Question

I’ve got a table called “double_select” that looks like this:

itemID | orderID | status
------   -------   ------
   1        1       ready
   2        1       ready
   3        1      waiting
   4        2      complete
   5        3       ready
  ...      ...       ...

And with one SQL command I want to retrieve:

• all items with status “ready”
• all other items for any orderID returned above

So in the above table snippet, I’d like to return itemID’s 1, 2, 3 & 5.

I can do this with this statement:

SELECT `a`.* FROM `double_select` AS `a`, `double_select` AS `b` WHERE `a`.`status` = "ready" OR (`a`.`status` != "ready" AND `b`.`orderID` = `a`.`orderID` AND `b`.`status` = "ready") GROUP BY `a`.`itemID`;

But it doesn’t seem very clean. Is there a better way of doing this?

Table details are below, thanks a lot,

James

CREATE TABLE `double_select` (`itemID` int(11) NOT NULL auto_increment,`orderID` int(11) default NULL,  `status` varchar(128) default NULL,  PRIMARY KEY  (`itemID`),  KEY `Status` (`status`)) ENGINE=MyISAM AUTO_INCREMENT=11 DEFAULT CHARSET=latin1;
insert  into `double_select`(`itemID`,`orderID`,`status`) values (1,1,'ready'),(2,1,'waiting'),(3,1,'waiting'),(4,2,'complete'),(5,2,'ready'),(6,3,'ready'),(7,3,'ready'),(8,4,'complete'),(9,5,'failed'),(10,6,'complete');

Answer 1

Subquery example:

select * 
from double_select
where orderID in 
    (select orderID 
     from double_select 
     where `status` = 'ready')