On Mon, Aug 13, 2012 at 5:13 AM, Linda Alvord <[email protected]> wrote: > How can I amend adjacent columns to remove duplicates? > > 2 10$'ABCDEEEFGH' > ABCDEEEFGH > ABCDEEEFGH > > 2 7$'ABCDEFG' > ABCDEFG > ABCDEFG
Note that "amend" is a J primitive which preserves the shape of its right argument, so I am going to presume that you are only hinting at what you want done here. If you can relax the "adjacent" part, it's simple: ~.&.|: 2 10$'ABCDEEEFGH' ABCDEFGH ABCDEFGH If you really need to preserve duplicate columns which are not a part of an adjacent sequence, the complications that Henry spelled out will accomplish that task. FYI, -- Raul ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
