topic Using PROC inside code transform in Data Studio in SAS Procedures

Using PROC inside code transform in Data Studio

alisio_meneses — Thu, 03 Nov 2022 20:12:11 GMT

Hello there.

Is it possible to use this PROC inside a code transform in Data Studio (Something like the following code)?

proc tsmodel data={{_dp_inputCaslib}}.{{_dp_inputTable}} seasonality=7 
		outarray={{_dp_outputCaslib}}.{{_dp_outputTable}}(replace=YES);
	by var1 var2;
	id data interval=Day setmissing=0 FORMAT=_DATA_;
	var var3 / accumulate=total;
run;

The documentation clearly states that CASL and DATA Step code languages are available to use, but as a noob I still struggle to identify which is which.

Thank You!

Re: Using PROC inside code transform in Data Studio

SASKiwi — Thu, 03 Nov 2022 21:04:46 GMT

If the code runs fine in a "normal" SAS Studio session then I'd say yes. In any case no harm in trying.

Re: Using PROC inside code transform in Data Studio

alisio_meneses — Fri, 04 Nov 2022 10:33:22 GMT

Hi SASKiwi. Thanks for your reply. Just tested the code in a Data Studio code transformation. The plan executed without complaints but the session table did not show the proc result and kept the same table from the previous transformation. Any tips on what am I doing wrong?

Re: Using PROC inside code transform in Data Studio

LinusH — Fri, 04 Nov 2022 10:40:27 GMT

Please attach the log, and maybe a print screen about the session table.

Re: Using PROC inside code transform in Data Studio

alisio_meneses — Fri, 18 Nov 2022 10:57:13 GMT

Sure thing. Here you go.

The screenshot below was taken after "successfully" running the code transform. By "successfully" I mean the plan ran without errors (log attached) but the session table didn't change. The expected result is shown on the next screenshot.

The screenshot below was taken in SAS StudioV and shows the expected result. This is the table I was expecting to see when running the code transform in the Data Studio plan.

This is the code inside the transform:

/* BEGIN data step with the output table data */
data {{_dp_outputTable}} (caslib={{_dp_outputCaslib}} promote="no");
/* Set the input set */
set {{_dp_inputTable}} (caslib={{_dp_inputCaslib}} );
/* END data step run */
run;

proc means nway order=freq missing;proc means nway order=freq missing;
   class country;
   freq num_sold;
   output out={{_dp_outputCaslib}}.{{_dp_outputTable}}(drop=_type_ index=(country)) / levels;
run;

Dataset used for testing: train.csv