• Here's one similar to Luis, but handles detecting the start of a word differently and works on nvarchar(4000) input.

    create function [dbo].[propercase](@text nvarchar(4000))

    returns table with schemabinding

    as

    return (

    with seed1 (a)

    as

    (

    select 1 union all select 1 union all select 1 union all select 1 union all select 1 union all select 1 union all select 1 union all select 1 union all

    select 1 union all select 1 union all select 1 union all select 1 union all select 1 union all select 1 union all select 1 union all select 1

    ),

    numbers (n) as

    (

    select top (datalength(@text)) row_number() over (order by (select null))

    from seed1 s1, seed1 s2, seed1 s3

    )

    select a.b.value('(./text())[1]', 'nvarchar(4000)') as [text]

    from (

    select

    case

    when n = 1 then upper(substring(@text, n, 1))

    when substring(@text, n - 1, 2) like N'[^a-z][a-z]' collate Latin1_General_CI_AI then upper(substring(@text, n, 1))

    else lower(substring(@text, n, 1))

    end

    from numbers

    for xml path (''), type

    ) a (b)

    )

    On the small test strings it performs about 50% slower than Luis original one, but that is just because of the NVARCHAR(4000) compatibility.

    Below is a varchar(255) version that is comparable to Luis, but with the extra word start checks.

    create function [dbo].[propercase](@text varchar(255))

    returns table with schemabinding

    as

    return (

    with seed1 (a)

    as

    (

    select 1 union all select 1 union all select 1 union all select 1 union all select 1 union all select 1 union all select 1

    ),

    numbers (n) as

    (

    select top (datalength(@text)) row_number() over (order by (select null))

    from seed1 s1, seed1 s2, seed1 s3

    )

    select a.b.value('(./text())[1]', 'varchar(255)') as [text]

    from (

    select

    case

    when n = 1 then upper(substring(@text, n, 1))

    when substring(@text, n - 1, 2) like '[^a-z][a-z]' collate Latin1_General_CI_AI then upper(substring(@text, n, 1))

    else lower(substring(@text, n, 1))

    end

    from numbers

    for xml path (''), type

    ) a (b)

    )

    EDIT: modified to cope with Accénted characters...Damn you Luis! I wanted to sleep:-D

    MM



    select geometry::STGeomFromWKB(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

  • Forum Etiquette: How to post Reporting Services problems
  • [/url]
  • Forum Etiquette: How to post data/code on a forum to get the best help - by Jeff Moden
  • [/url]
  • How to Post Performance Problems - by Gail Shaw
  • [/url]