Then maybe move your control file to another folder and do not take into account "last modified date" for file deletion. Or create TODO_DONE list (data set) where you keep inventory of your files (FILENAME column) and STATUS column ("TODO" or "DONE" values) and/or DATE_DELETED column (. if to be deleted, and some real date when it was done). Just solve the problem instead of creating one. ... View more

Hi PerBundgaardLarsen, The bigger question to ask is why do you want to change a file's "modified date" when you do not modify the file, but only read it? I suspect you are solving the wrong problem... ... View more

Hi PerBundgaardLarsen, You can "touch" your file using the following 2 commands: x "cd ""&indir"""; x "copy /b ""&infile""+"; Hope this helps. ... View more

Hi mounikag, You can do it using the following approach: data x; length subj astdt 8 inj1-inj4 $10; infile datalines truncover; input subj astdt yymmdd10. inj1-inj4 $; format astdt yymmdd10.; datalines; 101 2020-06-03 2019-07-18 2019-08-15 2019-09-10 2019-11-14 102 2019-08-07 2019-06-12 2019-07-08 2019-08-12 2019-09-30 103 2020-01-09 2019-11-13 2019-12-16 2020-01-08 2020-03-04 104 2019-12-13 2019-11-13 2019-12-16 2020-01-08 2020-03-04 105 2019-05-15 2019-05-13 106 2019-08-05 2019-07-17 2019-08-14 107 2019-08-21 2019-07-17 2019-08-14 108 2020-06-30 2020-05-27 2020-06-24 2020-07-22 ; DATA WANT; set x; array inj $ inj1-inj4; array injn injn1-injn4; no_injs_before_astdt = 0; do i=1 to dim(inj); injn[i] = input(inj[i],yymmdd10.); no_injs_before_astdt + (astdt > injn[i] > .); end; run; You can drop any variables you don't like to keep. Also, in your data last line you have an error, No.of injs before astdt should be equal 2, not 3.   This code essentially converts you character dates into numeric, and then loops through array elements and increments no_injs_before_astdt every time when (astdt > injn[i] > .)   Hope, this helps.     ... View more

SASKiwi, I think SAS will be available on quantum computers as soon as quantum computers will be available for SAS. ... View more

@SASKiwi, I am glad you chimed in. You got it almost right, with the reversed precision: my answer was "not yes and not no" at the same time. 🙂 I like the idea you suggested (even though I prefer dogs over cats), although I wouldn't call it "deviation" as these are all parts of the same - our pitiful attempt to explain the inexplicable world. ... View more

Thank you, leelee, for your very thoughtful comment (up until now I thought that everyone took this picture for granted). You say, you are "a little late reading this article". However, relatively speaking, you are a month early... And the answer to your question is Yes. No.  😀 ... View more

Hi contactnishan,   You can try this: libname inlib cvp 'c:\source_folder' inencoding='wlatin1'; libname outlib 'c:\target_folder' outencoding='UTF-8'; proc copy in=inlib out=outlib noclone; select dataset1 dataset2; run; For more information, see this blog post https://blogs.sas.com/content/sgf/2020/08/12/expanding-lengths-of-all-character-variables-in-sas-data-sets/ .   Hope this helps. ... View more

Hi MarieT,   You can do it this way, for example:   %if &nb=0 %then %do; data _null_; abort return; run; %end; See ABORT Statement. Hope this helps.       ... View more

@bedzekim , be careful with using cat(zip) eq '00201 ! This will not work for strings with leading zeros. Run this code: data _null_; zip=00201; if cat(zip) eq '00201' then put 'Equal'; else put 'Not equal'; run; and you will see in the SAS log: Not equal This is because leading zeros are dropped from numeric value.   ... View more

Hi pacman94,   You can achieve this with the following code: options missing=' '; data test; set raw; average_values = (value1+value2+value3+value4) / 4; run;   Hope this helps. ... View more

Indeed, this might be a classic case of improper design. What is the final goal there? When that is formulated then a simple solution can be found, most likely without splitting your data set into many. ... View more