sqlserver中的搜索序列模式

我有这样的记录，我想在SQL Server中按模式和顺序搜索，如（50,54,50）在值字段中，它应该返回02,03,04任何人都可以这样做

======================================
Id             Date             Value
01             2020-01-01       50
02             2020-01-02       50
03             2020-01-03       54
04             2020-01-04       50
05             2020-01-05       35
06             2020-01-06       98
07             2020-01-07       13
======================================

---

如果我得到你的要求正确，你可以尝试下面的逻辑开发的帮助下滞后和领先-

---

在用户语音网站上有一个请求，我相信这会允许的

同时，这应该满足您的需要

WITH T AS
(
SELECT *, 
       LAG(Id) OVER (ORDER BY Id) AS PrevId, 
       LAG(value) OVER (ORDER BY Id) AS PrevValue,
       LEAD(Id) OVER (ORDER BY Id) AS NextId, 
       LEAD(value) OVER (ORDER BY Id) AS NextValue
FROM YourTable
)
SELECT PrevId, Id, NextId
FROM T
WHERE PrevValue = 50 AND Value =54 AND NextValue = 50

---

如果您想要更灵活的方法，可以使用

交叉应用

：

select t2.*
from t cross apply
     (select string_agg(id, ',') within group (order by date) as ids,
             string_agg(value, ',') within group (order by date) as vals
      from (select top (3) t2.*
            from t t2
            where t2.date >= t.date
            order by t2.date
           ) t2
     ) t2
where vals = '50,54,50';

他是一把小提琴

如果

string\u agg（）

作为窗口函数受支持，则可以使用：

select t.*
from (select t.*,
             string_agg(id, ',') within group (order by date) over (order by id rows between current row and 2 following) as ids,
             string_agg(value, ',') within group (order by date) over (order by id rows between current row and 2 following) as vals
      from t
     ) t
where vals = '50,54,50';

---

但遗憾的是，事实并非如此。

你尝试过什么？你在哪里卡住了？请为你的逻辑和要求添加更多的解释，看起来IN子句可以满足你的要求。如果Where子句中的序列太长，这是不可能的。那么你要匹配多长的序列？@MartinSmith。我希望看到对数组的完全支持，这也可以用于此目的。如果where子句中的序列太长，这是不可能的。请提及最坏的可能性。是的，我也考虑了

string\u agg

。他们可能只是

string_agg

整批，然后根据序列可能包含的行数使用正则表达式来查找他们的模式

select t.*
from (select t.*,
             string_agg(id, ',') within group (order by date) over (order by id rows between current row and 2 following) as ids,
             string_agg(value, ',') within group (order by date) over (order by id rows between current row and 2 following) as vals
      from t
     ) t
where vals = '50,54,50';