DB_Table.running
runningstatisticofasgroup_byorder_byon_problems
Group: Values
Aliases: cumulative
Documentation
Adds a new column to the table with a running calculation.
Arguments
statistic
: The running statistic to calculate.of
: The existing column to run the statistic over.as
: The name of the new column.group_by
: Specifies the columns to group by. The running statistic is calculated separately for each group. By default, all rows are treated as a single group.order_by
: Specifies the columns to order by. Defaults to the order of the rows in the table. The running statistic is calculated according to the specified ordering.
Errors
- If the columns specified in
group_by
ororder_by
are not present in the table, aMissing_Input_Columns
error is raised. - If the column with the same name as provided by
as
already exists, aDuplicate_Output_Column_Names
problem is reported and the existing column is renamed to avoid the clash. - If grouping on floating point numbers, a
Floating_Point_Equality
problem is reported.
Remarks
Ordering of rows
Note that the ordering of rows from the original table is preserved in all cases. The grouping and ordering settings affect how the running statistic is calculated for each row, but the order of the rows itself is not changed by this operation.