Compare value in next row and update the current row | Oracle SQL

Question

I have the data in the following format in my database:

Name    Values          Start_of_week  End_of_week
Name1   1_2_2_1_1_2_1   22-Dec-19      28-Dec-19
Name1   1_2_2_1_2_2_1   29-Dec-19      04-Jan-20
Name1   1_2_2_2_2_2_1   05-Jan-20      11-Jan-20
Name1   1_2_2_2_2_2_1   12-Jan-20      18-Jan-20
Name1   1_2_2_2_2_2_1   19-Jan-20      25-Jan-20
Name1   1_2_2_2_2_2_1   26-Jan-20      01-Feb-20
Name1   1_2_2_2_2_2_1   02-Feb-20      08-Feb-20
Name1   1_2_2_2_2_2_1   09-Feb-20      15-Feb-20
Name1   1_2_2_2_2_2_1   16-Feb-20      22-Feb-20
Name1   1_2_2_2_2_2_1   23-Feb-20      29-Feb-20
Name1   1_2_2_2_2_2_1   01-Mar-20      07-Mar-20
Name2   1_2_2_1_1_2_1   22-Dec-19      28-Dec-19
Name2   1_2_2_2_2_2_2   29-Dec-19      04-Jan-20
Name2   1_2_2_2_2_2_2   05-Jan-20      11-Jan-20
Name2   1_2_2_2_2_2_2   12-Jan-20      18-Jan-20
Name2   1_2_2_2_2_2_2   19-Jan-20      25-Jan-20
Name2   1_2_2_2_2_2_2   26-Jan-20      01-Feb-20
Name2   1_2_2_2_2_2_2   02-Feb-20      08-Feb-20
Name2   1_2_2_2_2_2_2   09-Feb-20      15-Feb-20
Name2   1_2_2_2_2_2_2   16-Feb-20      22-Feb-20
Name2   1_2_2_2_2_2_2   23-Feb-20      29-Feb-20
Name2   1_2_2_2_2_2_2   01-Mar-20      07-Mar-20

I need the values column to be compared for each name column and update the end_of_week. For example, first row and second row has different values column so no need to update the end_of_week column. Third and fourth column has same values column, so the fourth row's end of week should be updated for the third row resulting like this.

Then this row should be compared with the next row and if the values column is same, then end_of_week should be taken from the next row and updated in this row. This should be happening for every set of rows of each name column value.

I tried to compare the rows using the lead() function but unable to compare with the next set of rows after update.

Name   Values         start_of_week  end_of_week
Name1  1_2_2_1_1_2_1  22-Dec-19      28-Dec-19
Name1  1_2_2_1_2_2_1  29-Dec-19      04-Jan-20
Name1  1_2_2_2_2_2_1  05-Jan-20      07-Mar-20
Name2  1_2_2_1_1_2_1  22-Dec-19      28-Dec-19
Name2  1_2_2_2_2_2_2  29-Dec-19      07-Mar-20

Answer 1

This is a gaps-and-islands problem. A simple solution is the difference of row numbers:

select name, value,
       min(week_start), max(week_end)
from (select t.*,
             row_number() over (partition by name order by week_start) as seqnum,
             row_number() over (partition by name, value order by week_start) as seqnum_2
      from t
     ) t
group by name, value, (seqnum - seqnum_2);

Why this works is a little tricky to explain. But if you look at the results of the subquery, you will see how the difference of row numbers identifies adjacent rows with the same values.

Answer 2

Looking at sample data, I think It is not a gap and island problem. You can achieve the desired output using group by .

Select name, value,
       Min(start_of_week),
       Max(end_of_week)
From your_table
Group by name, value;

Cheers!!