James, probably a view is not slower, but if you need this data several times...
I wanted to do an update because we are using this data several times in the process, and I did not want to add the missing pieces each time. gert 2013/11/7 James K. Lowden <jklow...@schemamania.org> > On Mon, 4 Nov 2013 13:01:37 +0100 > Gert Van Assche <ger...@datamundi.be> wrote: > > > Thanks James -- the select query is something I could do, but the > > update one I could not get that one right. > > I was considering to create a new table based on the select query, > > but since the real data set is millions of records, an update was the > > best solution. > > You're welcome, Gert. Implicitly I was recommending a view: instead of > periodically updating a table with derived information, let SQLite > always provide up-to-date information by deriving it on demand. > > With a dataset of millions of records, is performance on such a view > much different from that of a table? I don't happen to have such a > sample handy, and I'm curious what you've seen. > > --jkl > > _______________________________________________ > 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
