This step merges input formats of two defined inputs into a single output format. This merge is based on defined columnMappings. Data are read from their individual inputs (in_a or in_b) and written to the output. Therefore, in the output there are merged-format data records, where each row contains data read from either in_a or in_b input.

        <step id='i1' className='cz.adastra.cif.tasks.flow.Union'>
                <properties>
                        <columnMappings>
                                <columnMapping in_a='name' dest='name'/>
                                <columnMapping in_b='surname' dest='surname'/>
                        </columnMappings>
                </properties>
        </step>