Re: Using b-tree index for >= condition when joining

2020-05-17 Thread Łukasz Dąbek
On 17/05/2020 at 04:22, Tom Lane wrote:> Note that you're not really getting the same plan that way: it's not a left join anymore, because you put a strict constraint on the join's inner relation, so the planner realizes it doesn't have to produce any null-extended rows. You could make it work w

Re: Using b-tree index for >= condition when joining

2020-05-16 Thread Tom Lane
=?UTF-8?B?xYF1a2FzeiBExIViZWs=?= writes: > I am having a problem with nudging postgres to choose a good plan for > a query involving a left join and an inequality constraint on a column > with b-tree index. > ... > It looks like the inequality on date isn't pushed down below the left > join? Nope

Using b-tree index for >= condition when joining

2020-05-16 Thread Łukasz Dąbek
Hello All! I am having a problem with nudging postgres to choose a good plan for a query involving a left join and an inequality constraint on a column with b-tree index. Let's say both tbl1 and tbl2 tables have date column with an index on it. Queries like "SELECT * FROM tbl1 WHERE date >= CONST