On Tue, Oct 23, 2012 at 1:52 PM, Raul Miller <rauldmil...@gmail.com> wrote: > You will need some way of identifying each of these types of data. > > You have said that there's nothing unique about Attribute 0 except > that it appears before Attribute 1, but this does not seem like a > complete description of the rules about how identify all of these.
Raul, I wasn't clear. In this example, "Attribute 1", "Attribute 2", "Attribute 3", and "Location" are the explicit attribute names used in the file. Those are not the real ones, but the attribute names do create a known set. The example nouns I gave would be legitimate if simple ones; I didn't use "Attribute 1" as a placeholder but as a real value. In other words, you can figure out the location of the values for Attribute 0 with a phrase something like 1 |. (<'Attribute 1') -:"0 d1 Attribute 1's value can be found by something like _1 |. (<'Attribute 1') -:"0 d1 For those used to using format/zulu, d1 looks like > applied to the output of a2b. Does that help? Bill -- Bill Harris http://facilitatedsystems.com/weblog/ ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
