outputOrder

Specifies the order by which the data should be presented.

Not specifying a value or specifying the empty set will output the data in the order it was received.

Consider a pivot table with the dimensions "country" and "region" and the measure "turnover". Specifying an output order of "-1" creates an output in which the country with the highest turnover is listed first. Within that country the region with the highest turnover is listed first and within each country the individual fact rows are ordered in descending order by the turnover value.

Note: Specifying an output order may cause large latency and memory consumption on large input data.

When an output order is specified, it is possible to specify a cutoff value called topN that limits the number of items displayed. Continuing with our example, a cutoff value of 2 would limit the output to the two top countries; within each country the two top regions; and within each region to the two highest fact rows.

Note: Specifying an output order currently limits the number of displayable aggregations to one per dimension. If more are specified, one is picked following a priority list. See computeTotal.

Type: Order specifier