MySQL: Count the distinct rows per day

Question

I have an interesting query I need to do. I have a table with an INT column containing ip address numbers (using INET_ATON ), and a timestamp column. I want to be able to count the number of unique ip address columns there are per day. That is, how many distinct ip rows there are in each day. So, for example, if an ip address is in the same day twice, it counts as 1 in the final count, however if the same ip address is in another day it'll be counted there will be a second count for it.

Example Data:

PK | FK  | ipNum      | timestamp
11 | 404 | 219395     | 2013-01-06 22:23:56
7  | 404 | 467719     | 2013-01-06 22:23:41
8  | 404 | 4718869    | 2013-01-06 22:23:42
10 | 404 | 16777224   | 2013-01-06 22:23:56
5  | 404 | 1292435475 | 2013-01-06 22:23:25
12 | 404 | 1526990605 | 2013-01-06 22:23:57
6  | 404 | 1594313225 | 2013-01-06 22:23:40
4  | 404 | 1610613001 | 2013-01-06 22:23:23
9  | 404 | 1628635192 | 2013-01-06 22:23:55
1  | 404 | 2130706433 | 2013-01-06 21:29:38
2  | 407 | 2130706433 | 2013-01-06 21:31:59
3  | 407 | 2130706433 | 2013-01-06 21:32:22

Answer 1

SELECT  DATE(timestamp) Date, COUNT(DISTINCT ipNum) totalCOunt
FROM    tableName
GROUP   BY  DATE(timestamp)

• SQLFiddle Demo

Answer 2

Here's how you'd get counts per day for the last 7 days:

select
    count(*) as count,
    date(timestamp) as date
from
    tablename
where
    timestamp >= date_sub(curdate(), interval 7 day)
group by 
    date;

+-------------+------------+
| count       | date       |
+-------------+------------+
| #forThatDay | 2020-02-21 |
| #forThatDay | 2020-02-22 |
| #forThatDay | 2020-02-22 |
| #forThatDay | 2020-02-23 |
| #forThatDay | 2020-02-24 |
| #forThatDay | 2020-02-25 |
| #forThatDay | 2020-02-26 |
+-------------+------------+
7 rows in set (0.03 sec)

group by the ipNum column first to get distinct counts of that column per day:

select
    count(*) as count,
    date(timestamp) as date
from
    tablename
where
    timestamp >= date_sub(curdate(), interval 7 day)
group by 
    ipNum, date;

Answer 3

$log_date     = date('Y-m-d H:i:s');
$log_date     = date('Y-m-d H:i:s', strtotime($log_date.' -1 hour'));
SELECT ipNum, COUNT(ipNum), COUNT(DISTINCT ipNum), DATE(timestamp), timestamp FROM tableName  WHERE `timestamp` > '".$log_date."' GROUP BY ipNum ORDER BY DATE(timestamp) DESC  

THIS WILL GIVE YOU A RESULT LIKE

 ip                  TIME               COUNTIPS
11.237.115.30     2018-01-27 19:13:51       1
21.744.133.52     2018-01-27 19:14:03       1
44.628.197.51     2018-01-27 19:48:12       14