* Dennis Cote <[EMAIL PROTECTED]> [2007-02-03 17:20]: > I suspect the reduction from executing three statements > (insert, select, insert) down to only two (insert insert) would > probably provide about the same performance increase as the 5% > to 10% speedup he saw by replacing the separate select with the > VDBE stack lookup hack.
Ah, you mean the SELECT / sometimes-INSERT / INSERT strategy might not be any faster than INSERT / INSERT-with-subselect because the former is 3 statements and the latter is just 2? Hmm, that’s something I’d definitely benchmark before deciding. It’s a pity that INSERT OR IGNORE (apparently?) does not set last_insert_id properly regardless of outcome, otherwise it could be reduced to just two INSERTs doing absolutely no duplicate work. Regards, -- Aristotle Pagaltzis // <http://plasmasturm.org/> ----------------------------------------------------------------------------- To unsubscribe, send email to [EMAIL PROTECTED] -----------------------------------------------------------------------------
