Clemens Ladisch wrote:
Luuk wrote:
On 01-01-18 03:14, Shane Dev wrote:
select * from nodes where not exists (select * from edges where
child=nodes.id);
Changing this to:
select * from nodes where not exists (select 1 from edges where
child=nodes.id);
saved in my test about 10% of time
Then I have to doubt your test; the generated code (see the EXPLAIN
output) is exactly the same.
Yes, the execution plans are the same. Still the second is faster!
Another query that uses EXCEPT instead of NOT IN and that has yet
another execution plan:
select id from nodes except select child from edges;
_______________________________________________
sqlite-users mailing list
sqlite-users@mailinglists.sqlite.org
http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users
