选择特定格式记录的sql查询（使用pivot或其他）

我有这样一个SQL表：

Animal1       Animal2       Corelation
---------+---------------+--------------
Cat             Cat           1       
Cat             Dog           0.6     
Cat             Mouse         0.8     
Dog             Cat           0.6 
Dog             Dog           1    
Dog             Mouse         0.4      
Mouse           Cat           0.8     
Mouse           Dog           0.4     
Mouse           Mouse         1   

use SymbolsDB
select * from [AnimalsTable]
pivot (
   [Corelation]
   for [Animal2] in (select * from [Animal2]
)

DECLARE @cols AS NVARCHAR(MAX),
    @query  AS NVARCHAR(MAX)

select @cols = STUFF((SELECT distinct ',' + QUOTENAME(Animal2) 
                    from animals
            FOR XML PATH(''), TYPE
            ).value('.', 'NVARCHAR(MAX)') 
        ,1,1,'')

set @query = 'SELECT Animal1, ' + @cols + ' from 
             (
                select animal1, animal2, Corelation
                from animals
            ) x
            pivot 
            (
                min(Corelation)
                for animal2 in (' + @cols + ')
            ) p '

execute(@query)

---

我正在寻找返回以下结果的SQL查询：

Animal 1       Cat              Dog               Mouse
---------+---------------+------------------+---------------+
Cat            1                  0.6               0.8
Dog            6                  1                 0.4
Mouse          0.8                0.4               1

基本上，我想要更具可读性的表格

我尝试像这样使用pivot：

Animal1       Animal2       Corelation
---------+---------------+--------------
Cat             Cat           1       
Cat             Dog           0.6     
Cat             Mouse         0.8     
Dog             Cat           0.6 
Dog             Dog           1    
Dog             Mouse         0.4      
Mouse           Cat           0.8     
Mouse           Dog           0.4     
Mouse           Mouse         1   

use SymbolsDB
select * from [AnimalsTable]
pivot (
   [Corelation]
   for [Animal2] in (select * from [Animal2]
)

DECLARE @cols AS NVARCHAR(MAX),
    @query  AS NVARCHAR(MAX)

select @cols = STUFF((SELECT distinct ',' + QUOTENAME(Animal2) 
                    from animals
            FOR XML PATH(''), TYPE
            ).value('.', 'NVARCHAR(MAX)') 
        ,1,1,'')

set @query = 'SELECT Animal1, ' + @cols + ' from 
             (
                select animal1, animal2, Corelation
                from animals
            ) x
            pivot 
            (
                min(Corelation)
                for animal2 in (' + @cols + ')
            ) p '

execute(@query)

但它不起作用。我不确定我是否了解pivot是如何工作的，是否可以在我的案例中使用它。还是有别的办法？（我试图避免循环，因为我有100万条记录）

---

感谢

在pivot中，列名必须是常量

为此，需要使用动态SQL

declare @options varchar(max) = ''
select @options  = @options + ', [' + animal2 + ']' from (select distinct animal2 from animals) v
select @options = substring(@options ,2, LEN(@options))

declare @sql nvarchar(4000) =
   'select * from [AnimalsTable] pivot (max([Corelation])
   for [Animal2] in (' + @options + ')) p'

   exec sp_executesql @sql

---

不能将

SELECT

语句放在

PIVOT

中以返回值列表，然后必须为常量。如果您正在寻找一个动态的

轴

，那么您将需要使用如下内容：

Animal1       Animal2       Corelation
---------+---------------+--------------
Cat             Cat           1       
Cat             Dog           0.6     
Cat             Mouse         0.8     
Dog             Cat           0.6 
Dog             Dog           1    
Dog             Mouse         0.4      
Mouse           Cat           0.8     
Mouse           Dog           0.4     
Mouse           Mouse         1   

use SymbolsDB
select * from [AnimalsTable]
pivot (
   [Corelation]
   for [Animal2] in (select * from [Animal2]
)

DECLARE @cols AS NVARCHAR(MAX),
    @query  AS NVARCHAR(MAX)

select @cols = STUFF((SELECT distinct ',' + QUOTENAME(Animal2) 
                    from animals
            FOR XML PATH(''), TYPE
            ).value('.', 'NVARCHAR(MAX)') 
        ,1,1,'')

set @query = 'SELECT Animal1, ' + @cols + ' from 
             (
                select animal1, animal2, Corelation
                from animals
            ) x
            pivot 
            (
                min(Corelation)
                for animal2 in (' + @cols + ')
            ) p '

execute(@query)

---

请参见

您是否希望返回所有100万条记录的结果集？（您将有1000个列）