Ragnarok

Thank-you very much for your help

I meen like if I have a table like this id|date page1|2006-01-13 page2|2006-01-26 page1|2006-01-26 page3|2006-02-05 page2|2006-02-09 it would find that there are 3 unique id (page1, page2, page3) names and only count those id names

I have been thinking, all I really need to do is count the different ids, it doesn't matter which date goes with it. Is it posible to do it that way? Also would this way be faster?

I am doing a versioning system so that every time you edit a page, the old page is still kept but the new page is shown insted. I have that working find but when I am trying to list the pages I am having problems getting a query to do it. I have a table like this: [code]CREATE TABLE `pages` (   `id` varchar(25) NOT NULL default '',   `date` datetime NOT NULL default '0000-00-00 00:00:00',   `title` varchar(25) NOT NULL default '',   `content` mediumtext NOT NULL,   PRIMARY KEY  (`id`,`date`),   FULLTEXT KEY `Search` (`title`,`content`) );[/code] I want to count all of the ID's with the most recent date. I have tried: [code]SELECT COUNT(*) as total FROM pages GROUP BY date HAVING MAX(date)[/code] returns 0 [code]SELECT COUNT(MAX(date)) FROM pages GROUP BY date[/code] but that returns this error [code]#1111 - Invalid use of group function[/code] Does anyone know of a solution?