Reading your response, thinks. I did however find that the incoming data did not conform to what I expects (contiguous series_id), so the code worked, my thinking did not.
On 5/10/10, Tim Romano <tim.romano...@gmail.com> wrote: > The select/group by part of your statement will group table SERIESDATA by > text-column series_id (aliased to id) and return the min and max data_index > for each grouping, assuming those columns are populated with data for each > row. The set will have three columns and some number of rows, one per id. > > id | min(data_index) | max(data_index) > > However, it is not clear to me where you want to put that aggregated set. Do > you have another *table* called SERIESID with those three columns in it? > > > Regards > Tim Romano > Swarthmore PA > > > On Mon, May 10, 2010 at 2:43 AM, Matt Young <youngsan...@gmail.com> wrote: > >> # series data looks like: >> create table seriesdata ( >> data_index INTEGER PRIMARY KEY autoincrement, >> series_id text, >> year numeric, >> value numeric); >> # and is filled >> insert into seriesid >> select >> s.series_id as id, min(data_index),max(data_index) >> from >> seriesdata as s >> group by >> id; >> >> # I intend seriesid to pick up the minand max values of data_index for >> each unique series_id >> _______________________________________________ >> sqlite-users mailing list >> sqlite-users@sqlite.org >> http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users >> > _______________________________________________ > sqlite-users mailing list > sqlite-users@sqlite.org > http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users > _______________________________________________ sqlite-users mailing list sqlite-users@sqlite.org http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
