SELECT WHERE date has less then "X" row count

[curb]

I have rows with the same unixtime date with different times. How can I select rows that has less then 3 duplicate dates.

The example would pull march 4th & 5th and not march 3rd since it's more then 3.

Example without unixtime:

march 3rd 2008 12:08pm
march 3rd 2008 10:28am
march 3rd 2008 09:02pm

march 4th 2008 12:08pm
march 4th 2008 10:28am

march 5th 2008 09:02pm

[bogu]

Try COUNT, HAVING AND GROUP BY...

[curb]

How would I count based on the Month and date, excluding the time when it comes to unixtime?

[kakki]

SELECT COUNT(*) as _times, MIN(unixtime) AS _unixtime
FROM xxx
GROUP BY unixtime
HAVING COUNT(*)>3

[nemonoman]

This sort of problem is why you want to use DATETIME and TIMESTAMP datatypes.

SELECT
FROM table
GROUP BY date_format(mytimestamp, '%Y%m%d')
HAVING COUNT ()>2