On 31 Jan 2020, at 9:27pm, Keith Medcalf <kmedc...@dessus.com> wrote:
> You are however correct that this is an "application consistency" problem > more than an SQLite problem and it is a lot of change for little actual > benefit. How about this ? A program (possibly part of the SQLite precompiled suite, possibly from a third party) goes through any SQLite database as an integrity checker. One of its jobs is to check that column constraints are not violated. This cannot possible be wrong. A SQLite database with violated CHECK clauses is, by definition, corrupt. However, because the checks are performed on the values input, not the values stored, some data in the database does violate a CHECK constraint. Can this happen, given the behaviour Keith identified ? If so, I would say that something is wrong. _______________________________________________ sqlite-users mailing list sqlite-users@mailinglists.sqlite.org http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users
