如何根据SQL

Question

我想创建一个基本的案例记录系统，当有人打开新问题时，该问题被分配了带有给定数量的SR_NUMBER。例如，sr_number 1为4小时，2为6小时，3个是8小时，4个小时为24小时。

现在，在时间戳记上增加数小时很容易，但是我需要考虑工作时间为09:00至17:00星期一至周五。基本上，截止日期为12个工作时间。计算应在同一天登录，剩余11个小时到下一个工作日的问题进行计算。

如果是太阳，应该直接考虑到星期一。

示例：

casus创建于：10/06/2015 12:04:39 PM-与SR_NUMBER 1（12小时）截止日期为现在：10/07/2015 12.05 PM 有意义吗？

另一个捕获量是我需要搁置数小时，这两个只需在工作时间内即可。 在某些情况下，星期六有些假期正在工作。

我应该继续进行。 我尝试执行DatePart，DateAdd和日期函数。但是我只能找到工作日。 我是SQL的新手。

最简单的会计延迟方法是剩下的几个小时，然后用递归的cte ptecte几天，直到工作时间比剩下的延迟较多的工作时间。

这里是它会给出的结果（我在10月10日至11日的周末期间添加了一些日期，以显示在周末期间的时间啤酒花）：

Answer 1

Id T sr_number

delay

COMPLETE_DATE2015-09-0817：00：00.000 我希望SQL Server 2014的此查询（请在小提琴中看到）可以用作基础（甚至是工作？）：

with
    wh as
    (
        select *, datediff(second, starttime, endtime) / 3600.0 wh
        from working_hours
    ),
    delays as
    (
        select
            row_number() over (order by contact_date) id, -- Will ease reading.
            contact_date t,
            a.sr_number,
            cast(floor(cast(contact_date as float)) as datetime) d,
            cast(contact_date as time) h,
            delay
        from assign a join contract c on c.sr_number = a.sr_number
    ),
    -- The day the contact occurred is special, as the timer began to run in the middle of the day.
    firstday as
    (
        select
            *, -- wh = working hours of the current day.
            -- If contact occurred after 1 working hour with a delay of 4 hours,
            -- it's as if it occurred at working hours start but with a delay of 5 hours.
            delay + case
                when h <= wh.starttime or wh.starttime is null then 0
                when h < wh.endtime then datediff(second, starttime, h) / 3600.0
                else wh
            end as remaining
        from delays left join wh on datepart(dw, t) between wh.startday and wh.endday
    ),
    running as
    (
        select id, t, sr_number, delay, d, wh, cast(remaining as float) remaining from firstday
        union all
        select
            -- Stable part (attributes of the call):
            id, t, sr_number, delay,
            -- Evolving part, relative to the day now explored:
            d + 1,
            (
                select wh.wh from wh where datepart(dw, d + 1) between wh.startday and wh.endday
                -- Here you could put hold days (specific to Name, which would then have to be included in the CTEs):
                --and not exists (select 1 from on_hold oh where oh.Name = running.Name and d + 1 between oh.startday and oh.endday)
            ),
            -- What remains in the delay after the previous day was elapsed:
            remaining - coalesce(wh, 0.0)
        from running
        where wh is null or remaining > wh
    )
select
    id, t, sr_number, delay,
    dateadd(second, datediff(second, 0, starttime) + remaining * 3600.0, d) complete_date
from running join wh on datepart(dw, d + 1) between wh.startday and wh.endday
where running.remaining <= running.wh;

	1	2015-09-0805：39：29.000	3	8
2	2015-10-0612：04：39.000	1	4	2015-10-0616：04：39.000
3	2015-10-0700：45：00.000	3	8	2015-10-0717：00：00.000
8	2015-10-1212：00：00.000	2	6	2015-10-1310：00：00.000
7	2015-10-1012：45：00.000	4	24	2015-10-1417：00：00.000
6	2015-10-0912：45：00.000	4	24	2015-10-1412：45：00.000
5	2015-10-0906：45：00.000	4	24	2015-10-1317：00：00.000
4	2015-10-0822：45：00.000	4	24	2015-10-1317：00：00.000
	尽管我找不到SQL Server 2005，