String Data manipulation query

[dennis_oz]

Hi all,

Appreciate some help on this .

Thanks in advance.

data a;
input word1 $ & 200.;
datalines;
101 Capital Region R 000s
101 Capital Region R Uwgt
101 Capital Region R Index
102 Central Coast R Unwgt
102 Central Coast R 000s
102 Central Coast R Vert percent
;
run;


want 
101 Capital Region
101 Capital Region
101 Capital Region 
102 Central Coast
102 Central Coast
102 Central Coast

[PaigeMiller]

So what are the rules for doing this truncation of the text? Is it always truncated after the third word, or before the "R", or do you always want the first 18 columns?

[dennis_oz]

Hi, Yes it is before the "R".

Thank you .

[ballardw]

One way:

data want;
   set a;
   word1 = substr(word1,1,findw(word1,'R')-1);
run;

Not the only by any means.

[PaigeMiller]

data want;
    set a;
    where=find(word1,' R ');
    word1=substr(word1,1,where-1);
    drop where;
run;

[dennis_oz]

sorry slight update to requirement , suppose the data is as below , i.e  a combination of R and M

data a;
input word1 $ & 200.;
datalines;

102 Central Coast R 000s
103 Capital Region M 000s

;
run;

want 
-----
102 Central Coast
103 Capital Region

[andreas_lds]

Which letter will be added to the separator list with the next update of the requirements?

[PaigeMiller · Accepted solution]

where=max(find(word1,' R '),find(word1,' M '));