如何填写MySQL中两个日期之间的所有时间段?

How to fill in all periods between two dates in MySQL?

我遇到以下情况,我有一些人有开始日期和结束日期:

ID | start_date | end_date
1    2015-02-15   2015-04-20
2    2015-03-10   2015-06-15
...   ...            ...

现在,我需要得出一个 table 个人和 他们的开始日期和结束日期之间的所有连续 30 天时间段(从 start_date).结果应如下所示:

ID | period | from_date   | to_date
1     1       2015-02-15    2015-03-17
1     2       2015-03-18    2015-04-17
2     1       2015-03-10    2015-04-09
2     2       2015-04-10    2015-05-10
2     3       2015-05-11    2015-06-10

你知道如何巧妙地在 MySQL 中创建这样一个 table 吗?如果 MySQL 对于这样的数据操作来说太麻烦,R 或 Excel 也适合我。

您可以生成一个数字范围,然后将该范围与所有记录交叉连接,向该行添加与返回的数字一样多的 30 天组。

类似这样的东西(没有测试所以请原谅任何打字错误):-

SELECT a.id, b.aNum, DATE_ADD(a.start_date, INTERVAL (b.aNum * 30) DAY) AS from_date, DATE_ADD(a.start_date, INTERVAL ((b.aNum + 1) * 30) DAY) AS to_date
FROM sometable a
CROSS JOIN
(
    SELECT tens.aCnt * 10 + units.aCnt AS aNum
    FROM
    (SELECT 1 AS aCnt UNION SELECT 2 UNION SELECT 3 UNION SELECT 4 UNION SELECT 5 UNION SELECT 6 UNION SELECT 7 UNION SELECT 8 UNION SELECT 9 UNION SELECT 0) units
    CROSS JOIN
    (SELECT 1 AS aCnt UNION SELECT 2 UNION SELECT 3 UNION SELECT 4 UNION SELECT 5 UNION SELECT 6 UNION SELECT 7 UNION SELECT 8 UNION SELECT 9 UNION SELECT 0) tens
) b
WHERE  DATE_ADD(a.start_date, INTERVAL (b.aNum * 30) DAY) <= end_date

此版本最多只能支持 100 组 30 天,但可以轻松扩展(但处理的组越多,速度越慢)

我必须说的棘手问题。

这是我在 R 中使用 data.table 的尝试。首先,我会确保您的数据中有正确的日期格式

library(data.table)
indx <- grep("date", names(df))
setDT(df)[, (indx) := lapply(.SD, as.Date), .SDcols = indx]

然后,我们将计算每个 ID30 天的间隔,同时将累积索引添加到开始和结束列

df[, 
     {
      temp <- seq.Date(start_date, end_date, by = "30 days")
      indx <- seq_along(temp[-(1L:2L)])
      .(
        Period = c(indx, length(temp) - 1L),
        from = c(temp[1L], temp[-c(1L, length(temp))] + indx),
        to = c(temp[2L], temp[-c(1L:2L)] + indx)
       )
      }
, by = ID]

#    ID Period       from         to
# 1:  1      1 2015-02-15 2015-03-17
# 2:  1      2 2015-03-18 2015-04-17
# 3:  2      1 2015-03-10 2015-04-09
# 4:  2      2 2015-04-10 2015-05-10
# 5:  2      3 2015-05-11 2015-06-10

我稍微调整了 Kickstart 的代码来解决我对原始 post 的所有要求,也许它可以帮助遇到类似问题的人:

SELECT a.pid, b.aNum+1 as period, DATE_ADD(a.start_date, INTERVAL (b.aNum * 31) DAY) AS from_date, 
DATE_ADD(DATE_ADD(a.start_date, INTERVAL (b.aNum * 31) DAY), INTERVAL 30 DAY) AS to_date
FROM any_table a
CROSS JOIN
(
SELECT hundreds.aCnt*100 + tens.aCnt * 10 + units.aCnt AS aNum
FROM
(SELECT 1 AS aCnt UNION SELECT 2 UNION SELECT 3 UNION SELECT 4 UNION SELECT 5 UNION SELECT 6 UNION SELECT 7 UNION SELECT 8 UNION SELECT 9 UNION SELECT 0) units
CROSS JOIN
(SELECT 1 AS aCnt UNION SELECT 2 UNION SELECT 3 UNION SELECT 4 UNION SELECT 5 UNION SELECT 6 UNION SELECT 7 UNION SELECT 8 UNION SELECT 9 UNION SELECT 0) tens
 CROSS JOIN
(SELECT 1 AS aCnt UNION SELECT 2 UNION SELECT 3 UNION SELECT 4 UNION SELECT 5 UNION SELECT 6 UNION SELECT 7 UNION SELECT 8 UNION SELECT 9 UNION SELECT 0) hundreds
) b
WHERE  DATE_ADD(a.start_date, INTERVAL (b.aNum * 30)+30 DAY) <= end_date

现在,下一个周期在上一个周期结束后一天开始,并且个人的最后 30 天周期在 end_date 之前结束。