我刚在Oracle 10g中找到了这个，如果你用的就是它的话。为了进行电话号码比较，我必须去掉所有的特殊字符。

regexp_replace(c.phone, '[^0-9]', '')

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

下面是使用iTVF删除非字母字符的另一种方法。首先，需要一个基于模式的字符串分配器。以下是Dwain Camp文章中的一段:

-- PatternSplitCM will split a string based on a pattern of the form 
-- supported by LIKE and PATINDEX 
-- 
-- Created by: Chris Morris 12-Oct-2012 
CREATE FUNCTION [dbo].[PatternSplitCM]
(
       @List                VARCHAR(8000) = NULL
       ,@Pattern            VARCHAR(50)
) RETURNS TABLE WITH SCHEMABINDING 
AS 

RETURN
    WITH numbers AS (
        SELECT TOP(ISNULL(DATALENGTH(@List), 0))
            n = ROW_NUMBER() OVER(ORDER BY (SELECT NULL))
        FROM
        (VALUES (0),(0),(0),(0),(0),(0),(0),(0),(0),(0)) d (n),
        (VALUES (0),(0),(0),(0),(0),(0),(0),(0),(0),(0)) e (n),
        (VALUES (0),(0),(0),(0),(0),(0),(0),(0),(0),(0)) f (n),
        (VALUES (0),(0),(0),(0),(0),(0),(0),(0),(0),(0)) g (n)
    )

    SELECT
        ItemNumber = ROW_NUMBER() OVER(ORDER BY MIN(n)),
        Item = SUBSTRING(@List,MIN(n),1+MAX(n)-MIN(n)),
        [Matched]
    FROM (
        SELECT n, y.[Matched], Grouper = n - ROW_NUMBER() OVER(ORDER BY y.[Matched],n)
        FROM numbers
        CROSS APPLY (
            SELECT [Matched] = CASE WHEN SUBSTRING(@List,n,1) LIKE @Pattern THEN 1 ELSE 0 END
        ) y
    ) d
    GROUP BY [Matched], Grouper

现在你有了一个基于模式的拆分器，你需要拆分匹配模式的字符串:

[a-z]

然后将它们连接起来以得到想要的结果:

SELECT *
FROM tbl t
CROSS APPLY(
    SELECT Item + ''
    FROM dbo.PatternSplitCM(t.str, '[a-z]')
    WHERE Matched = 1
    ORDER BY ItemNumber
    FOR XML PATH('')
) x (a)

样本

结果:

| Id |              str |              a |
|----|------------------|----------------|
|  1 |    test“te d'abc |     testtedabc |
|  2 |            anr¤a |           anra |
|  3 |  gs-re-C“te d'ab |     gsreCtedab |
|  4 |         M‚fe, DF |          MfeDF |
|  5 |           R™temd |          Rtemd |
|  6 |          ™jad”ji |          jadji |
|  7 |      Cje y ret¢n |       Cjeyretn |
|  8 |        J™kl™balu |        Jklbalu |
|  9 |       le“ne-iokd |       leneiokd |
| 10 |   liode-Pyr‚n‚ie |    liodePyrnie |
| 11 |         V„s G”ta |          VsGta |
| 12 |        Sƒo Paulo |        SoPaulo |
| 13 |  vAstra gAtaland | vAstragAtaland |
| 14 |  ¥uble / Bio-Bio |     ubleBioBio |
| 15 | U“pl™n/ds VAsb-y |    UplndsVAsby |

我知道SQL不擅长字符串操作，但我没想到它会这么难。下面是一个简单的函数，用于从字符串中剥离所有数字。当然还有更好的办法，但这只是个开始。

CREATE FUNCTION dbo.AlphaOnly (
    @String varchar(100)
)
RETURNS varchar(100)
AS BEGIN
  RETURN (
    REPLACE(
      REPLACE(
        REPLACE(
          REPLACE(
            REPLACE(
              REPLACE(
                REPLACE(
                  REPLACE(
                    REPLACE(
                      REPLACE(
                        @String,
                      '9', ''),
                    '8', ''),
                  '7', ''),
                '6', ''),
              '5', ''),
            '4', ''),
          '3', ''),
        '2', ''),
      '1', ''),
    '0', '')
  )
END
GO

-- ==================
DECLARE @t TABLE (
    ColID       int,
    ColString   varchar(50)
)

INSERT INTO @t VALUES (1, 'abc1234567890')

SELECT ColID, ColString, dbo.AlphaOnly(ColString)
FROM @t

输出

ColID ColString
----- ------------- ---
    1 abc1234567890 abc

第2轮-数据驱动黑名单

-- ============================================
-- Create a table of blacklist characters
-- ============================================
IF EXISTS (SELECT * FROM sys.tables WHERE [object_id] = OBJECT_ID('dbo.CharacterBlacklist'))
  DROP TABLE dbo.CharacterBlacklist
GO
CREATE TABLE dbo.CharacterBlacklist (
    CharID              int         IDENTITY,
    DisallowedCharacter nchar(1)    NOT NULL
)
GO
INSERT INTO dbo.CharacterBlacklist (DisallowedCharacter) VALUES (N'0')
INSERT INTO dbo.CharacterBlacklist (DisallowedCharacter) VALUES (N'1')
INSERT INTO dbo.CharacterBlacklist (DisallowedCharacter) VALUES (N'2')
INSERT INTO dbo.CharacterBlacklist (DisallowedCharacter) VALUES (N'3')
INSERT INTO dbo.CharacterBlacklist (DisallowedCharacter) VALUES (N'4')
INSERT INTO dbo.CharacterBlacklist (DisallowedCharacter) VALUES (N'5')
INSERT INTO dbo.CharacterBlacklist (DisallowedCharacter) VALUES (N'6')
INSERT INTO dbo.CharacterBlacklist (DisallowedCharacter) VALUES (N'7')
INSERT INTO dbo.CharacterBlacklist (DisallowedCharacter) VALUES (N'8')
INSERT INTO dbo.CharacterBlacklist (DisallowedCharacter) VALUES (N'9')
GO

-- ====================================
IF EXISTS (SELECT * FROM sys.objects WHERE [object_id] = OBJECT_ID('dbo.StripBlacklistCharacters'))
  DROP FUNCTION dbo.StripBlacklistCharacters
GO
CREATE FUNCTION dbo.StripBlacklistCharacters (
    @String nvarchar(100)
)
RETURNS varchar(100)
AS BEGIN
  DECLARE @blacklistCt  int
  DECLARE @ct           int
  DECLARE @c            nchar(1)

  SELECT @blacklistCt = COUNT(*) FROM dbo.CharacterBlacklist

  SET @ct = 0
  WHILE @ct < @blacklistCt BEGIN
    SET @ct = @ct + 1

    SELECT @String = REPLACE(@String, DisallowedCharacter, N'')
    FROM dbo.CharacterBlacklist
    WHERE CharID = @ct
  END

  RETURN (@String)
END
GO

-- ====================================
DECLARE @s  nvarchar(24)
SET @s = N'abc1234def5678ghi90jkl'

SELECT
    @s                  AS OriginalString,
    dbo.StripBlacklistCharacters(@s)   AS ResultString

输出

OriginalString           ResultString
------------------------ ------------
abc1234def5678ghi90jkl   abcdefghijkl

我对读者的挑战是:你能让这个过程更有效率吗?那么使用递归呢?

Here's a solution that doesn't require creating a function or listing all instances of characters to replace. It uses a recursive WITH statement in combination with a PATINDEX to find unwanted chars. It will replace all unwanted chars in a column - up to 100 unique bad characters contained in any given string. (E.G. "ABC123DEF234" would contain 4 bad characters 1, 2, 3 and 4) The 100 limit is the maximum number of recursions allowed in a WITH statement, but this doesn't impose a limit on the number of rows to process, which is only limited by the memory available. If you don't want DISTINCT results, you can remove the two options from the code.

-- Create some test data:
SELECT * INTO #testData 
FROM (VALUES ('ABC DEF,K.l(p)'),('123H,J,234'),('ABCD EFG')) as t(TXT)

-- Actual query:
-- Remove non-alpha chars: '%[^A-Z]%'
-- Remove non-alphanumeric chars: '%[^A-Z0-9]%'
DECLARE @BadCharacterPattern VARCHAR(250) = '%[^A-Z]%';

WITH recurMain as (
    SELECT DISTINCT CAST(TXT AS VARCHAR(250)) AS TXT, PATINDEX(@BadCharacterPattern, TXT) AS BadCharIndex
    FROM #testData
    UNION ALL
    SELECT CAST(TXT AS VARCHAR(250)) AS TXT, PATINDEX(@BadCharacterPattern, TXT) AS BadCharIndex
    FROM (
        SELECT 
            CASE WHEN BadCharIndex > 0 
                THEN REPLACE(TXT, SUBSTRING(TXT, BadCharIndex, 1), '')
                ELSE TXT 
            END AS TXT
        FROM recurMain
        WHERE BadCharIndex > 0
    ) badCharFinder
)
SELECT DISTINCT TXT
FROM recurMain
WHERE BadCharIndex = 0;

DECLARE @vchVAlue NVARCHAR(255) = 'SWP, Lettering Position 1: 4 Ω, 2: 8 Ω, 3: 16 Ω, 4:  , 5:  , 6:  , Voltage Selector, Solder, 6, Step switch, : w/o fuseholder '


WHILE PATINDEX('%?%' , CAST(@vchVAlue AS VARCHAR(255))) > 0
  BEGIN
    SELECT @vchVAlue = STUFF(@vchVAlue,PATINDEX('%?%' , CAST(@vchVAlue AS VARCHAR(255))),1,' ')
  END 

SELECT @vchVAlue