Thanks.. ... View more

This is a community, not a free support site. If you need immediate help, hire a consultant. Also, it's easier to help if you provide logs that are readable. Have them in an attachment if you got problem pasting them into the text input field in the browser. So how are you executing this code? From a Linux shell command prompt, Enterprise Guide connection, or? Since it seems you are not executing the pipe in Linux. To narrow down the problem, test to put the path explicitly in the pipe string (avoid the macro variable). The test the exact same command on a Linux shell prompt. ... View more

Hi:   A minor correction. ODS and TAGSETS.EXCELXP do not create a "genuine" XLSX spreadsheet. TAGSETS.EXCELXP creates Spreadsheet Markup Language XML (designed for Office 2003), which was the XML precursor to the XLSX format. PROC EXPORT can create true XLSX files. There is a new ODS destination on the horizon that will also create XLSX files, but it is still pre-production, as described here: http://blogs.sas.com/content/sasdummy/2014/09/21/ods-excel-and-proc-export-xlsx/ cynthia ... View more

Pleased it is now working. It would be good if you could mark your post as answered. ... View more

Well, names of variables in SAS are restricted to underscores, letters and numbers.  Dashes and periods would not be valid characters for naming a variable. There are ways around this, but they make future programming very cumbersome. ... View more

Thanks everybody for help... ... View more

START_DATE - 3*60*60 ... View more

Also try this: %let path=c:\temp\; data _null_; infile "dir/b/a:-d  ""&path"" " pipe truncover; length ren $500.; input name $20.; newname=prxchange('s/test(14|0)/test14/',-1,name); ren=catx(' ','rename',quote(cats("&path",name)),quote(newname)); infile ren pipe filevar=ren; run; ... View more

I resolved it. I used the following condition in the loop %if &wday_cnt. = 1 and (&dsn1_var_cnt. = 0  or &dsn2_var_cnt. = 0) %then %do; %end; ... View more

I will be running it on daily basis and checking for the same day.   Yes I would check for records for that day so that I can send email based on which dataset has 0 records. ... View more

Or year = floor(Date/100); This is what I would do. ... View more

Thanks a lot for the help... ... View more

This seems to work, but today is April 1st, so month4 actually. If you're going to specify the month then you'll need to modify the code according rather than using today's date function. data have;   input id $2. mth1-mth24 2.;   cards; 10 2 4 5 8 9 13 4 6 7 9 8 2 5 7 6 4 6 8 9 5 2 6 11 8 12 5 6 8 4 2 12 5 6 8 6 8 2 5 7 6 4 5 8 9 5 2 4 16 2 ; run; *let date='31Mar2014'd; data want; set have; array current_year(12) mth24-mth13; array prev_year(12) mth12-mth1; month=month(today()); *month=month(&date.); current_YTD=0; previous_YTD=0; do i=1 to month;      current_YTD=current_YTD+current_year(i);      previous_YTD=previous_YTD+prev_year(i);      put i; end; run; ... View more

The number of brand_count_mth columns in the transposed output will correspond to the maximum number of observations per mktid-drgid-terrid-measure combination. One way to add a new column to the transposed output is to add a fourth (dummy) observation in the input dataset to a group which already has three. You could also add the missing column(s) to the output in a following datastep. data want2; array x brand_count_mht1-brand_count_mht4; set want1; run; PG ... View more

Simple just add the missing months to your data. ... View more