Date:         Wed, 17 Feb 1999 12:53:43 -0500
Reply-To:     Lary Jones <ljones@BINGHAMTON.EDU>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@UGA.CC.UGA.EDU>
From:         Lary Jones <ljones@BINGHAMTON.EDU>
Subject:      Re: string variables
Comments: To: "Mack, Karin" <kim9@CDC.GOV>
In-Reply-To:  <3D839FBCE868D21181C400805FE6E24801DF3CDB@mcdc-atl-65.cdc.g ov>
Content-Type: text/plain; charset="us-ascii"

At 10:32 AM 2/17/1999 -0500, Mack, Karin wrote:
>Can someone tell me the language for taking a long string variable and
>breaking it up into smaller parts (e.g., SPSS6.0 into two variables, say,
>SPSS and 6.0)?
 
Breaking a string into shorter parts is done with the SUBSTR function.  If
one wants to divide a string into 4 character segments, then one could do
the following:
 
STRING Long (A12)  Part1 Part2 Part3 (A4).
COMPUTE Part1 = SUBSTR(long,1,4).
COMPUTE Part2 = SUBSTR(long,5,4).
COMPUTE Part3 = SUBSTR(long,9,4).
 
SUBSTR extracts 4 characters from long beginning at the 1st, 5th, and 9th
character.
 
The above could be done with a DO REPEAT or LOOP, as well.
 
If you wish to do the split based on the presence of a substring, something
like the following will work:
 
NUMERIC spot (F2).
STRING stuff (A10) part1 part2 (A8).
 
COMPUTE stuff = 'SPSS6.1 '.
 
COMPUTE spot = INDEX(stuff,'0123456789',1) .
 
COMPUTE part1 = SUBSTR(stuff,1,spot-1).
COMPUTE part2 = SUBSTR(stuff,spot).
 
PRINT / '  stuff=', stuff, '  part1=', part1,
        '  part2=', part2, '  spot=', spot .
 
            _____ output from above _____
 
  stuff=SPSS6.1      part1=SPSS       part2=6.1        spot= 5
 
 
This uses the INDEX function to find any 1 character part of '0123456789'
in the variable stuff.
 
Let me know if you wish more details!
 
-Lary Jones
_______________________________________________________
Lary Jones          %     Statistical Computing Analyst
Computing Services     %     ..........................
Binghamton University    %        LJones@Binghamton.EDU
Binghamton, NY 13902-6000   %            (607) 777-2879