home *** CD-ROM | disk | FTP | other *** search
- Comments: Gated by NETNEWS@AUVM.AMERICAN.EDU
- Path: sparky!uunet!paladin.american.edu!auvm!UDELVM.BITNET!ACS03043
- Message-ID: <SAS-L%92091011201144@VTVM2.CC.VT.EDU>
- Newsgroups: bit.listserv.sas-l
- Date: Thu, 10 Sep 1992 10:36:44 EDT
- Reply-To: "John R. Gerlach" <ACS03043@UDELVM.BITNET>
- Sender: "SAS(r) Discussion" <SAS-L@UGA.BITNET>
- From: "John R. Gerlach" <ACS03043@UDELVM.BITNET>
- Subject: The RETAIN Statement
- Lines: 35
-
- Greetings,
-
- There seems to be a problem with the RETAIN statement. Consider
- the code below which produces a single observation data set containing
- n-variables initialized to zero using the RETAIN statement. At the
- 96th, 97th, and 98th iteration, SAS generates an "Internal compiler
- stack overflow" error message. A patch to this problem is to use
- two RETAIN statements. Any comments?
-
- The SAS (6.06) code was executed on an IBM 3090 running VM/CMS.
-
-
- --- John Gerlach
-
-
-
- %macro nth_run(n);
- data ds;
- retain var1-var&n. 0;
- run;
- proc print;
- title "The &n.th iteration";
- run;
- proc datasets library=work;
- delete ds;
- run;
- %mend nth_run;
-
- %macro doit;
- %do i = 90 %TO 100;
- %nth_run(&i.)
- %end;
- %mend doit;
-
- %doit
-