Question

为什么这在SQLite中不起作用，但在MySQL中起作用？如果您能帮助我，那就太好了！

其他问题：我可以用联接替换子选择吗？

SELECT h1.drivers.name, SUM((
    SELECT kilometers FROM history AS h2
    WHERE h2.timestamp >= h1.timestamp
    AND h1.id != h2.id
    GROUP BY h2.timestamp HAVING MIN(h2.timestamp) LIMIT 1
    ) - h1.kilometers) AS driven_kilometers
FROM history AS h1
INNER JOIN drivers ON drivers.id = h1.driver_id
GROUP BY h1.driver_id

为澄清我的问题：

╔════╦═════════╦═════════════════════╦═══════════════╗
║ id ║driver_id║      timestamp      ║ kilometers    ║
╠════╬═════════╬═════════════════════╬═══════════════╣
║  1 ║    1    ║ 2018-07-01 12:00:04 ║             0 ║
║  2 ║    2    ║ 2018-07-01 16:05:12 ║           120 ║
║  3 ║    1    ║ 2018-07-06 11:13:59 ║           220 ║
║  4 ║    2    ║ 2018-07-09 09:17:19 ║           250 ║
║  5 ║    1    ║ 2018-07-12 10:02:00 ║           300 ║
║  6 ║    2    ║ 2018-07-16 07:11:21 ║           320 ║
╚════╩═════════╩═════════════════════╩═══════════════╝

我想计算行驶的公里数。司机开始开车时会添加一个数据集，以便插入其他人开车的公里数。

示例：驾驶员2的公里数我要计算：220-120 + 300-250 = 150

驾驶员1的公里我要计算：120-0 + 250-220 + 320-300 = 170

Answer 1

在子查询中，分组没有意义，因为时间戳始终是唯一的。

要从具有下一个最大时间戳的行中获取值，只需将ORDER BY与LIMIT一起使用：

SELECT drivers.name,
       SUM((SELECT kilometers
            FROM history AS h2
            WHERE h2.timestamp > h1.timestamp
            ORDER BY h2.timestamp 
            LIMIT 1
           ) - h1.kilometers
          ) AS driven_kilometers
FROM history AS h1
INNER JOIN drivers ON drivers.id = h1.driver_id
GROUP BY h1.driver_id;

查询在SQLite中不起作用

1 个答案: