I am connecting to MongoDB Atlas using mongodb-spark-connector v10.3
I am using a SamplePartitioner with a pipeline that matches on a certain date range and projects only the _id column: