CODEWITHSUNDEEP
sas
Zorran
Zorran

Reputation: 13

How to differentiate between one space and multiple spaces in SAS

I would like to parse a sas variable into several variables based on the the space (or special characters) between 2 words. I would like the code to place the word in a new column ONLY when there is > 2 spaces (or special chars) between the words. for example below I have used ^ instead of space here.

    if TEXT="                   
            LEVEL^OF^RENAL^IMPAIRMENT^^^^^^^SEVERE,^n^(%)^^^^^^^52^(^1.9)^^^^^^^^^ 133^(^1.5)"

   then col1= "LEVEL^OF^RENAL^IMPAIRMENT"
  col2= "SEVERE,^n^(%)"
  col3="52^(^1.9)"
  col4="133^(^1.5)"

it would be great if the code can count the number of columns for each record upto a max of say 15 columns.........Regards, Z

Upvotes: 1

Views: 1385

Answers (1)

Jeff
Jeff

Reputation: 1807

You can use a regex to convert your data in delimted file that the scan() function can understand. For example this statement,

text2=prxchange('s/(\^| ){2,}/~/',-1,TEXT);

results in text2 of

LEVEL^OF^RENAL^IMPAIRMENT~SEVERE,^n^(%)~52^(^1.9)~133^(^1.5)

which is a version of your data that uses ~ as a delimeter. Now you can use scan(), like this:

col1 = scan(text2,1,'~');
col2 = scan(text2,2,'~');
col3 = scan(text2,3,'~');
col4 = scan(text2,4,'~');

You might have to modify the regex to capture all the spaces and special characters that can make up a delimiter. In my example, the (\^| ) means "a string that is either a single ^ or a single space". Also, if you're data contains ~ you'll probably want to choose some other characther.

Upvotes: 2

Related Questions