@RW9: 3)~      - could mean anything, I use this as split in report for instance. Actually, it is also an alternative form of the NOT operator, e.g. 408 %put %eval(1~=2); 1 ... View more

This code will read a directory and find the creation times for .xlsx files: filename indir '<directory>'; data test; dirid=dopen('indir'); do filenum=1 to dnum(dirid); name=dread(dirid,filenum); if upcase(scan(name,-1,'.')) ne 'XLSX' then continue; fileid=mopen(dirid,name); if fileid=0 then do; put _all_; continue; end; Created=finfo(fileid,'Create Time'); output; rc=fclose(fileid); end; keep name created; run; The created variable will be a datetime, but formatted according to your national language settings in Windows, you then have to find the correct informat for reading it into a numeric datetime. After that, it should be as simple as to sort by the date and take the last record. ... View more

No, it won't. The program runs without any errors, but it is not correct. ... View more

The problem is that your NAME variable is not read correctly. It consists of first name and last name, but the way you read it only the first name gets into the NAME variable. You can do it like this, provided that you do not suddenly get a person with a middle name: data club; infile datalines; length idnumber 8 firstname lastname $10 team $6 startwi endwi 8; input idnumber--endwi; datalines; 1023 DAVID shaw RED 189 155 1049 AMELIA verna YELLOW 145 125 1219 ALAN ranju red 210 192 1246 RAVI shankar YELLOW 177 177 1078 ASHELY valley RED 127 118 1221 JIM rose YELLOW 220 220 ; run; ... View more

If I understand you correctly, you are writing a function for calculating distances on a sphere (the Earth). SAS already has a function fot that, it is called GEODIST. ... View more

You may get away with using the DBSASTYPE option: data new; set dblib.mydata(dbsastype=(orig_var='CHAR(8)')); run; ... View more

As @ChrisNZ wrote, one option is to read the variable as character and then clean it.   If you are absolutely sure that you do not have any ">" characters anywhere else, the fast and dirty solution can be to add it as a delimiter: DATA filename; infile datalines delimiter=' >'; INPUT age class $; Datalines; 13 B 18 C >36 A 21 B ; Run; Another quick fix is to modify the infile buffer: DATA filename; input @; if _infile_=:'>' then _infile_=substr(_infile_,2); INPUT age class $; Datalines; 13 B 18 C >36 A 21 B ; Run; ... View more

Using the index to get at the duplicates is a very good idea, you may try something like this:   data big; set dupes(keep=<key variables>); first=1; do until(0); modify big key=<your index name>; if _iorc_ then do; _error_=0; /* when _iorc_ is set, an error is provoked */ leave; end; if not first then remove; else first=0; /* the next obs is not the first */ end; run;   ... View more

Why not just insert the number of observations previously selected, which is returned in the SQLOBS macro variable: create table DX.SAS_Table as select * from connection to oracle ( select * from my table ); execute by Oracle( insert into TEST_SAS_INSERT values(&sqlobs) ); If you want to insert data from SAS tables into Oracle tables, the easiest is normally to allocate the Oracle connection as a SAS library: libname Oralib Oracle <connection options here>; Proc sql; INSERT into Oralib.TEST_SAS_INSERT select count(*) from SAS_Table ; quit; ... View more

I would use PROC FCMP to write a function, e.g.: proc fcmp outlib=work.funcs.str; function palindrome(str $,options $); str=left(str); options=upcase(options); if index(options,'I') then str=upcase(str); l=length(str); if l=1 then return(1); half=floor(l/2); if substr(str,1,half)=reverse(substr(str,l-half+1,half)) then return(1); return(0); endsub; run; options cmplib=work.funcs; data test; input str $; p=palindrome(str,''); pi=palindrome(str,'i'); cards; a aga aggA aggb ;run; ... View more

You can do it with a function macro, like this: %macro tarr(vector); %local date i r plus; %do i=1 %to %sysfunc(countw(&vector,+)); %let date=%sysfunc(inputn(%scan(&vector,&i,+)01,yymmdd6.)); %let r=&r.&plus.%str(%')%sysfunc(intnx(month,&date,0,e),date9.)%str(%')d; %let plus=+; %end; %unquote(&r) %mend; %let tarr=%tarr(&vector); ... View more

This is one of the occasions where PRX is really good: data want; Vector='0,0,5,11,9,10'; pos=prxmatch('/\b[1-9]/',Vector); run; A short explanation of the search string: \b means a word boundary (could be your comma, or the beginning of the string), [1-9] means one of the digits 1 to 9. If there is no match, the function returns 0. ... View more