Test Data:
CREATE TABLE your_table ( usr, start_date, end_date ) AS (
SELECT 'A', DATE '2017-06-01', DATE '2017-06-03' FROM DUAL UNION ALL
SELECT 'B', DATE '2017-06-02', DATE '2017-06-04' FROM DUAL UNION ALL -- Overlaps previous
SELECT 'C', DATE '2017-06-06', DATE '2017-06-06' FROM DUAL UNION ALL
SELECT 'D', DATE '2017-06-07', DATE '2017-06-07' FROM DUAL UNION ALL -- Adjacent to previous
SELECT 'E', DATE '2017-06-11', DATE '2017-06-20' FROM DUAL UNION ALL
SELECT 'F', DATE '2017-06-14', DATE '2017-06-15' FROM DUAL UNION ALL -- Within previous
SELECT 'G', DATE '2017-06-22', DATE '2017-06-25' FROM DUAL UNION ALL
SELECT 'H', DATE '2017-06-24', DATE '2017-06-28' FROM DUAL UNION ALL -- Overlaps previous and next
SELECT 'I', DATE '2017-06-27', DATE '2017-06-30' FROM DUAL UNION ALL
SELECT 'J', DATE '2017-06-27', DATE '2017-06-28' FROM DUAL; -- Within H and I
Query:
SELECT SUM( days ) AS total_days
FROM (
SELECT dt - LAG( dt ) OVER ( ORDER BY dt ) + 1 AS days,
start_end
FROM (
SELECT dt,
CASE SUM( value ) OVER ( ORDER BY dt ASC, value DESC, ROWNUM ) * value
WHEN 1 THEN 'start'
WHEN 0 THEN 'end'
END AS start_end
FROM your_table
UNPIVOT ( dt FOR value IN ( start_date AS 1, end_date AS -1 ) )
)
WHERE start_end IS NOT NULL
)
WHERE start_end = 'end';
Output:
TOTAL_DAYS
----------
25
Explanation:
SELECT dt, value
FROM your_table
UNPIVOT ( dt FOR value IN ( start_date AS 1, end_date AS -1 ) )
This will UNPIVOT
the table so that the start and end dates are in the same column (dt
) and are given a corresponding value of +1 for a start and -1 for an end date.
SELECT dt,
SUM( value ) OVER ( ORDER BY dt ASC, value DESC, ROWNUM ) AS total,
value
FROM your_table
UNPIVOT ( dt FOR value IN ( start_date AS 1, end_date AS -1 ) )
Will give the start and end dates and the cumulative sum of those generated values. The start of a range will always have value=1
and total=1
and the end of a range will always have total=0
. If a date is mid-way through a range then it will either have total>1
or value=-1
and total=1
. Using this, if you multiply value
and total
then the start of a range is when value*total=1
and the end of a range is when value*total=0
and any other value indicates a date that is midway through a range.
Which is what this gives:
SELECT dt,
CASE SUM( value ) OVER ( ORDER BY dt ASC, value DESC, ROWNUM ) * value
WHEN 1 THEN 'start'
WHEN 0 THEN 'end'
END AS start_end
FROM your_table
UNPIVOT ( dt FOR value IN ( start_date AS 1, end_date AS -1 ) )
You can then filter out the dates when the start_end
is NULL
which will leave you with a table with alternating start
and end
rows which you can use LAG
to calculate the number of days difference:
SELECT dt - LAG( dt ) OVER ( ORDER BY dt ) + 1 AS days,
start_end
FROM (
SELECT dt,
CASE SUM( value ) OVER ( ORDER BY dt ASC, value DESC, ROWNUM ) * value
WHEN 1 THEN 'start'
WHEN 0 THEN 'end'
END AS start_end
FROM your_table
UNPIVOT ( dt FOR value IN ( start_date AS 1, end_date AS -1 ) )
)
WHERE start_end IS NOT NULL
All you need to do then is to SUM
all the differences for the end - start
; which gives the query above.