Remove Duplicates

Scenario

Files (or database records) can often show up with duplicate data. Often it is OK, and sometimes it is required to ignore duplicate records.

Duplicate data, or data with duplicate keys are a feature of most enterprise systems. PhixFlow has lots of ways of dealing with duplicated data, and how a model uses them depends entirely on the system requirements. In this case, we just want to ignore duplicates.

Solution:

• Load all data (including duplicates) into a stream
• Create a new stream from this stream - make it an aggregate stream.
• Make the pipe linking the 2 streams an aggregate pipe, grouped on the field with duplicated data, and sorted by another field, depending on which record you want. e.g. if you want the lastest record, you could sort by the updated date.
• In the second stream, reference data coming from the input pipe using an array index . i.e. in[1].value to just retrieve the first of the grouped records.