如何在某些条件下计算 datediff？

Question

我有一个表格，它的日期时间对应于不同的状态：为每个用户点击、认领和购买。

我折叠了每个用户的记录，并将每个状态对应的每个日期时间放在他们自己的列中：clicked_date、claim_date和buy_date。我也有一个截止日期。

现在我需要另一个列来计算截止日期的天数。新列（number_days）应该总是减去deadline_date 减去clicked_date。但是，如果 clicked_date 设置为“2999-12-31”，那么它应该减去deadline_date 减去claim_date。如果clicked_date 设置为“2999-12-31”并且claimed_date 设置为=“299-12-31”，那么它应该减去截止日期减去购买日期。如果 clicked_date 设置为“2999-12-31”，claim_date 设置为“2999-12-31”，buy_date 设置为“2999-12-31”或deadline_date = '/N'，那么它应该为0。

In the examples above,
The first record should be (deadline_date - bought_date)
The second record should be (deadline_date - clicked_date) 
The third record should be 0.
The fourth record should be (deadline_date - claimed_date)
The fifth record should be (deadline_date - claimed_date) 
The sixth record should be (deadline_date - clicked_date) 

我无法制定此查询。有谁知道我如何根据点击、认领和购买的三个不同日期列中的每一个来计算天数？

我尝试了类似的方法，但它不起作用：

select 
user,
(array_agg(STATUS) within group(order by UPDATED_AT_DATETIME desc)[0])::varchar as last_status,
   coalesce(max(case when STATUS = 'clicked'  THEN UPDATED_AT_DATETIME END),'2999-12-31'::datetime) as clicked_date,
   coalesce(max(case when STATUS = 'claimed'   THEN UPDATED_AT_DATETIME END),'2999-12-31'::datetime) as claimed_date,
   coalesce(max(case when STATUS = 'bought' THEN ACCOUNT_REWARD_UPDATED_AT_DATETIME END),'2999-12-31'::datetime) as bought_date,
   **case when CLICKED_DATE <> '2999-12-31 00:00:00.000' then DATEDIFF('days', TRY_TO_TIMESTAMP(clicked_date) ,TRY_TO_TIMESTAMP(DEADLINE_DATETIME))
   when claimed_date <> '2999-12-31 00:00:00.000' then DATEDIFF('days', TRY_TO_TIMESTAMP(claimed_date) ,TRY_TO_TIMESTAMP(DEADLINE_DATETIME))
   when bought_date <> '2999-12-31 00:00:00.000' then DATEDIFF('days', TRY_TO_TIMESTAMP(bought_date) ,TRY_TO_TIMESTAMP(DEADLINE_DATETIME))
   END as number_days**
from TBL_A a
group by 1,2

Answer 1

不运行它，这应该可以工作：

SELECT 
    user,
    FIRST_VALUE(status) OVER (PARITION BY user ORDER BY updated_at_datetime DESC) AS last_status,
    MAX(IFF(status = 'clicked', UPDATED_AT_DATETIME, null)) AS clicked_date,
    MAX(IFF(status = 'claimed', UPDATED_AT_DATETIME, null)) AS claimed_date,
    MAX(IFF(status = 'bought', UPDATED_AT_DATETIME, null)) AS bought_date,
    COALESCE(bought_date, claimed_date, clicked_date, '2999-12-31'::datetime) as date_of_interest,
    DATEDIFF('days', date_of_interest, deadline_datetime) AS number_days
FROM tbl_a AS a
GROUP BY 1,2;

如果它们确实已经是字符串，则可能需要像在示例中那样转换“日期”。

“date_of_interest”可以内联。 MAX 和 COALESCE 不太可能发挥得很好，以防万一，就像子句一样：

SELECT a.*
    COALESCE(bought_date, claimed_date, clicked_date, '2999-12-31'::datetime) as date_of_interest,
    DATEDIFF('days', date_of_interest, deadline_datetime) AS number_days
FROM (
    SELECT 
        user,
        FIRST_VALUE(status) OVER (PARITION BY user ORDER BY updated_at_datetime DESC) AS last_status,
        MAX(IFF(status = 'clicked', UPDATED_AT_DATETIME, null)) AS clicked_date,
        MAX(IFF(status = 'claimed', UPDATED_AT_DATETIME, null)) AS claimed_date,
        MAX(IFF(status = 'bought', UPDATED_AT_DATETIME, null)) AS bought_date
    FROM tbl_a AS a
    GROUP BY 1,2
) AS a;