Re: BUG #14107: Major query planner bug regarding subqueries and indices
От | Victor Yegorov |
---|---|
Тема | Re: BUG #14107: Major query planner bug regarding subqueries and indices |
Дата | |
Msg-id | CAGnEboiCo+NuGab0PEnUQ2Nf0DXZv2Rvo2xWhMbi88n=guZB5w@mail.gmail.com обсуждение исходный текст |
Ответ на | Re: BUG #14107: Major query planner bug regarding subqueries and indices (Mathias Kunter <mathiaskunter@gmail.com>) |
Ответы |
Re: BUG #14107: Major query planner bug regarding subqueries
and indices
|
Список | pgsql-bugs |
2016-04-22 17:02 GMT+03:00 Mathias Kunter <mathiaskunter@gmail.com>: > EXPLAIN SELECT * FROM book WHERE name = 'Harry Potter' OR authorId IN > (SELECT id FROM author WHERE surname = 'Rowling'); > QUERY PLAN > > ------------------------------------------------------------------------------------------- > Seq Scan on book (cost=13.68..26792.88 rows=576709 width=40) > ... > > EXPLAIN SELECT * FROM book WHERE name = 'Harry Potter' UNION SELECT * FROM > book WHERE authorId IN (SELECT id FROM author WHERE surname = 'Rowling'); > QUERY PLAN > > ------------------------------------------------------------------------------------------------------------ > HashAggregate (cost=454.87..455.99 rows=112 width=29) > Queries return different number of rows, meaning they're not fully equivalent. Can you provide full DDL for the tables, constraints and indexes in use? -- Victor Y. Yegorov
В списке pgsql-bugs по дате отправления: