Sql server 查询以从包含ORDERBY子句的当前行中查找上一行和下一行

我执行以下查询：

SELECT quotesid 
FROM quotestable 
WHERE quotesauthor LIKE '%Mahatma%'  
ORDER BY popularity DESC

这给了我以下结果：quotesid-是主键

 968
 158
 159
 160
 161

我的要求： 我需要下一个和上一个quotesid顺序，例如，如果我在158记录。下一个和上一个值应该分别是159和968，我需要在单个查询中使用这两个值

我的问题 这是行不通的。这只给了我一张唱片——159张。 有没有一种方法可以编写单个查询并获得所需的结果

小提琴：

---

我想这会给你你想要的：

;WITH t AS (
    SELECT *, ROW_NUMBER() OVER (ORDER BY popularity DESC) rn
    FROM quotestable
    WHERE quotesauthor LIKE '%Mah%')
SELECT *
FROM t
WHERE 1 = ABS(t.rn - (SELECT rn FROM t WHERE quotesid = 158))

您可以使用此选项将上一行和下一行合并为一行：

;WITH t AS (
    SELECT *, ROW_NUMBER() OVER (ORDER BY popularity DESC) rn
    FROM quotestable
    WHERE quotesauthor LIKE '%Mah%')
SELECT t1.quotesid, t2.quotesid PrevID, t3.quotesid NextID
FROM t t1 
    LEFT JOIN t t2 ON t1.rn = t2.rn - 1
    LEFT JOIN t t3 ON t1.rn = t3.rn + 1
WHERE t1.rn = (SELECT rn FROM t WHERE quotesid = 158)

---

也许，这对你有用。它给出您的查询行号索引，然后将外部联接本身左移+1和-1记录，以生成[next]和[previous]列

结果:

quotesid Next   Previous
------------------------
968      158    NULL
158      159    968
159      160    158
160      161    159
161      NULL   160

sql fiddle：

---

从quesid>=158或quesid order by primary key的表格@venkatselect*发布您想要的输出。问题是我需要一个不同键的订单。我建议你再增加一个名为“受欢迎程度和顺序”的专栏。它会起作用，因为你的人气是1,2,3,4,5，。请将100200,90101,5之类的值混洗，然后重试。@Venkat ok更新sql FIDLE并回答。原则上，查询是按流行度描述a排序的。当流行度不为空时，它似乎工作得很好。如果该值为空，则无法正常工作。@Venkat我检查了您的链接，但只能看到单选，没有数据。我还尝试了空值查看此链接：并且似乎效果良好。这不起作用。让我输入SQLFIDLE并向这里的每个人打开这个查询。

with cte as
(
    select quotesid, ROW_NUMBER() over (order by popularity desc) as [RowNo]
    from quotestable
)
select cte.quotesid, [next].quotesid as [Next], [prev].quotesid as [Previous]
from cte left outer join cte as [next] on cte.RowNo = ([next].RowNo - 1)
         left outer join cte as [prev] on cte.RowNo = ([prev].RowNo + 1)

quotesid Next   Previous
------------------------
968      158    NULL
158      159    968
159      160    158
160      161    159
161      NULL   160

select quotestable.quotesid, b.quotesid nextQuoteID, c.quotesid previousQuoteID from quotestable
join (
    select a.quotesID, max(b.popularity) previousQuoteID, min(c.popularity) nextQuoteID
    from quotestable a
    left join quotestable b
    on a.popularity > b.popularity
    left join quotestable c
    on a.popularity < c.popularity
    where a.author like '%Mahatma%'
    group by a.quotesid
) x
on quotestable.quotesid = x.quotesid
left join quotestable b
on b.popularity = x.previousquoteid
left join quotestable c
on c.popularity = x.nextquoteid
order by quotestable.popularity desc