The question is far too vague. What is your requirements and constraints? What sw do you have available, or are you looking for an investment? How much and what kind of data you have? You talk about "upload", what does that mean to you? ... View more

https://communities.sas.com/message/145485#145485 ... View more

This does not look like a problem where you would need to get into the complexity of MACRO programming.  Normal programming will do just fine. It will probably work better if your two source datasets had some type of ID variable to identify the rows so that you can match the SAL to the proper DAYS.  Perhaps MONTH? If you are willing to assume that the tables should match row by row then you can use merge without a BY statement or two independent SET statements to read one row from each source table per loop of the data step. data main;   input sal; cards; 100 200 run; data mon_sal;   input days; cards; 2 3 run; data mon_sal;   set main;   set mon_sal;    mo_days = days*sal; run; ... View more

data l; input id age; cards; 2 2 2 2 run; data l2; input  amt rupees; cards; 3 4 5 6 run; proc sql; select count(*) into :cnt from l2; quit; %macro cal; %do i =1 %to &cnt; data _null_; set l; call symput('id',id); call symput('age',age); run; data l2; set l2; amt =amt *&id; rupees=rupees*&age; run; %end; %mend; %cal; output shd be for L2 6 8 10 12 ... View more

Not sure what you are asking.  Take a look at the code at the bottom of the program at: http://www.sascommunity.org/mwiki/images/d/da/Hebrew_islamic_dates.sas It sends emails to addresses, from a SAS dataset, like those shown in your example. ... View more

TABS are not treated the sames as spaces unless you tell it differently on the INFILE statement. You could use DLM='2009'x to treate space and tab as delimiters Or use EXPANDTABS to have it convert the tabs to spaces on the way into the input buffer. ... View more

thqs it worked ... View more

EASY. data have; input id name & $20.     value2  & $20.            value3 & $20.; cards; 1  a,b      a:1,b:1              a:2,b:2 2  a,b      a:3,b:4              a:3,b:4 3  a        a:5                   a:6 4  a,b,c,d  a:10,b:20 c:30 d:40   a:10,b:20 c:30 d:40 ; run; data want(drop=value: i n name); set have; length _name _value2 _value3 $ 20; n=max(countw(value2,', '),countw(value3,', ')); do i=1 to n;   _name=scan(name,i,', ');   _value2=scan(value2,i,', ');   _value3=scan(value3,i,', ');   output; end; run; Ksharp ... View more

Liked. Art's solution makes PRX solution of mine looks like a chunky hamburger from Macdonald's.  ... View more

It is very intesting . I don't realize it will appear so weird result. Try it. data l2; input id no; cards; 1 100 2 200 3 300 3 400 ; run; proc sql; select count(*)  into :l2 from (select *  from l2 group by id having id=3 and no=max(no) ) ; quit; %put &l2; Ksharp ... View more

if it is having 5 words and length <10 and length greate than 200 Actually,I do not understand what you mean. Since you want length<10 ,why do you need > 200 ? data l; infile cards  truncover; input add $ 1-100; cards; q w e r t y r t t t t qwerret w a d f g g run; data l2; set l; length add_QC1 $ 100; if countw(add) eq 5 and LENGTH(COMPRESS(add, ' ') ) <10 or  LENGTH(COMPRESS(add, ' ')) >200 then add_QC1=' '; else add_QC1=add; run; Ksharp ... View more

Edit: Art is right, there is no way to do it using proc datasets. if using data step; data want; length sex $100.; set class; run; Regards, Haikuo ... View more

Is it at least 9.1.3?  The modifiers weren't even added until then.  However, if you are on 9.1.3, then you could just specify exactly which characters you want to keep in the second field (which is currently only ,, ). Regardless, is sounds like you already have an acceptable solution. ... View more