Date:         Fri, 16 Jan 2004 16:18:33 -0500
Reply-To:     Howard Schreier <Howard_Schreier@ITA.DOC.GOV>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Howard Schreier <Howard_Schreier@ITA.DOC.GOV>
Subject:      Re: Automatic numeric/character type conversion: suggestions
              wanted for paper
Comments: To: Bruce Gilsen <bruce.gilsen@FRB.GOV>

Should be interesting. Right now I have just a couple of quibbles regarding
Case #1:

I would say that the number is too precise, rather than too big, since the
BESTw. format will always preserve magnitude at the sacrifice of
significant digits. Consider this:

   data _null_;
   do big = 1499, 1501;
      put big best4. +2 big best3.;
      end;
   run;

Result:

   1499  1E3
   1501  2E3

This also demonstrates that the consequence of not having space for all
significant digits is rounding rather than truncation.

On Thu, 8 Jan 2004 18:54:25 +0000, Bruce F. Gilsen <bruce.gilsen@FRB.GOV>
wrote:

>Hi, I'm working on a possible future conference paper, and one part is
about
>problems related to automatic type conversion between numeric and
character.
>You know, the ubiquitous notes like
>
>  NOTE: Numeric values have been converted to character
>        values at the places given by: (Line):(Column).
>        55:9
>  NOTE: Character values have been converted to numeric
>        values at the places given by: (Line):(Column).
>        52:8
>
>I've got 2 simple problem cases identified so far.
>
>1. NUMVAR1 is too big for the BEST12. format, so the value gets
>truncated to 1234567.8901.
>
>  data one;
>    numvar1 = 1234567.8901234;
>    call symput('macvar1',numvar1);
>  run;
>
>2. Before concatenation, NUM1 gets converted to
>"           1" (since BEST12. right justifies).
>If CHAR1 does not exist or exists with length 13 or greater,
>CHAR1 is "2           1" (11 spaces between 2 and 1).  If
>CHAR1 exists with length 12 or less, CHAR1 is "1".
>
>  data one;
>    num1=1;
>    char1 = "2" || num1 ;
>
>If anyone has any suggestions for other automatic type conversion
>problems, I would greatly appreciate hearing from you (on the list,
>off-line, whatever).  I'll post a summary to the list if I get
>off-line suggestions, and of course I'll cite you in the
>paper when it gets written.
>
>Thanks in advance,
>Bruce Gilsen
>speaking only for myself (and my fantasy baseball team that *almost* won
>for the 2nd straight year - I finished 2nd)