STUFF - 1

SQLServerCentral is supported by Red Gate Software Ltd.

Popular Topics

Home

Members

Calendar

Who's On

Home » Article Discussions » Article Discussions by Author » Discuss content posted by bitbucket » STUFF - 1

23 posts, Page 2 of 3

««1 2 3 »»

STUFF - 1

Rate Topic

Display Mode

Topic Options

Author

Message

Toreador

Posted Friday, November 23, 2012 2:10 AM

Ten Centuries

Group: General Forum Members
Last Login: Friday, May 17, 2013 9:52 AM
Points: 1,356, Visits: 4,761

There's an explanation of sorts here
http://msdn.microsoft.com/en-us/library/ms188043(SQL.105).aspx
(in the comments section at the end)

Post #1388081

Andre Ranieri

Posted Friday, November 23, 2012 8:46 AM

SSC Veteran

Group: General Forum Members
Last Login: Today @ 4:56 PM
Points: 277, Visits: 201

Nice question. I learned something new about T-SQL today. Thanks!

Andre Ranieri

Post #1388188

Revenant

Posted Friday, November 23, 2012 10:28 AM

Hall of Fame

Group: General Forum Members
Last Login: Today @ 2:42 PM
Points: 3,457, Visits: 2,532

kalyani.k478 (11/22/2012)

Post #1388211

john.arnott

Posted Friday, November 23, 2012 11:04 AM

UDP Broadcaster

Group: General Forum Members
Last Login: Wednesday, April 17, 2013 10:57 PM
Points: 1,491, Visits: 3,008

First, thank you for prodding me to read about STUFF, a function I'd seen mentioned but which I haven't used.

Your explanation of answer starts with

I could not find an explanation of the "lengthing" of the declared item....

This was the point I considered after reading the BOL entry and before choosing my answer. The given QOD code does not assign the modified value to either of the variables. The function may return data with a type inherited from the input, but the length of the function's output is not constrained by the lenghth of either input string. Consider what you may naturally expect if you were to add these lines to the end of the given script:

DECLARE @string2 VARCHAR(50)

set @string2 = STUFF(@string1, 11,0,@Stuffing)
select @string2
	  ,len(@string2)

A truncated value of "Microsoft ******" or "Microsoft ********************" would hardly be acceptable.

Post #1388218

KWymore

Posted Friday, November 23, 2012 12:54 PM

SSCrazy

Group: General Forum Members
Last Login: Today @ 1:57 PM
Points: 2,575, Visits: 1,533

Nice question about a function that I do not use enough. Thanks for the question!

Post #1388242

@Cassie

Posted Saturday, November 24, 2012 2:39 AM

Mr or Mrs. 500

Group: General Forum Members
Last Login: Wednesday, April 24, 2013 2:28 PM
Points: 515, Visits: 119

Nice Stuff.

T-SQL is Great

Post #1388292

L' Eomot Inversé

Posted Saturday, November 24, 2012 2:47 AM

SSCertifiable

Group: General Forum Members
Last Login: Today @ 3:07 PM
Points: 7,096, Visits: 7,156

Nice question - and a reminder to me to use STUFF more often.

I don't understand why the explanation worries about the result length. . I agree with John Arnott's comments above about this. String functions either do or don't have a length parameter, and the ones that don't have a resulting length that can be any valid length (except MAX, unless a parameter of the function is declared with length MAX). Why should anyone expect STUFF to behave differently from + in this respect?

Tom
Que conclure à la fin de tous mes longs propos? C'est que les préjugés sont la raison des sots. (Voltaire, 1756)

Post #1388293