I am looking for a solution on row aggregation as follows. We have in the example of the recording of three users.
We look at the author's ID: if the next line (s) belong to the author and the dates in these next lines are adjacent, then there must be a cluster of adjacent (or current dates). October 12,14 are not neighboring, 10,10,10,11,11,11,12 are neighboring. 14,15,16,16,16,16 - neighboring.
then we aggregate, otherwise we skip.
If a cluster of neighboring dates is found in the author's records, then replace it with the record:
- the first column is the sum of the values of all the numbers in the first column of the cluster of records
- the second is the average date.
- The third is the sum of the numbers in the corresponding column.
- The fifth is the sum of the numbers in the corresponding column.
That is, instead of records we insert one "aggregated" record. All other values of this record in other columns are put down with values from any record from the cluster (the same values are there).
