2009/1/26 Grzegorz Jaśkiewicz <gryz...@gmail.com> > Hey folks, > > I have question really for all mighty developers, but don't want to > spam -hackers with it. > > why : > select * from foo where X in (1,1,1,1,1,1,1,1) --- same values in search. > or select * from foo where (x,y) in > ((1,2),(1,2),(1,2),(1,2),(1,2),(1,2),(1,2)); > > never gets optimized by planner, etc ?
I would guess that optimizing silly-written queries was always a low-priority task... IMHO this is good topic for -hackers list.. and probably not so hard to implement :) BTW, test on CVS HEAD: CREATE TABLE atest(id integer primary key); insert into atest select x from generate_series(1,100000) x(x); ANALYZE atest; EXPLAIN ANALYZE SELECT * FROM atest where id in (1,2,3,3,3,3,3,3,3,5,5,5,5,5,5,5,5,5,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1); EXPLAIN ANALYZE SELECT * FROM atest where id in (1,2,3,5); shows that second query is 2.5 times faster than the first ( 0.170 ms / 0.070 ms). > > Is it just not worth optimizing from pg side? I am sure, it would make > sense to actually reorder these values, so that index/whatnot could > pick it up faster. > > Just another one of those, 'why' (not) questions from my side. > > thanks. > > -- > GJ > > -- > Sent via pgsql-general mailing list (pgsql-general@postgresql.org) > To make changes to your subscription: > http://www.postgresql.org/mailpref/pgsql-general > -- Filip Rembiałkowski
