SQL 2005在分隔符上拆分逗号分隔列

My google searches on how to split a string on a delimiter have resulted in some useful functions for splitting strings when the string is known (i.e. see below):

我的谷歌搜索如何在分隔符上拆分字符串已经导致一些有用的功能,用于在字符串已知时分割字符串(即见下文):

SET ANSI_NULLS ON
GO
SET QUOTED_IDENTIFIER ON
GO
ALTER FUNCTION [dbo].[Split] (@String varchar(8000), @Delimiter char(1))     
   returns @temptable TABLE (items varchar(8000))       
   as       
   begin       
       declare @idx int       
        declare @slice varchar(8000)       

        select @idx = 1       
            if len(@String)<1 or @String is null  return       

       while @idx!= 0       
       begin       
           set @idx = charindex(@Delimiter,@String)       
           if @idx!=0       
               set @slice = left(@String,@idx - 1)       
           else       
              set @slice = @String       

           if(len(@slice)>0)  
               insert into @temptable(Items) values(@slice)       

           set @String = right(@String,len(@String) - @idx)       
           if len(@String) = 0 break       
       end   
   return       
   end

This works well for a known string like:

这适用于已知字符串,如:

SELECT TOP 10 * FROM dbo.Split('This,Is,My,List',',')

However, I would like to pass a column to a function, and have it unioned together with my other data in it's own row... for example given the data:

但是,我想将一个列传递给一个函数,并将它与我自己的行中的其他数据联合在一起...例如,给定数据:

CommaColumn   ValueColumn1   ValueColumn2
-----------   ------------   -------------
ABC,123       1              2
XYZ, 789      2              3

I would like to write something like:

我想写一些类似的东西:

SELECT Split(CommaColumn,',') As SplitValue, ValueColumn1, ValueColumn2 FROM MyTable

And get back

然后回来

SplitValue    ValueColumn1   ValueColumn2
----------    ------------   ------------
ABC           1              2
123           1              2
XYZ           2              3
789           2              3

Is this possible, or has anyone done this before?

这可能,或者之前有人这样做过吗?

5 个解决方案

#1

Yes, it's possible with CROSS APPLY (SQL 2005+):

是的,可以使用CROSS APPLY(SQL 2005+):

with testdata (CommaColumn, ValueColumn1, ValueColumn2) as (
  select 'ABC,123', 1, 2 union all
  select 'XYZ, 789', 2, 3
  ) 
select 
  b.items as SplitValue
, a.ValueColumn1
, a.ValueColumn2
from testdata a
cross apply dbo.Split(a.CommaColumn,',') b

Notes:

You should add an index to the result set of your split column, so that it returns two columns, IndexNumber and Value.

您应该为拆分列的结果集添加索引,以便它返回两列,IndexNumber和Value。
In-line implementations with a numbers table are generally faster than your procedural version here.

使用数字表的内联实现通常比此处的过程版本更快。

eg:

create function [dbo].[Split] (@list nvarchar(max), @delimiter nchar(1) = N',')
returns table
as
return (
  select 
    Number = row_number() over (order by Number)
  , [Value] = ltrim(rtrim(convert(nvarchar(4000),
        substring(@list, Number
        , charindex(@delimiter, @list+@delimiter, Number)-Number
        )
    )))
  from dbo.Numbers
  where Number <= convert(int, len(@list))
    and substring(@delimiter + @list, Number, 1) = @delimiter
  )

Erland Sommarskog has the definitive page on this, I think: http://www.sommarskog.se/arrays-in-sql-2005.html

我认为Erland Sommarskog有关于此的最终页面:http://www.sommarskog.se/arrays-in-sql-2005.html

#2

Fix it the right way--make that column a related table. No good ever comes of comma-separated scalar columns.

以正确的方式修复它 - 使该列成为一个相关的表。以逗号分隔的标量列没有好处。

#3

+1 to the anti-CSV comments, but if you must do this you would use CROSS APPLY or OUTER APPLY.

+1反垃圾评论,但如果你必须这样做,你将使用CROSS APPLY或OUTER APPLY。

#4

alter procedure [dbo].[usp_split](@strings varchar(max)) as  
begin  
    Declare @index int  
    set @index=1  
    declare @length int  
    set @length=len(@strings)  
    declare @str varchar(max)  
    declare @diff int  
    declare @Tags table(id varchar(30))  
    while(@index<@length)  
    begin  
        if(@index='1')  
        begin  
            set @str=(SELECT substring(@strings, @index, (charindex(',',(substring(@strings, @index,(@length)))))-1))  
            insert into @Tags values(@str)  
                set @index=(charindex(',',(substring(@strings, @index,(@length)))))  
        end  
        else  
        begin  
            set @diff=@length-@index  
            if(@diff !=0)  
            begin  
                set @str=(select substring(@strings, @index, (charindex(',',(substring(@strings,@index,@diff))))-1))  
                if(@str is not null and @str!='')  
                begin  
                    insert into @Tags VALUES(@str)  
                end  
                set @index=@index +(charindex(',',(substring(@strings, @index,@diff))))  
            end  
        end
    end
    set @str=(select right(@strings,(charindex(',',(substring(reverse(@strings),1,(@length)))))-1)) 
    insert into @Tags VALUES(@str)   
    select id from @Tags  
end

Usage:

exec usp_split '1212,21213,1,3,133,1313131,1,231313,5'

#5

You could try something like:

你可以尝试类似的东西:

SELECT s.Items AS SplitValue, ValueColumn1, ValueColumn2 
FROM MyTable, Split(CommaColumn,',') AS s

#1