The import infrastructure uses formatters to interpret the incoming data and convert it for insertion into the database. If you use the CSV or TSV formatter, you can control how the data is interpreted by setting additional properties associated with those formatters. For example, the following configuration for the Kafka importer includes the formatter property blank specifying that blank entries should generate an error, rather than being interpreted as null or empty values:

deployment:
  import:
    configuration:
    - nickname: employees
      type: kafka
      format: csv
      enabled: true
      property:
      - name: brokers
        value: kafkasvr:9092
      - name: topics
        value: employees
      - name: procedure
        value: EMPLOYEE.insert
      - name: blank
        value: error

You include the formatter properties in the import configuration along with the other import connector properties. Table 15.9, “CSV and TSV Formatter Properties” lists the allowable properties for the CSV and TSV import formatters.