Sql 内置函数，用于大写每个单词的第一个字母

如果SQL Server中已经存在这样的函数，我不想为此创建自定义函数

输入字符串：这是我要转换的字符串

---

预期输出：这是我要转换的字符串

AFAIK，SQL Server没有用于此的内置函数。 您必须为它编写自定义函数

试试这个

CREATE FUNCTION [dbo].[CamelCase]
(@Str varchar(8000))
RETURNS varchar(8000) AS
BEGIN
  DECLARE @Result varchar(2000)
  SET @Str = LOWER(@Str) + ' '
  SET @Result = ''
  WHILE 1=1
  BEGIN
    IF PATINDEX('% %',@Str) = 0 BREAK
    SET @Result = @Result + UPPER(Left(@Str,1))+
    SubString  (@Str,2,CharIndex(' ',@Str)-1)
    SET @Str = SubString(@Str,
      CharIndex(' ',@Str)+1,Len(@Str))
  END
  SET @Result = Left(@Result,Len(@Result))
  RETURN @Result
END  

输出：

---

我必须去不，那是不存在的。这是基于几年来对T-SQL中可用字符串函数的仔细阅读，以及SQLServer2008R2中最近的一些为期5天的课程

当然，我仍然可能是错的：。

我的策略

如果名称的大小写已经混合，请相信它是正确的。 如果名称不是混合大小写，请执行以下操作： 修剪名称以消除空白 说明以“Mc”开头的名称，如“McDavid” 用像O'Reilly这样的撇号来解释名字 连字符姓名的帐户已婚姓名“Anderson Johnson” 解释多个单词名称，如“La Russa” 确保名称字段中包含的后缀适当大写 代码

以下是我的原始帖子：

---

如果操作的目标是美化名称字符串，那么可以将适当的大写字母定义为每个单词的第一个字母，由非字母字符分隔

其他解决方案未考虑：

保留间距，尤其是尾随间距。 保留NULL、空字符串或仅包含空格的字符串。 处理的不仅仅是空格，例如破折号、逗号、下划线等。。。 处理单词/标记之间的多个非字母字符。 处理例外情况，如James William中的McDonald或III 下牙III。 注意：我的解决方案不处理异常。 如果您非常关心这些，那么我建议为它们编写一个CLR C程序集，因为这将是一个棘手的问题，而字符串是C擅长的领域。 这里的另一个解决方案试图解释这一点，但它仍然需要ivan可怕的iv和输出**iv***可怕的iv*

这就是我提出的功能：

IF  EXISTS (SELECT * FROM sys.objects WHERE object_id = OBJECT_ID(N'[dbo].[fs_PascalCase]') AND type in (N'FN', N'IF', N'TF', N'FS', N'FT'))
DROP FUNCTION [dbo].[fs_PascalCase]
GO
SET ANSI_NULLS ON
GO
SET QUOTED_IDENTIFIER ON
GO
CREATE FUNCTION [dbo].[fs_PascalCase]
(
    @Text nVarChar(MAX)
)
RETURNS nVarChar(MAX)
AS
BEGIN
        SET @Text = LOWER(@Text)--This step is optional.  Keep if you want the code below to control all casing. - 11/26/2013 - MCR.
    DECLARE @New nVarChar(MAX) = (CASE WHEN @Text IS NULL THEN NULL ELSE '' END)--Still return null when source is null. - 11/26/2013 - MCR.
    DECLARE @Len   Int = LEN(REPLACE(@Text, ' ', '_'))--If you want to count/keep trailing-spaces, you MUST use this!!! - 11/26/2013 - MCR.
    DECLARE @Index Int = 1--Sql-Server is 1-based, not 0-based.
    WHILE (@Index <= @Len)
        IF (SUBSTRING(@Text, @Index, 1) LIKE '[^a-z]' AND @Index + 1 <= @Len)--If not alpha and there are more character(s).
            SELECT @New = @New + UPPER(SUBSTRING(@Text, @Index, 2)), @Index = @Index + 2
        ELSE
            SELECT @New = @New +       SUBSTRING(@Text, @Index, 1) , @Index = @Index + 1

    --If @Text is null, then @Len will be Null, and everything will be null.
    --If @Text is '',   then (@Len - 1) will be -1, so ABS() it to use 1 instead, which will still return ''.
    RETURN ( UPPER(LEFT(@New, 1)) + RIGHT(@New, ABS(@Len - 1)) )
END
GO

输出如下所示：

---

这里有一件简单的事情，不要把它复杂化

甲骨文：

---

从dual中选择initcaplower“这是我要转换的字符串”

使用SQL 2017，函数可以如下所示：

create function dbo.cap_words (@str varchar(max))
returns varchar(max)
as
begin
    declare @result varchar(max);
    select @result = string_agg( upper(left(value,1)) + substring(value,2,999),' ') from string_split(lower(@str),' ') 
    return @result;
end

---

“从技术上讲，这就是帕斯卡卡塞。”托马斯，从技术上讲，这两者都不是，只是一串大写字母words@nathan冈萨雷斯-我站在正确的立场上。我喜欢你这样做。然而，这段代码并没有在SQLServer2008R2下为我执行。错误为必须声明标量变量@NameCamelCase..Thank。我试图让它变得简单一点，但错过了一些东西。现在更新…投票赞成：如果名称已经是混合大小写，请相信它是正确的。很好，但只将空格作为分隔符处理。例如，不会处理破折号/连字符，但仍会回答问题。也不会处理NULL，而是返回一个空字符串。也不会处理单字母字符串，即第一个-中间的姓氏作为单独的字符串：dbo.CamelCasep.FirstName、dbo.CamelCasep.LastName、dbo.CamelCasep.middlename问题是关于sqlserver而不是oracle。您可以删除答案以避免得到负分

CREATE FUNCTION [dbo].[GetCamelCaseName]
(
    @Name varchar(50)
)
RETURNS VARCHAR(50) WITH SCHEMABINDING
AS
BEGIN
    -- Declare the return variable here
    DECLARE @NameCamelCase VARCHAR(50)  

    -- This is determining whether or not the name is in camel case already (if the 1st character is uppercase
    -- and the third is lower (because the 2nd could be an apostrophe).  To do this, you have to cast the 
    -- character as varbinary and compare it with the upper case of the character cast as varbinary.  

    IF (CAST(SUBSTRING(@Name, 1,1) as varbinary) = CAST(SUBSTRING(UPPER(@Name), 1, 1) as varbinary)         
            AND ((CAST(SUBSTRING(@Name, 2,1) as varbinary) = CAST(SUBSTRING(LOWER(@Name), 2, 1) as varbinary)
                    AND SUBSTRING(@Name, 2,1) != '''')
                or
                (CAST(SUBSTRING(@Name, 4,1) as varbinary) = CAST(SUBSTRING(LOWER(@Name), 4, 1) as varbinary)
                    AND SUBSTRING(@Name, 2,1) = '''')))

        BEGIN
            SELECT @NameCamelCase = RTRIM(LTRIM(@Name))
            SELECT @NameCamelCase = REPLACE(@NameCamelCase, ' sr', ' Sr')           
            SELECT @NameCamelCase = REPLACE(@NameCamelCase, ' jr', ' Jr')       
            SELECT @NameCamelCase = REPLACE(@NameCamelCase, ' ii', ' II')   
            SELECT @NameCamelCase = REPLACE(@NameCamelCase, ' iii', ' III')
            SELECT @NameCamelCase = REPLACE(@NameCamelCase, ' DE ', ' de ')
            SELECT @NameCamelCase = REPLACE(@NameCamelCase, 'macdonald', 'MacDonald')

            if (@NameCamelCase LIKE '% iv') -- avoid changing "Ivan" to "IVan"
                SELECT @NameCamelCase = REPLACE(@NameCamelCase, ' iv', ' IV')

            if ((@NameCamelCase = 'i') or (@NameCamelCase = 'ii') or (@NameCamelCase = 'iii') or (@NameCamelCase = 'iv'))
                SELECT @NameCamelCase = UPPER(@NameCamelCase)

            RETURN @NameCamelCase       

        END

    ELSE

        BEGIN       

            SELECT @NameCamelCase = RTRIM(LTRIM(@Name))

            -- "Mc-"
            SELECT @NameCamelCase = 
                CASE 
                    WHEN @Name LIKE 'mc%'
                        THEN UPPER(SUBSTRING(@Name, 1, 1)) + LOWER(SUBSTRING(@Name, 2, 1)) + UPPER(SUBSTRING(@Name, 3, 1))  + LOWER(SUBSTRING(@Name, 4, 47))
                    ELSE
                       UPPER(SUBSTRING(@Name, 1, 1)) + LOWER(SUBSTRING(@Name, 2, 49))
                END

            -- Apostrophes
            SELECT @NameCamelCase = 
                CASE 
                    WHEN @NameCamelCase LIKE '%''%'
                        THEN SUBSTRING(@NameCamelCase, 1, CHARINDEX('''', @NameCamelCase) - 1) + ''''  + UPPER(SUBSTRING(@NameCamelCase, CHARINDEX('''', @NameCamelCase) + 1, 1)) + SUBSTRING(@NameCamelCase, CHARINDEX('''', @NameCamelCase) + 2, 50)
                    ELSE
                        @NameCamelCase
                END 


            -- Hyphenated names (do it twice to account for double hyphens)
            SELECT @NameCamelCase = 
                CASE 
                    WHEN @NameCamelCase LIKE '%-%'
                        THEN SUBSTRING(@NameCamelCase, 1, CHARINDEX('-', @NameCamelCase) - 1) + '^'  + UPPER(SUBSTRING(@NameCamelCase, CHARINDEX('-', @NameCamelCase) + 1, 1)) + SUBSTRING(@NameCamelCase, CHARINDEX('-', @NameCamelCase) + 2, 50)
                    ELSE
                        @NameCamelCase
                END 

            SELECT @NameCamelCase = 
                CASE 
                    WHEN @NameCamelCase LIKE '%-%'
                        THEN SUBSTRING(@NameCamelCase, 1, CHARINDEX('-', @NameCamelCase) - 1) + '^'  + UPPER(SUBSTRING(@NameCamelCase, CHARINDEX('-', @NameCamelCase) + 1, 1)) + SUBSTRING(@NameCamelCase, CHARINDEX('-', @NameCamelCase) + 2, 50)
                    ELSE
                        @NameCamelCase
                END 

            SELECT @NameCamelCase = REPLACE(@NameCamelCase, '^', '-')

            -- Multiple word names (do it twice to account for three word names)
            SELECT @NameCamelCase = 
                CASE 
                    WHEN @NameCamelCase LIKE '% %'
                        THEN SUBSTRING(@NameCamelCase, 1, CHARINDEX(' ', @NameCamelCase) - 1) + '?'  + UPPER(SUBSTRING(@NameCamelCase, CHARINDEX(' ', @NameCamelCase) + 1, 1)) + SUBSTRING(@NameCamelCase, CHARINDEX(' ', @NameCamelCase) + 2, 50)
                    ELSE
                        @NameCamelCase
                END 

            SELECT @NameCamelCase = 
                CASE 
                    WHEN @NameCamelCase LIKE '% %'
                        THEN SUBSTRING(@NameCamelCase, 1, CHARINDEX(' ', @NameCamelCase) - 1) + '?'  + UPPER(SUBSTRING(@NameCamelCase, CHARINDEX(' ', @NameCamelCase) + 1, 1)) + SUBSTRING(@NameCamelCase, CHARINDEX(' ', @NameCamelCase) + 2, 50)
                    ELSE
                        @NameCamelCase
                END 

            SELECT @NameCamelCase = REPLACE(@NameCamelCase, '?', ' ')

            -- Names in Parentheses         
            SELECT @NameCamelCase = 
                CASE 
                    WHEN @NameCamelCase LIKE '%(%'
                        THEN SUBSTRING(@NameCamelCase, 1, CHARINDEX('(', @NameCamelCase) - 1) + '('  + UPPER(SUBSTRING(@NameCamelCase, CHARINDEX('(', @NameCamelCase) + 1, 1)) + SUBSTRING(@NameCamelCase, CHARINDEX('(', @NameCamelCase) + 2, 50)
                    ELSE
                        @NameCamelCase
                END 


            SELECT @NameCamelCase = REPLACE(@NameCamelCase, ' sr', ' Sr')           
            SELECT @NameCamelCase = REPLACE(@NameCamelCase, ' jr', ' Jr')           
            SELECT @NameCamelCase = REPLACE(@NameCamelCase, ' ii', ' II')
            SELECT @NameCamelCase = REPLACE(@NameCamelCase, ' iii', ' III')
            SELECT @NameCamelCase = REPLACE(@NameCamelCase, ' DE ', ' de ')
            SELECT @NameCamelCase = REPLACE(@NameCamelCase, 'macdonald', 'MacDonald')

            if (@NameCamelCase LIKE '% iv')
                SELECT @NameCamelCase = REPLACE(@NameCamelCase, ' iv', ' IV')

            if ((@NameCamelCase = 'i') or (@NameCamelCase = 'ii') or (@NameCamelCase = 'iii') or (@NameCamelCase = 'iv'))
                SELECT @NameCamelCase = UPPER(@NameCamelCase)

            -- Return the result of the function
            RETURN ISNULL(@NameCamelCase, '')

        END

    RETURN ISNULL(@NameCamelCase, '')

END

IF  EXISTS (SELECT * FROM sys.objects WHERE object_id = OBJECT_ID(N'[dbo].[fs_PascalCase]') AND type in (N'FN', N'IF', N'TF', N'FS', N'FT'))
DROP FUNCTION [dbo].[fs_PascalCase]
GO
SET ANSI_NULLS ON
GO
SET QUOTED_IDENTIFIER ON
GO
CREATE FUNCTION [dbo].[fs_PascalCase]
(
    @Text nVarChar(MAX)
)
RETURNS nVarChar(MAX)
AS
BEGIN
        SET @Text = LOWER(@Text)--This step is optional.  Keep if you want the code below to control all casing. - 11/26/2013 - MCR.
    DECLARE @New nVarChar(MAX) = (CASE WHEN @Text IS NULL THEN NULL ELSE '' END)--Still return null when source is null. - 11/26/2013 - MCR.
    DECLARE @Len   Int = LEN(REPLACE(@Text, ' ', '_'))--If you want to count/keep trailing-spaces, you MUST use this!!! - 11/26/2013 - MCR.
    DECLARE @Index Int = 1--Sql-Server is 1-based, not 0-based.
    WHILE (@Index <= @Len)
        IF (SUBSTRING(@Text, @Index, 1) LIKE '[^a-z]' AND @Index + 1 <= @Len)--If not alpha and there are more character(s).
            SELECT @New = @New + UPPER(SUBSTRING(@Text, @Index, 2)), @Index = @Index + 2
        ELSE
            SELECT @New = @New +       SUBSTRING(@Text, @Index, 1) , @Index = @Index + 1

    --If @Text is null, then @Len will be Null, and everything will be null.
    --If @Text is '',   then (@Len - 1) will be -1, so ABS() it to use 1 instead, which will still return ''.
    RETURN ( UPPER(LEFT(@New, 1)) + RIGHT(@New, ABS(@Len - 1)) )
END
GO

SELECT dbo.fs_PascalCase(NULL)[Null],
       dbo.fs_PascalCase('')[EmptyString],
       dbo.fs_PascalCase('hello   how are-you TODAY    ')[LongString]

create function dbo.cap_words (@str varchar(max))
returns varchar(max)
as
begin
    declare @result varchar(max);
    select @result = string_agg( upper(left(value,1)) + substring(value,2,999),' ') from string_split(lower(@str),' ') 
    return @result;
end