An earlier email I sent had this schema:

    <xs:element name="file1">
        <xs:complexType>
          <xs:sequence dfdl:separator="," dfdl:separatorPosition="infix"
            dfdl:separatorSuppressionPolicy="never">
            <xs:element name="given-name" type="xs:string" minOccurs="0" maxOccurs="3" />
            <xs:element name="surname" type="xs:string" minOccurs="0"/>
            <xs:element name="phone" type="xs:string" minOccurs="0" maxOccurs="6" />
          </xs:sequence>
        </xs:complexType>
      </xs:element>

Given that, and a default dfd:format with occursCountKind='implicit' I expect this data:

"madonna,,,,,,,,,"

To produce this DFDL Infoset (cast as XML):

<file1>

  <given-name>madonna</given-name>

</file1>

I don't expect to see any empty elements like <given-name></given-name> because all elements are optional, so a zero-length representation for a xs:string is the Empty representation, and optional empty is never added to the infoset.

Furthermore, this should "round trip" i.e., unparse back to the original input data.

Am I correct here?