Or just add the forma to the PUT statement: data _null_; thisdate=today(); rptdate = (today()+1); if (thisdate<rptdate) then put "WARNING: future RPTDATE. " rptdate= date9.; run; ... View more

If you're already using a SAS supplied format like percent. below sample code to demonstrate how you can extend it for a missing case. proc format; value missToUnknown . = 'Unknown' other=[percent.] ; run; data demo; do var=-0.5,.,0,0.5; output; end; run; proc print data=demo; format var missToUnknown.; run;   ... View more

Alternatively, start with a "long" dataset or create one from your existing data: proc transpose data=TempMax out=long; by id; run; Then you can use PROC SUMMARY (or PROC SQL) to obtain the desired output quite naturally: proc summary data=long; by id; var col1; id _name_; output out=want(drop=_:) max=maxdate maxid=varname; run; ... View more

You can use SAS array and examine the values of each column to match with 1, and count them. data have; input A B C D E; datalines; 1 2 1 1 2 2 2 1 2 2 2 3 3 2 2 1 2 3 1 1 ; run; data want; set have; array x[*] A B C D E; count = 0; do i = 1 to dim(x); if x[i] = 1 then count + 1; end; drop i; run; proc print data = want; run; ... View more

You can also redirect parts of the log with PROC PRINTTO.   Example: filename otherlog 'C:\Temp\routine_macro.log'; ... code of main program ... proc printto log=otherlog; run; %routine_macro_call(1) proc printto; run; ... more code of main program ... proc printto log=otherlog; run; %routine_macro_call(2) proc printto; run; ... more code of main program ... This would write the parts of the log created by your "routine program" (I've used a macro call as an example) to the external file C:\Temp\routine_macro.log and the more interesting "main" parts of the log to the default destination, so you can read both parts separately.   To suppress the code and notes from the PROC PRINTTO steps, you can add the NONOTES and NOSOURCE options suggested by Reeza and SAS_Kiwi: options nonotes nosource; proc printto log=otherlog; run; %routine_macro_call(...) proc printto; run; options notes source; Then only the first line of the above block of code will appear in the main log. EDIT: This would, of course, suppress the notes and source code of the "routine program" as well. So, depending on how much you want to see from the "otherlog", you may want to wrap the two OPTIONS statements only around the PROC PRINTTO steps.     ... View more

Hi:   I don't recommend fiddling with the tables in Word, after the RTF file is created. I would suggest trying to find out the final table size and then use that size like this: ods rtf file='c:\temp\samesize.rtf' startpage=no; proc report data=sashelp.class(obs=5) Style(Report)={frame=hsides rules=cols width=6.5in}; column name age height; run; proc report data=sashelp.class(obs=5) Style(Report)={frame=hsides rules=cols width=6.5in}; column name sex age height weight; run; proc report data=sashelp.shoes(obs=5) Style(Report)={frame=hsides rules=cols width=6.5in}; column region subsidiary product sales inventory returns; run; ods rtf close; Assuming you have margins of just 1 inch on the right and left side, then the available space on the page would be 6.5 inches (but notice how I just specify width= for each PROC REPORT step, which "spreads" the table as needed without overcontrolling the width of the individual cells. Of course, you can adjust this but you'll want all of your tables to be the same size, which is something that might be known ahead of time in the documentation or the standards document. I just used startpage=no, so the 3 tables would line up on the same page and you could see that each table is lined up with the one beneath it in width.   Hope this helps, Cynthia ... View more

Maybe not the most elegant but should get you started. Note the ADDED variable ROW. data Temp1; input Col1 $4. Var1 Var2 Var3 Var4; row + 1; datalines; Var1 . 45 200 223 11 Var2 0.43 . 43 99 Var3 0.98 0.89 . 39 Var4 0.99 0.77 0.99 . ; run; proc report data=Temp1 style(Header)=[backgroundcolor = lightgrey]; columns row Col1 Var1 Var2 Var3 Var4; define row /noprint order; compute var1; if row=2 then call define(3, "style", "style=[backgroundcolor=yellow]"); if row in (3,4) then call define(3, "style", "style=[backgroundcolor=lightgreen]"); endcomp; run; ROW has the NOPRINT property so it does not appear in the table but values can be checked in a compute block. Note that call define is setting the number of the column based on the Columns statement definition. So the 3rd column is Var1 after Row and Col1. And exercise for the interested student to do the columns for Var2, var3 and var4 as needed. ... View more

Speaking of looping over elements in a correlation matrix, I recently wrote about how to convert the correlation matrix to a "long format" that shows the pairwise correlations in a list format: "Display correlations in a list format"   The DATA step loop for that purpose is similar to @Tom's loop here since it processes the same configuration of data.   ... View more

thats what i expected. But I couldn't find any confirmation of it anywhere. Thanks. ... View more

Forgot to mention.  All users are using Enterprise Guide.   ... View more

There are some differences between the SQL unions and the data step SET statements when the structures are different.   The SQL code will set the length of character variables to the maximum length it has on any of the input datasets. The SET statement will set the length of character variables to the length it has on the first dataset where it appears.   There is a limit on the number of datasets you can combine in a single SQL statement.  I don't think there is a similar limit on the number of datasets you can reference in a SET statement. ... View more

Once it's in CSV you can also use this macro to clean up the carriage returns:  https://core.sasjs.io/mp__cleancsv_8sas.html Another option is a tool my team have produced, which is free for up to 5 users, and lets you import any version of excel (with data on any sheet, in any row, with columns in any order).  It can also be used to fix the carriage return issue:  https://datacontroller.io ... View more

Hi,   Sorry for multiple replies.. I tried this code and it worked. A thin dash line is visible but it won't affect the printing.   proc report data=Temp1 split = '@'   Style(Report)=[rules=None Frame=HSides cellspacing=0 ]   style(Header)=[background=lightgrey borderleftwidth=0.1pt borderleftcolor=black borderrightwidth=0.1pt borderrightcolor=black ]   style(Column)=[just=c bordertopcolor=black bordertopwidth=0.1pt borderleftcolor=white borderrightcolor=white borderbottomwidth=0.1pt borderbottomcolor=black]; Columns ("^S={borderleftwidth=0.1pt borderleftcolor=black}" (col1)) ("^S={borderleftwidth=0.1pt borderleftcolor=black}" (col2)) ("^S={borderbottomwidth=0.1pt borderbottomcolor=black borderbottomstyle=solid}Span1" (Col3 Col4)) ("^S={borderbottomwidth=0.1pt borderbottomcolor=black borderbottomstyle=solid}Span2" (Col5 Col6)); define Col1 /display; define Col2 /analysis; define Col3 /analysis; define Col4 /analysis; define Col5 /analysis; define Col6 /analysis; run; ... View more