On Tue, 24 Jun 2014 21:02:22 +0100 "Dave Wellman" <[email protected]> wrote:
> 2) Assuming that my processing follows this pattern: empty table > T1 completely, insert a number of rows, insert/select from T1 into > T2. On the 'select' processing will the 'rowid' > ** always ** start at 1? http://www.schemamania.org/sql/#rank.rows If the use of the sequential number is ephemeral, why store it? Just generate it with a self-join. If the use of the sequential number is to identify the row in perpetuity, you may find it's more trouble than it's worth. --jkl _______________________________________________ sqlite-users mailing list [email protected] http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
