Question

目前我有一个包含3个主要维度的交易列表：Keys，Time_Index和Cumulative_Amount。

Keys      Time_Index     Cumulative_Amount
100          1          103
100          3          222
101          2          101

我正在尝试“填充所有Time_Index”。在这种情况下，虽然在时间2关键100没有发生任何事情，但我想填充它。结果应如下（最大时间指数= 3）：

Keys      Time_Index     Cumulative_Amount
100          1          103
100          2          103
100          3          222
101          2          101
101          3          101

请注意，我没有为密钥101填充时间1，因为密钥101存在的最早时间是时间2.我将填充每个密钥，直到max_time_index = 3。

Answer 1

I would first generate a list of all possible combinations of Keys and Time_Index, followed by a usual RIGHT / LEFT OUTER JOIN where you can do a condition for NULL. Something like this:

with t as
(
    select distinct x = a.number, y = b.number
    from master..[spt_values] a
    cross join 
    master..[spt_values] b
    where a.number between 1 and 4
    and b.number between 100 and 103
)
select
    case when x.Cumulative_Amount is null then -1 else x.Cumulative_Amount  end, t.Keys, t.Time_Index
from dbo.myTable x 
right join t on t.Keys = x.Keys

对于按时间索引的SQL表，添加行以填充缺少的时间索引

1 个答案: