Prev: How to merge with duplicate observation on both sides
Next: Unable to sort dataset with 6.7 Million records. ERROR: Utility file write failed. Probable disk full condition.
From: Kenneth M. Lin on 21 Mar 2010 22:50
By that I mean just scan the program to check for syntax errors and not
actually process the data but assume that all datasets exist. My client
won't give me access to Oracle where the data reside and I am writing the
program rather blindly.

From: Amar Mundankar on 22 Mar 2010 02:02
On Mar 22, 7:50 am, "Kenneth M. Lin" <kenneth_m_...(a)sbcglobal.net>
wrote:
> By that I mean just scan the program to check for syntax errors and not
> actually process the data but assume that all datasets exist.  My client
> won't give me access to Oracle where the data reside and I am writing the
> program rather blindly.

Hi Kenneth,
For Proc sql you can use NOEXEC option

proc sql noexec;
select * from table;
/*--You can write multiple querries here.--*/
quit;

This will only check the syntax and will not execute the query.

For Data step, you can set SAS System option OBS = 0 and run the
program.
And if you dont have the actual dataset create some sample dataset
without any record in that.
e.g. Create a dataset called TEST with o observation and some varibles
in it.

data TEST;
attrib a b c;
run;

Then set OBS = 0 like.
OPTIONS OBS = 0;
/*--Write your code below--*/
data TEST2;
set TEST;
/*---Write SAS Statements here--*/
run;

Hope this willl help you.

Regards,
Amar Mundankar.
 | 
Pages: 1
Prev: How to merge with duplicate observation on both sides
Next: Unable to sort dataset with 6.7 Million records. ERROR: Utility file write failed. Probable disk full condition.