Update dataset onto itself.... WARNING

[NKormanik]

Please look over the following code, particularly toward the bottom where there is a NOTE and WARNING message:

1    proc summary data=sas_1.combined;
2    var
3    i_50401_Z
4    i_50402_Z
5    i_50403_Z
6    i_50404_Z
7    i_50405_Z
8    i_50408_Z
9    i_50501_Z
10   i_50502_Z
11   i_50503_Z
12   i_50504_Z
13   i_50505_Z
14   i_50508_Z
15   ;
16   output out=sas_1.combined_temp (drop= _:) p80= /autoname;
17   run;

NOTE: Multiple concurrent threads will be used to summarize data.
NOTE: There were 5000 observations read from the data set SAS_1.COMBINED.
NOTE: The data set SAS_1.COMBINED_TEMP has 1 observations and 12 variables.
NOTE: PROCEDURE SUMMARY used (Total process time):
      real time           0.07 seconds
      user cpu time       0.01 seconds
      system cpu time     0.03 seconds
      memory              13816.03k
      OS Memory           30012.00k
      Timestamp           02/11/2022 07:13:37 PM
      Step Count                        1  Switch Count  0


18
19   Proc sql;
20   create table sas_1.combined as
21   select a.*,
22   (a.i_50401_Z >= b.i_50401_Z_P80) as i_50401_Z_top20pct ,
23   (a.i_50402_Z >= b.i_50402_Z_P80) as i_50402_Z_top20pct ,
24   (a.i_50403_Z >= b.i_50403_Z_P80) as i_50403_Z_top20pct ,
25   (a.i_50404_Z >= b.i_50404_Z_P80) as i_50404_Z_top20pct ,
26   (a.i_50405_Z >= b.i_50405_Z_P80) as i_50405_Z_top20pct ,
27   (a.i_50408_Z >= b.i_50408_Z_P80) as i_50408_Z_top20pct ,
28   (a.i_50501_Z >= b.i_50501_Z_P80) as i_50501_Z_top20pct ,
29   (a.i_50502_Z >= b.i_50502_Z_P80) as i_50502_Z_top20pct ,
30   (a.i_50503_Z >= b.i_50503_Z_P80) as i_50503_Z_top20pct ,
31   (a.i_50504_Z >= b.i_50504_Z_P80) as i_50504_Z_top20pct ,
32   (a.i_50505_Z >= b.i_50505_Z_P80) as i_50505_Z_top20pct ,
33   (a.i_50508_Z >= b.i_50508_Z_P80) as i_50508_Z_top20pct
34
35   from sas_1.combined as a,  sas_1.combined_temp as b;
NOTE: The execution of this query involves performing one or more Cartesian product joins that can not be optimized.
WARNING: This CREATE TABLE statement recursively references the target table. A consequence of this is a possible data integrity problem.
NOTE: Table SAS_1.COMBINED created, with 5000 rows and 447 columns.

36
37   quit;

I'm creating 'flag' variables, to designate the top 20% of the values for various variables.  (Huge thanks to @ballardw for the code.)

I'd prefer to keep the name of the dataset the same.  But that means overwriting the initial dataset.  SAS gives a WARNING.  However, the dataset seems to get updated just fine nonetheless.

What do you think?  Just overlook the WARNING?  Or is there a better way of going about this?

Thanks!

Nicholas Kormanik

[Ksharp]

I am afraid you need to create a new table .

[Kurt_Bremser · Accepted solution]

Do not overwrite a dataset in a single step, a slight miscue will otherwise destroy it.

For simplicity of coding, I would use a DATA step and arrays.

[NKormanik]

Thanks guys....

/*  Only works if the desired changed-to filename does not exist.... */
/*
proc datasets library=sas_1;
change combined_2=combined;
run;
*/


data sas_1.combined;
set sas_1.combined_2;
run;

proc datasets lib=sas_1;
delete combined_2;
delete combined_temp;
run;

quit;