Question

我有一张这样的桌子：

TICKET ID | OPEN DATE  | CLOSE DATE
----------+------------+------------
1         | 2018-12-30 | 2019-01-01 
2         | 2019-01-30 | 2019-02-01 
3         | 2019-01-20 | 2019-01-22

我必须生成一个视图，该视图为每个票证ID为当年的每个月创建几个状态（打开，关闭，积压）的条目。如果打开日期为月份，则状态为OPEN；如果关闭日期为月份，则状态为CLOSED；如果打开日期为上个月的日期，则为LOGLOG；如果关闭日期为上个月的日期，则为STATLOG。

因此在下面的示例中，输出表将是：

TICKET ID | MONTH  | STATUS
----------+--------+---------
1         | JAN-19 | BACKLOG
1         | JAN-19 | CLOSED
2         | JAN-19 | OPEN
3         | JAN-19 | OPEN
3         | JAN-19 | CLOSED
2         | FEB-19 | CLOSED

有没有办法在SQL Server的纯SQL中做到这一点？

Answer 1

这是您想要的吗？您可以通过调整日期表来调整运行周期。

declare @tickets table (ticketId int, openDate datetime, closeDate datetime)

insert into @tickets (ticketId, openDate, closeDate) values (1, '30 Dec 2018', '1 Jan 2019')
insert into @tickets (ticketId, openDate, closeDate) values (2, '30 Jan 2019', '1 Feb 2019')
insert into @tickets (ticketId, openDate, closeDate) values (3, '20 Jan 2019', '22 Jan 2019')

; with dates as (
    select cast('1 Jan 2019' as datetime) startMon, cast('31 Jan 2019' as dateTime) endMon
    union all
    select dateAdd(mm, 1, startMon), DATEADD(dd, -1, dateAdd(mm, 2, startMon)) from dates where startMon < '01 Dec 2019'
)
select ticketId, startMon, 'BACKLOG' [Status] from @tickets t inner join dates d on t.openDate between dateAdd(m, -1, startMon) and startMon and closeDate < endMon and closeDate >= startMon
union
select ticketId, startMon, 'OPEN' from @tickets t inner join dates d on t.openDate between startMon and endMon
union
select ticketId, startMon, 'CLOSED' from @tickets t inner join dates d on t.closeDate between startMon and endMon
order by startMon, ticketId

Answer 2

您的描述表明：

    select t.ticket_id, v.mon, v.status
    from t cross apply
         (values ('OPEN', datefromparts(year(open_date), month(open_date), 1)),
                 ('CLOSED', datefromparts(year(open_date), month(open_date), 1)),
                 ('BACKLOG', datefromparts(year(dateadd(month, 1, open_date)), month(dateadd(month, 1, open_date)), 1))
        ) v(status, mon)
where status <> 'BACKLOG' or
      datediff(month, open_date, close_date) > 0;

您可以在where子句中添加特定时间段的过滤条件，例如2019年的日期。

这应该完全满足您对“待办事项”的要求：

BACKLOG（如果打开日期是上个月，而关闭日期不是上个月）。

但是，您没有考虑多个月的积压。如果有可能，请问一个 new 问题，并提供适当的样本数据和所需的结果。

SQL查询产生月份报告状态

2 个答案: