reason to send through a pipeline rather than writing to R2 directly: the buffering policy stops being something each producer decides and starts being a property of the destination.
Since
1.3.0