John, Sort, at least on distibuted platforms, enables you to compare in a JOIN like syntax where you can decide to keep OUTER (of the join) only. If all records are duplicated, you'll get an empty file. else you'll get 1 or more records in the outer condition. So, if the files goes to Unix, sort can help.
ITschak On Thu, Aug 15, 2013 at 7:11 PM, John Gilmore <[email protected]> wrote: > Eliminating duplicates gives > > A > B > > Eliminating all duplicated values is a different and rare requirement. > Eliminating duplicate values ordinarily means that if there are d >= > 1 instances of a value v in a list, file, whatever, then d - 1 of > these values v are eliminated. > > John Gilmore, Ashland, MA 01721 - USA > > ---------------------------------------------------------------------- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to [email protected] with the message: INFO IBM-MAIN > ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
