The Dataflow step “Filter” is a powerful tool for deleting unneeded data according to a certain logic. For example, if you receive warehouse transaction data from a source system for two storage locations, but want to look at both areas separately (for example, with a key figure for one area and the other separately), you would typically create two data flows and delete the data for the other area using a filter right at the beginning. This turns a mixed data source into two separate data sources.
The logic works with a Lua expression like in the Create Colum step. Only the return value changes. If the expression returns false, the line is deleted, if it returns true, it remains in the resultset. The following example shows how to delete all rows that contain the value “N/A” in the City column.
Filter By Date/Time Columns
Data in a dataflow can also be filtered using a previously created [time data source] (https://help.peakboard.com/data_sources/en-date-and-time.html). For this, the time format of the data source must be identical to the comparison time from the dataflow.