这个解决方案受到Allen先生的解决方案的启发，需要一个整数的Numbers表(如果您想进行具有良好性能的严肃查询操作，您应该手头有这个表)。它不需要CTE。您可以更改NOT IN(…)表达式以排除特定字符，或将其更改为IN(…)只保留某些字符的OR LIKE表达式。

SELECT (
    SELECT  SUBSTRING([YourString], N, 1)
    FROM    dbo.Numbers
    WHERE   N > 0 AND N <= CONVERT(INT, LEN([YourString]))
        AND SUBSTRING([YourString], N, 1) NOT IN ('(',')',',','.')
    FOR XML PATH('')
) AS [YourStringTransformed]
FROM ...

虽然这篇文章有点老了，但我想说以下几点。 我有上述解决方案的问题是，它没有过滤出字符，如ç， ë， ï等。我调整了一个函数如下(我只使用80 varchar字符串来节省内存):

create FUNCTION dbo.udf_Cleanchars (@InputString varchar(80)) 
RETURNS varchar(80) 
AS 

BEGIN 
declare @return varchar(80) , @length int , @counter int , @cur_char char(1) 
SET @return = '' 
SET @length = 0 
SET @counter = 1 
SET @length = LEN(@InputString) 
IF @length > 0 
BEGIN WHILE @counter <= @length 

BEGIN SET @cur_char = SUBSTRING(@InputString, @counter, 1) IF ((ascii(@cur_char) in (32,44,46)) or (ascii(@cur_char) between 48 and 57) or (ascii(@cur_char) between 65 and 90) or (ascii(@cur_char) between 97 and 122))
BEGIN SET @return = @return + @cur_char END 
SET @counter = @counter + 1 
END END 

RETURN @return END

我把它放在调用PatIndex的两个地方。

PatIndex('%[^A-Za-z0-9]%', @Temp)

为上面的自定义函数RemoveNonAlphaCharacters并重命名为RemoveNonAlphaNumericCharacters

信不信由你，在我的系统中，这个丑陋的函数比G masters的优雅函数表现得更好。

CREATE FUNCTION dbo.RemoveSpecialChar (@s VARCHAR(256)) 
RETURNS VARCHAR(256) 
WITH SCHEMABINDING
    BEGIN
        IF @s IS NULL
            RETURN NULL
        DECLARE @s2 VARCHAR(256) = '',
                @l INT = LEN(@s),
                @p INT = 1

        WHILE @p <= @l
            BEGIN
                DECLARE @c INT
                SET @c = ASCII(SUBSTRING(@s, @p, 1))
                IF @c BETWEEN 48 AND 57
                   OR  @c BETWEEN 65 AND 90
                   OR  @c BETWEEN 97 AND 122
                    SET @s2 = @s2 + CHAR(@c)
                SET @p = @p + 1
            END

        IF LEN(@s2) = 0
            RETURN NULL

        RETURN @s2

使用CTE生成的数字表来检查每个字符，然后FOR XML连接到一个保留值的字符串，您可以…

CREATE FUNCTION [dbo].[PatRemove](
    @pattern varchar(50),
    @expression varchar(8000) 
    )
RETURNS varchar(8000)
AS
BEGIN
    WITH 
        d(d) AS (SELECT d FROM (VALUES (0),(1),(2),(3),(4),(5),(6),(7),(8),(9)) digits(d)),
        nums(n) AS (SELECT ROW_NUMBER() OVER (ORDER BY (SELECT NULL)) FROM d d1, d d2, d d3, d d4),
        chars(c) AS (SELECT SUBSTRING(@expression, n, 1) FROM nums WHERE n <= LEN(@expression))
    SELECT 
        @expression = (SELECT c AS [text()] FROM chars WHERE c NOT LIKE @pattern FOR XML PATH(''));

    RETURN @expression;
END

SQL Server 2017+的另一个可能的选项，没有循环和/或递归，是使用TRANSLATE()和REPLACE()的基于字符串的方法。

t - sql声明:

DECLARE @pattern varchar(52) = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ'

SELECT 
   v.[Text], 
   REPLACE(
      TRANSLATE(
         v.[Text],
         REPLACE(TRANSLATE(v.[Text], @pattern, REPLICATE('a', LEN(@pattern))), 'a', ''),
         REPLICATE('0', LEN(REPLACE(TRANSLATE(v.[Text], @pattern, REPLICATE('a', LEN(@pattern))), 'a', '')))
      ),
      '0',
      ''
   ) AS AlphabeticCharacters
FROM (VALUES
   ('abc1234def5678ghi90jkl#@$&'),
   ('1234567890'),
   ('JAHDBESBN%*#*@*($E*sd55bn')
) v ([Text])

或作为一个函数:

CREATE FUNCTION dbo.RemoveNonAlphabeticCharacters (@Text varchar(1000)) 
RETURNS varchar(1000)
AS BEGIN

   DECLARE @pattern varchar(52) = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ'
   SET @text = REPLACE(
      TRANSLATE(
         @Text,
         REPLACE(TRANSLATE(@Text, @pattern, REPLICATE('a', LEN(@pattern))), 'a', ''),
         REPLICATE('0', LEN(REPLACE(TRANSLATE(@Text, @pattern, REPLICATE('a', LEN(@pattern))), 'a', '')))
      ),
      '0',
      ''
   )
   
   RETURN @Text
END