The essence of pivot data analysis is the capability to perform hierarchical grouping, filtering and sorting. When row or column data groupings are processed, each data grouping operates on a subset of the data set records. Data groupings are processed from parent to children (e.g. outer to inner), which subsequently decreases the number of records in the current scope.

The pivot processing starts with a FULL recordset. The FULL recordset is then filtered with the global pivot filters. The resulting recordset is the input scope for all root row and column data groupings. To better describe grouping, filtering and sorting we will assume that we have a pivot with no global filters, which is connected to the following fictional data set:

The id column is a 1-based index of the records in the table. We will be using the id column to define the records and their order in a recordset (scope).

The samples in this chapter denote the recordset in this format: recordset={r1,r2..rn} - for example the FULL recordset in the default data set order is fullRS={1,2,3,4,5,6,7,8,9}, while the recordset for the Maison company records is maisonRS={4,5,6}.