Mysql 查询以获取'；大于零的不为空

我有一个问题：

SELECT
    taskDeadline,
    subtasksDeadline,
    REPLACE(
        LEAST(
            COALESCE(t.deadline, 2147483647),
            COALESCE(sub.deadline, 2147483647)
        ), 2147483647, NULL
    ) AS Deadline
FROM
    tasks t
    LEFT OUTER JOIN subtasks sub ON sub.task_id = t.id

这会产生如下结果：

taskDeadline | subtasksDeadline | Deadline
1338501600   | 1333058400       | 1333058400
1334268000   | NULL             | 1334268000
NULL         | 1328050800       | 1328050800
NULL         | NULL             | NULL
1353798000   | 0                | 0
0            | 1353798000       | 0

---

我快到了。我使用2147483647作为数据类型（Mysql INT）的最大值，唯一的问题是，如果我有两个值，其中一个值为0，那么我想要得到另一个值。我已经花了一些时间，但无法到达那里。有人能解释一下吗？

如果我理解正确，您不希望

t.deadline

和

sub.deadline

的值为零，如果零是最小值。所以您可以将0转换为NULL，这将从

COALESCE

SELECT
    taskDeadline,
    subtasksDeadline,
    REPLACE(
        LEAST(
            COALESCE(IF(t.deadline = 0, NULL, t.deadline), 2147483647),
            COALESCE(IF(sub.deadline = 0, NULL, sub.deadline), 2147483647)
        ), 2147483647, NULL
    ) AS Deadline
FROM
    tasks t
    LEFT OUTER JOIN subtasks sub ON sub.task_id = t.id

---

如果我理解正确，您不希望

t.deadline

和

sub.deadline

的值为零（如果零是最小值）。所以您可以将0转换为NULL，这将从

COALESCE

SELECT
    taskDeadline,
    subtasksDeadline,
    REPLACE(
        LEAST(
            COALESCE(IF(t.deadline = 0, NULL, t.deadline), 2147483647),
            COALESCE(IF(sub.deadline = 0, NULL, sub.deadline), 2147483647)
        ), 2147483647, NULL
    ) AS Deadline
FROM
    tasks t
    LEFT OUTER JOIN subtasks sub ON sub.task_id = t.id

---

为什么不使用案例陈述

SELECT
    taskDeadline,
    subtasksDeadline,
    CASE
        WHEN t.deadline IS NULL AND sub.deadline IS NULL THEN NULL
        WHEN t.deadline IS NULL THEN sub.deadline
        WHEN sub.deadline IS NULL THEN t.deadline
        WHEN t.deadline > sub.deadline THEN sub.deadline
        WHEN sub.deadline > t.deadline THEN t.deadline
        WHEN sub.deadline = t.deadline THEN sub.deadline -- or this could be t.deadline
     END AS Deadline
FROM
    tasks t
    LEFT OUTER JOIN subtasks sub 
        ON sub.task_id = t.id

---

为什么不使用案例陈述

SELECT
    taskDeadline,
    subtasksDeadline,
    CASE
        WHEN t.deadline IS NULL AND sub.deadline IS NULL THEN NULL
        WHEN t.deadline IS NULL THEN sub.deadline
        WHEN sub.deadline IS NULL THEN t.deadline
        WHEN t.deadline > sub.deadline THEN sub.deadline
        WHEN sub.deadline > t.deadline THEN t.deadline
        WHEN sub.deadline = t.deadline THEN sub.deadline -- or this could be t.deadline
     END AS Deadline
FROM
    tasks t
    LEFT OUTER JOIN subtasks sub 
        ON sub.task_id = t.id

---

尝试使用此查询-

SELECT
    taskDeadline,
    subtasksDeadline,
    LEAST(
      IF(t.deadline = 0 OR t.deadline IS NULL, sub.deadline, t.deadline),
      IF(sub.deadline = 0 OR sub.deadline IS NULL, t.deadline, sub.deadline)
    ) AS Deadline
FROM
    tasks t
    LEFT OUTER JOIN subtasks sub ON sub.task_id = t.id

---

尝试使用此查询-

SELECT
    taskDeadline,
    subtasksDeadline,
    LEAST(
      IF(t.deadline = 0 OR t.deadline IS NULL, sub.deadline, t.deadline),
      IF(sub.deadline = 0 OR sub.deadline IS NULL, t.deadline, sub.deadline)
    ) AS Deadline
FROM
    tasks t
    LEFT OUTER JOIN subtasks sub ON sub.task_id = t.id

---

或者只是将NULLIF添加到现有合并中-

SELECT
    taskDeadline,
    subtasksDeadline,
    NULLIF(
        LEAST(
            COALESCE(NULLIF(t.deadline, 0), 2147483647),
            COALESCE(NULLIF(sub.deadline, 0), 2147483647)
        ), 2147483647
    ) AS Deadline
FROM
    tasks t
    LEFT OUTER JOIN subtasks sub ON sub.task_id = t.id

---

或者只是将NULLIF添加到现有合并中-

SELECT
    taskDeadline,
    subtasksDeadline,
    NULLIF(
        LEAST(
            COALESCE(NULLIF(t.deadline, 0), 2147483647),
            COALESCE(NULLIF(sub.deadline, 0), 2147483647)
        ), 2147483647
    ) AS Deadline
FROM
    tasks t
    LEFT OUTER JOIN subtasks sub ON sub.task_id = t.id

---

更多信息：我一直认为案例的速度会很慢。这就是我跳过此选项的原因。取决于数据集的大小。它完全满足了您解决这个特定问题的需要，而且可读性也更高。更多信息：我一直认为CASE什么时候会比较慢。这就是我跳过此选项的原因。取决于数据集的大小。它完全满足了您解决这个特定问题的需要，而且可读性更高。快速简单。我更喜欢这个解决方案。我不知道NULLIF，谢谢！又快又容易。我更喜欢这个解决方案。我不知道NULLIF，谢谢！