Delete parts of variables, but how??

[Wouter]

Hello,

I've got a dataset in which I want to remain everything after a particular word, like "debtor". Like:

ID | Variable_1
1 xxx debtor profile 7
2 x debtor profile 3 to 1
3 xx debtor profiles 133 to 155
4 xxxx debtor profile 3

So the final result would be:

ID | Variable_1
1 profile 7
2 profile 3 to 1
3 profiles 133 to 155
4 profile 3

In other words, I need a statement with which I could delete everything to the left of a particular word. Anyone any idea? Thanks in advance!

[Olivier]

Hi.
What you can try is check where "debtor" is with the INDEW function, then pick the string beginning after that place with a SUBSTR function.
[pre]
DATA test ;
INFILE CARDS MISSOVER DLM="," ;
INPUT Variable_1 :$50. ;
posDebtor = INDEXW(variable_1, "debtor") ;
afterDebtor = LEFT(SUBSTR(variable_1, posDebtor+LENGTH("debtor"))) ;
CARDS ;
xxx debtor profile 7
x debtor profile 3 to 1
xx x x x x xx debtor profiles 133 to 155
xxxx debtor profile 3
;
RUN ;
[/pre]
Regards.
Olivier

[Wouter]

Thanks, works perfect!! Right now, I don't fully understand the complete code (the statements INDEXW, SUBSTR), but the idea I fully understand. Most of the time, I've got the idea, but lacking the ability to 'convert' it into SAS codes... Right now, I'm going to check the statements. Thanks!!

[sbb]

Consider using the SCAN function in a DATA step to extract / parse your SAS variable. Your delimiter argument can be a something other than a single character, which will work for your circumstances.

Related topic links are provided for your reference - found on the SAS support website. http://support.sas.com/

Scott Barry
SBBWorks, Inc.

SUGI 2007 paper:

http://www2.sas.com/proceedings/forum2007/035-2007.pdf

SAS DOC on SCAN function:

http://support.sas.com/documentation/cdl/en/lrdict/59540/HTML/default/a000214639.htm

SAS Support SAMPLES:

ftp://ftp.sas.com/techsup/download/sample/datastep/scanfunction.html