Date:         Fri, 28 Apr 2000 14:54:54 +0100
Reply-To:     John Whittington <John.W@MEDISCIENCE.CO.UK>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         John Whittington <John.W@MEDISCIENCE.CO.UK>
Subject:      Re: length of variable assigned to substr function.
Comments: To: "Workman, Robert" <rworkman@CARBOMEDICS.COM>
In-Reply-To:  <200004281325.OAA02906@vicar.netnames.net>
Content-Type: text/plain; charset="us-ascii"

At 08:19 28/04/00 -0500, Workman, Robert wrote:

>Unfortunately it doesn't seem to be that simple when using functions to set
>the initial value of the variables.  "The rule" holds true when using
>assignment statements to assign a literal string to the variable.  However,
>different functions seem to have different rules for determining the length.

Robert, yes, you're dead right - and Ian W has explained the 'thinking'
that the SAS compiler uses when the first encounter with a variable is an
assignment with the value returned by a function like SUBSTR() - I had
never noticed that the compiler was so 'clever'!

However, I think the bottom line still remains the same - that, unless one
is sure that one knows what one is doing, and equally sure of all possible
ata situations, one is advised to explicitly define the length of a
character variable that has not come from another data set.

Thanks for pointing out my error!

Kind Regards,


John

----------------------------------------------------------------
Dr John Whittington,       Voice:    +44 (0) 1296 730225
Mediscience Services       Fax:      +44 (0) 1296 738893
Twyford Manor, Twyford,    E-mail:   John.W@mediscience.co.uk
Buckingham  MK18 4EL, UK             mediscience@compuserve.com
----------------------------------------------------------------