我正在尝试理解一位同事编写的 SQL，但我正在努力理解 GROUPING() 函数在这种情况下是如何工作的。为什么最后两行中date_level_year_quarter_month的值等于 3 和 7？先感谢您！
SELECT
t_year,
t_quarter,
t_month,
GROUPING(t_year, t_quarter, t_month) date_level_year_quarter_month
FROM
d_time
WHERE t_year IN (2016)
GROUP BY
ROLLUP(t_year, t_quarter, t_month)
ORDER BY 1, 2, 3;
+---------+------------+------------+-------------------------------+
| t_year | t_quarter | t_month | date_level_year_quarter_month |
+---------+------------+------------+-------------------------------+
| 2016 | 1 | 1 | 0 |
+---------+------------+------------+-------------------------------+
| 2016 | 1 | 2 | 0 |
+---------+------------+------------+-------------------------------+
| 2016 | 1 | 3 | 0 |
+---------+------------+------------+-------------------------------+
| 2016 | 1 | | 1 |
+---------+------------+------------+-------------------------------+
| 2016 | ... | ... | ... |
+---------+------------+------------+-------------------------------+
| 2016 | | | 3 |
+---------+------------+------------+-------------------------------+
| | | | 7 |
+---------+------------+------------+-------------------------------+
编辑：看起来对于“添加”的每个维度，GROUPING 函数的值是“先前分组”+ 1 值的两倍，但是我仍然不清楚该函数首先如何计算这些值。
SELECT
T_YEAR
, T_QUARTAL
, T_MONTH
, T_WEEK
, T_DATE
, GROUPING (T_YEAR ) AS LEVEL_YEAR
, GROUPING (T_YEAR, T_QUARTER) AS LEVEL_YEAR_QUARTAL
, GROUPING (T_YEAR, T_QUARTER, T_MONTH) AS LEVEL_YEAR_QUARTAL_MONTH
, GROUPING (T_YEAR, T_QUARTER, T_MONTH, T_WEEK) AS LEVEL_YEAR_QUARTAL_MONTH_WEEK
, GROUPING (T_YEAR, T_QUARTER, T_MONTH, T_WEEK, T_DAY) AS LEVEL_YEAR_QUARTAL_MONTH_WEEK_DAY
FROM D_TIME
WHERE T_YEAR=2016 AND T_WEEK = 1
GROUP BY
ROLLUP (T_YEAR, T_QUARTER, T_MONTH, T_WEEK, T_DAY)
ORDER BY 5,1,2,3,4;
