> Clients can be referred to by one or more names and so there's
> another table:
>
> CREATE TABLE client_names (
> id integer,
> name text
> );
>
> Names aren't unique. Two clients can have the same name.
But the combination of id and name are unique: hence try this:
CREATE TABLE client_names (
id integer,
name text,
primary key (id, name))
That creates a unique index on (id, name). But you won't ever be able to
drop it. If you will be modifying your schema in future (perhaps!) you
can do:
CREATE TABLE client_names (
id integer,
name text)
CREATE UNIQUE INDEX client_names_primary_key ON client_names (id, name)
Hugh
_______________________________________________
sqlite-users mailing list
sqlite-users@sqlite.org
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
