Re: HELP WITH A QUERY
| От | Jasen Betts | 
|---|---|
| Тема | Re: HELP WITH A QUERY | 
| Дата | |
| Msg-id | ickegp$lia$1@reversiblemaps.ath.cx обсуждение исходный текст | 
| Ответ на | HELP WITH A QUERY (JORGE MALDONADO <jorgemal1960@gmail.com>) | 
| Ответы | Re: HELP WITH A QUERY | 
| Список | pgsql-novice | 
On 2010-11-24, JORGE MALDONADO <jorgemal1960@gmail.com> wrote: > --0015174c188eeae5060495d4ae01 > Content-Type: text/plain; charset=ISO-8859-1 > > I have 2 tables and each one has a date field. Is it possible to write a > simple query to get the records with the dates in table1 but not in table2? > For example: > > TABLE 1 > > ----------------------------------------------------------------------------- > Branch_Name Sales Date > ----------------------------------------------------------------------------- > Los Angeles 1500 May-15-2010 > San Diego 250 May-27-2010 > Los Angeles 300 May-28-2010 > Boston 700 May-28-2010 > > > TABLE 2 > > ----------------------------------------------------------------------------- > Date Sales > ----------------------------------------------------------------------------- > May-27-2010 250 > May-20-2010 535 > May-21-2010 320 > May-22-2010 750 > The query result should be: > May-15-2010 > May-28-2010 all examples untested eg "NOT IN": select date from table1 where date not in ( select distinct date from table2); or this "LEFT JOIN WHERE NULL: select table1.date from table1 left outer join table2 on table2.date=table1.date where table2.date is null. or this "EXCEPT": select date from table1 except select date from table2 If you want a tested example make it easy for us, post the SQL to make those tables. > I guess that a query with a subselect and the EXCEPT keyword would work but > I would like to know if there is a simpler alternative? EXCEPT is probably the best way, but all three methods will probably produce similar query plans and have similar performance. except is not really a subquery as both queries are on the same level. "not in" uses a subquery. -- ⚂⚃ 100% natural
В списке pgsql-novice по дате отправления: