Hi I have a table with the following columns and some sample data:



Basically, this is a log from my webstats, and I would like to make the following SQL query:

Count distinct IP addresses for a day, and if an IP is repeated within thirty minutes of its first occurrence, count it only as one. If this IP address occurs again after this thirty minutes count it as new visit and perform the same check.

Basically, with this query, I would get the unique visitors to my site.

Can someone please help?

Thank you,
Giorgio

-----
www.bannerkozpont.hu

are you sure?

what if the visits were like this --

V...7min...V...7min...V...7min...V...7min...V...7m in...V

the way i would look at it, that's only one visit

by your definition, it's two

Hi r397,

I would like to thank you for your kind response.

You are right that it can be considered as one visit as well. In fact that would make my life easier as I would just count distinct IPs for the entire day and that is it.

Yet, most webstatistics companies put this 30 minute limit for 1 unique visit, and even they admit that it is not perfect.

Giorgio

www.bannerkozpont.hu

Hi,

Would someone know if this is even possible in mysql?

Is there a time function that I can use in the select statements perhaps, like datediff?

Or could I use "interval" to group identical IPs into 30 minute groups and count the number of these groups?