When first running a pipeline with heterogeneous data, you might need to
do a lot of column additions. Support a batch path for updating columns
so that we don't have to fallback to retrying the whole batch - also
it makes the experience much faster.
When first running a pipeline with heterogeneous data, you might need to do a lot of column additions. Support a batch path for updating columns so that we don't have to fallback to retrying the whole batch - also it makes the experience much faster.