Alteryx Designer Desktop Discussions

Find answers, ask questions, and share expertise about Alteryx Designer Desktop and Intelligence Suite.
SOLVED

Converting SAS to Alteryx

jokeffer
7 - Meteor

Hi All, I'm very new to Alteryx and looking for some suggestions on how to convert the below SAS to Alteryx. I'm thinking it would be a combination of a Formula along with Regex but since just learning Alteryx not sure

 

This is my data and you'll notice that sometime there's an extra semicolon and sometimes there's a space missing or semicolon missing.

 

NM

VB

VB00026

VB = STD ; HIRE ; ; YEARS ; 1 ; 902NZ ; 2 ; 902P0 ; 3 ;902P1 ; 4 ; 902P2 ; 5 ; 902P3 ; AFTER ; 9029R

VB05067

VB = STD ; EMPL ; 20091101 ; MONTHS ; 12 ; 90ZHCAFTER ; 914DV

VB05089

VB = STD ; EMPL ; 20040101 ; MONTHS ; 6 ; 90ZP2 ;12 ; 90ZP5 ; 24 ; 90ZP6 ; AFTER ; 914HH

VB10090

VB = STD ; PCAR ; ; MONTHS ; 6 ; 91ZHH ; 12 ; 91ZHJ ;AFTER ; 91ZHK

 

Here's my SAS datastep I'm trying to convert to Alteryx:

 

data vbs1;

set vbs;

if index(vb, 'MONTHS ; 6')> 0 then first_6 = substr(vb, index(vb, 'MONTHS ; 6')+13, 5);

if index(vb, ' ; 12 ;') >0 then month_12 = substr(vb, index(vb, ' ; 12 ;')+8, 5);

if index(vb, '; YEARS ; 1 ;')>0 then year_1 = substr(vb, index(vb, '; YEARS ; 1 ;')+14, 5);

if index(vb, '; 2 ;')>0 then year_2 = strip(trim(substr(vb, index(vb, '; 2 ;')+5, 6)));

if index(vb, '; 3 ;')>0 then year_3 = strip(trim(substr(vb, index(vb, '; 3 ;')+5, 6)));

if index(vb, '; 4 ;')>0 then year_4 = strip(trim(substr(vb, index(vb, '; 4 ;')+5, 6)));

if index(vb, '; 5 ;')>0 then year_5 = strip(trim(substr(vb, index(vb, '; 5 ;')+5, 6)));

if index(vb, 'AFTER ;')>0 then after = substr(vb, index(vb, 'AFTER ;')+8, 5);

run;

 

And this is how the data looks after SAS datastep and how I need it to look in Alteryx:

 

NM

VB

first_6

month_12

year_1

year_2

year_3

year_4

year_5

after

VB00026

VB = STD ; HIRE ; ; YEARS ; 1 ; 902NZ ; 2 ; 902P0 ; 3 ;902P1 ; 4 ; 902P2 ; 5 ; 902P3 ; AFTER ; 9029R

 

 

902NZ

902P0

902P1

902P2

902P3

9029R

VB05067

VB = STD ; EMPL ; 20091101 ; MONTHS ; 12 ; 90ZHCAFTER ; 914DV

 

90ZHC

 

 

 

 

 

914DV

VB05089

VB = STD ; EMPL ; 20040101 ; MONTHS ; 6 ; 90ZP2 ;12 ; 90ZP5 ; 24 ; 90ZP6 ; AFTER ; 914HH

90ZP2

 

 

 

 

 

 

914HH

VB10090

VB = STD ; PCAR ; ; MONTHS ; 6 ; 91ZHH ; 12 ; 91ZHJ ;AFTER ; 91ZHK

91ZHH

91ZHJ

 

 

 

 

 

91ZHK

 

Thanks for any and all help!

 

4 REPLIES 4
jamielaird
14 - Magnetar

Hi @jokeffer 

 

Personally I would recommend first converting your data into a tabular format, and secondly using Formula tools to create the new columns. It will be much easier to write out the logic once you have distinct fields for each of the different parts of your data.

danilang
19 - Altair
19 - Altair

hi @jokeffer  

 

Since you've got varying number of columns in your VB, using @jamielaird's solution may be tricky.  A more straightforward approach would be to build a Formula tool with translation of the SAS to Alteryx

 

formula.png 

 

The translation is very direct. IndexOf()<>0 maps to Contains and substring(field,Indexof() maps to substring([field],findstring()).  You just have to watch the indexes.  Sometimes, but not always, they're off by one

 

I've created the 1st 3 columns for you resulting in 

 

 

Result.png

 

If you've only got these 8 fields to build out, using this approach will work well.  If you have hundreds of datasteps to convert, you'll need to come up with a different strategy. 

 

Dan

jokeffer
7 - Meteor

Yes, this will work. It has flushed out a couple additional string types to deal with but 10 isn't too bad. Thank you for your suggestions and the code to get it started! 

jokeffer
7 - Meteor

Thank you for your suggestion, however my data has varying lengths so opted for the SAS to Alteryx conversion. Really appreciate the help though. 

Labels