On Tue, Aug 16, 2011 at 8:20 AM, Rich Shepard <[email protected]>wrote:
> On Tue, 16 Aug 2011, Roderick A. Anderson wrote: > > > Can we see another snapshot of the data? And (did I miss it?) which three > > columns. > > Rod, > > Yep, and yep. > > Data: > > \N CVS 1994-01-20 Conductance, Specific 460 uS/cm t > \N \N \N > \N CVS 1994-01-20 Conductance, Specific 522 uS/cm t > \N \N \N > > (Fred: I think that pg_dump does use tabs as column separators, and there > are spaces within a column as the above demonstrates. These data were > extracted from Excel spreadsheets.) > > The three columns are the second, third, and fourth, named loc_name, > sample_date, and param. > > The current client staff can't figure out either how there could be two > different values for specific conductance at the same location on the same > date when both were supposedly checked for quality (the 't' in the seventh > column). > > Rich > I have no idea how to actually do it, but how is this as a strategy? Add a unique column (record #) Remove col. 2. Remove duplicate entries in the result. Note which records have been removed and remove them from the original. Repeat for cols. 3 and 4. -Denis _______________________________________________ PLUG mailing list [email protected] http://lists.pdxlinux.org/mailman/listinfo/plug
