Format dates in a unique format

[NewUsrStat]

Hi guys, 

I have a data set with dates in really really all types of "formats" like:

01/1/2001, 1/1/2001, 1/01/2001, 1/1/00, 01/1/00, 1/01/00. Is there a quick way to format all these exceptions in dd/mm/yyyy? 

Thank you very much. If I use for example: format date*, does it will take all the exceptions? The records are really many and so I cannot check for each single case. I must be sure that all the dates are in the same file format.

Thank you in advance

[Tom · Accepted solution]

The only way this question makes any sense is if the variable is a character string instead of a DATE.  (SAS stores dates as number of days since 1960).

If those strings are all supposed to represent day, month, and year values then you can use the DDMMYY informat to convert the string into a date value.  You could then leave it as a numeric value and apply any date type FORMAT to it you want. 

data want;
  set have;
   date = input(date_string,ddmmyy10.);
   format date yymmdd10.;
run;

Or you could use the PUT() function to use the format to generate a new character string.  

Then if you want you could overwrite the original character variable's vlaue.

data want;
  set have;
   date_string = put(input(date_string,ddmmyy10.),yymmdd10.);
run;

NOTE: I have used the YYMMDD format in my code so that the dates will be displayed in Year, Month, Day order.  If you display dates in DMY or MDY order you will confuse 50% of your audience.  If you really insist on seeing the dates in dd/mm/yyyy style then just use the DDMMYYS10. format instead.