Question

I’m trying to do something like

SELECT `v1`.`date`, COUNT(DISTINCT(`v2`.`id`)) AS `visits` 
FROM `visitor_visits` AS `v1` JOIN `visitor_visits` AS `v2` 
ON (`v1`.`date` = `v2`.`date`) GROUP BY `v1`.`date`, `v2`.`date` 
ORDER BY `v1`.`date

To get distinct dates, and a count of all other dates that match.

I’m pretty sure this is the wrong query since the date column is non-unique.

Answer 1

What’s wrong with:

SELECT v1.date, COUNT(v1.id) AS visits 
FROM visitor_visits AS v1
GROUP BY v1.date
ORDER BY v1.date

?

ps. I assume that id is primary key (unique).