Dear all, With https://www.sqlite.org/withoutrowid.html in mind, assuming something simple as:
create table if not exists SHEETS( NODE integer not null, SHEET integer not null, NAME text, -- often NULL, sometimes short text value (keyword) ..., -- 2 or 3 other small properties primary key(NODE,SHEET) ) without rowid; -- makes sense because small row and natural primary key is not integer Now what if an index on NAME becomes valuable? Will such a secondary index have a higher storage cost or performance cost than with an ordinary table? create index if not exists SHEETS_NAME on SHEETS(NAME); -- Best Regards, Meilleures salutations, Met vriendelijke groeten, Olivier Mascia _______________________________________________ sqlite-users mailing list [email protected] http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users
