mySql: count number of rows that have the same data in a column

5
votes

I am trying to select everything in a table, and also count the number of rows in a table that have the same data.

SELECT *, COUNT(thedate) daycount FROM `table` ORDER BY thedate DESC

My hope is to have one query that outputs the date and number of rows associated with that date, and the looped output will be something like this:

Jan 1, 2000 (2 rows)
col1, col2, col3, col4
col1, col2, col3, col4

Jan 1, 2000 (3 rows)
col1, col2, col3, col4
col1, col2, col3, col4
col1, col2, col3, col4

Jan 1, 2000 (6 rows)
col1, col2, col3, col4
col1, col2, col3, col4
col1, col2, col3, col4
col1, col2, col3, col4
col1, col2, col3, col4
col1, col2, col3, col4

etc...

Does this make sense?

5
mysql

5 Answers

3
votes

If you have a table that looks like this:

CREATE TABLE yourtable
(
    datefield DATETIME,
    col1 VARCHAR(20),
    col2 INT NOT NULL,
    col3 TINYINT NOT NULL,
    col4 CHAR(5)
);

and you wanted the counts of duplicate col1.. col4 per given day, you would run this query

SELECT
    COUNT(datefield) datefield_count,
    LEFT(all_fields,10) datefield,
    SUBSTR(all_fields,11) all_other_fields
FROM
(
    SELECT
        DATE(datefield) datefield,
        CONCAT(DATE(datefield),'|',
        COALESCE(col1,'< NULL >'),'|',
        COALESCE(col2,'< NULL >'),'|',
        COALESCE(col3,'< NULL >'),'|',
        COALESCE(col4,'< NULL >'),'|') all_fields
    FROM
         yourtable
) A
GROUP BY all_fields;

Here is some sample data and the result of the query:

mysql> DROP TABLE IF EXISTS yourtable;
Query OK, 0 rows affected (0.04 sec)

mysql> CREATE TABLE yourtable
    -> (
    ->     datefield DATETIME,
    ->     col1 VARCHAR(20),
    ->     col2 INT,
    ->     col3 TINYINT,
    ->     col4 CHAR(5)
    -> );
Query OK, 0 rows affected (0.11 sec)

mysql> INSERT INTO yourtable VALUES
    -> (DATE(NOW() - INTERVAL 1 DAY),'rolando',4,3   ,'angel'),
    -> (DATE(NOW() - INTERVAL 1 DAY),'rolando',4,3   ,'angel'),
    -> (DATE(NOW() - INTERVAL 1 DAY),'rolando',4,3   ,'angel'),
    -> (DATE(NOW() - INTERVAL 1 DAY),'rolando',4,NULL,'angel'),
    -> (DATE(NOW() - INTERVAL 1 DAY),'rolando',4,NULL,'angel'),
    -> (DATE(NOW() - INTERVAL 2 DAY),'rolando',4,2   ,'angel'),
    -> (DATE(NOW() - INTERVAL 2 DAY),'rolando',4,2   ,'angel'),
    -> (DATE(NOW() - INTERVAL 2 DAY),'rolando',4,2   ,'angel'),
    -> (DATE(NOW() - INTERVAL 2 DAY),'rolando',4,2   ,'angel'),
    -> (DATE(NOW() - INTERVAL 2 DAY),'rolando',4,NULL,'edwards'),
    -> (DATE(NOW() - INTERVAL 2 DAY),'rolando',4,NULL,'angel'),
    -> (DATE(NOW() - INTERVAL 3 DAY),'rolando',5,2   ,'angel'),
    -> (DATE(NOW() - INTERVAL 3 DAY),'rolando',5,2   ,'angel'),
    -> (DATE(NOW() - INTERVAL 3 DAY),'rolando',4,2   ,'angel'),
    -> (DATE(NOW() - INTERVAL 3 DAY),'pamela' ,4,2   ,'angel'),
    -> (DATE(NOW() - INTERVAL 3 DAY),'pamela' ,4,NULL,'edwards'),
    -> (DATE(NOW() - INTERVAL 3 DAY),'pamela' ,5,2   ,'angel'),
    -> (DATE(NOW() - INTERVAL 3 DAY),'pamela' ,5,2   ,'angel'),
    -> (DATE(NOW() - INTERVAL 3 DAY),'rolando',4,2   ,'angel'),
    -> (DATE(NOW() - INTERVAL 3 DAY),'rolando',4,2   ,'angel'),
    -> (DATE(NOW() - INTERVAL 3 DAY),'rolando',4,NULL,'edwards'),
    -> (DATE(NOW() - INTERVAL 3 DAY),'rolando',4,NULL,'angel')
    -> ;
Query OK, 22 rows affected, 3 warnings (0.03 sec)
Records: 22  Duplicates: 0  Warnings: 3

mysql> SELECT * FROM yourtable;
+---------------------+---------+------+------+-------+
| datefield           | col1    | col2 | col3 | col4  |
+---------------------+---------+------+------+-------+
| 2011-06-30 00:00:00 | rolando |    4 |    3 | angel |
| 2011-06-30 00:00:00 | rolando |    4 |    3 | angel |
| 2011-06-30 00:00:00 | rolando |    4 |    3 | angel |
| 2011-06-30 00:00:00 | rolando |    4 | NULL | angel |
| 2011-06-30 00:00:00 | rolando |    4 | NULL | angel |
| 2011-06-29 00:00:00 | rolando |    4 |    2 | angel |
| 2011-06-29 00:00:00 | rolando |    4 |    2 | angel |
| 2011-06-29 00:00:00 | rolando |    4 |    2 | angel |
| 2011-06-29 00:00:00 | rolando |    4 |    2 | angel |
| 2011-06-29 00:00:00 | rolando |    4 | NULL | edwar |
| 2011-06-29 00:00:00 | rolando |    4 | NULL | angel |
| 2011-06-28 00:00:00 | rolando |    5 |    2 | angel |
| 2011-06-28 00:00:00 | rolando |    5 |    2 | angel |
| 2011-06-28 00:00:00 | rolando |    4 |    2 | angel |
| 2011-06-28 00:00:00 | pamela  |    4 |    2 | angel |
| 2011-06-28 00:00:00 | pamela  |    4 | NULL | edwar |
| 2011-06-28 00:00:00 | pamela  |    5 |    2 | angel |
| 2011-06-28 00:00:00 | pamela  |    5 |    2 | angel |
| 2011-06-28 00:00:00 | rolando |    4 |    2 | angel |
| 2011-06-28 00:00:00 | rolando |    4 |    2 | angel |
| 2011-06-28 00:00:00 | rolando |    4 | NULL | edwar |
| 2011-06-28 00:00:00 | rolando |    4 | NULL | angel |
+---------------------+---------+------+------+-------+
22 rows in set (0.00 sec)

mysql> SELECT
    ->     COUNT(datefield) datefield_count,
    ->     LEFT(all_fields,10) datefield,
    ->     SUBSTR(all_fields,11) all_other_fields
    -> FROM
    -> (
    ->     SELECT
    ->         DATE(datefield) datefield,
    ->         CONCAT(DATE(datefield),'|',
    ->         COALESCE(col1,'< NULL >'),'|',
    ->         COALESCE(col2,'< NULL >'),'|',
    ->         COALESCE(col3,'< NULL >'),'|',
    ->         COALESCE(col4,'< NULL >'),'|') all_fields
    ->     FROM
    ->          yourtable
    -> ) A
    -> GROUP BY all_fields;
+-----------------+------------+----------------------------+
| datefield_count | datefield  | all_other_fields           |
+-----------------+------------+----------------------------+
|               1 | 2011-06-28 | |pamela|4|2|angel|         |
|               1 | 2011-06-28 | |pamela|4|< NULL >|edwar|  |
|               2 | 2011-06-28 | |pamela|5|2|angel|         |
|               3 | 2011-06-28 | |rolando|4|2|angel|        |
|               1 | 2011-06-28 | |rolando|4|< NULL >|angel| |
|               1 | 2011-06-28 | |rolando|4|< NULL >|edwar| |
|               2 | 2011-06-28 | |rolando|5|2|angel|        |
|               4 | 2011-06-29 | |rolando|4|2|angel|        |
|               1 | 2011-06-29 | |rolando|4|< NULL >|angel| |
|               1 | 2011-06-29 | |rolando|4|< NULL >|edwar| |
|               3 | 2011-06-30 | |rolando|4|3|angel|        |
|               2 | 2011-06-30 | |rolando|4|< NULL >|angel| |
+-----------------+------------+----------------------------+
12 rows in set (0.00 sec)

mysql>

I'll leave it to you imaginative creativity to loop through this and print

  • datefield
  • datefield_count
  • print all_other_fields 'datefield_count' times

Give it a Try !!!

1
votes

This is not what the OP have asked, rather what I have been looking for when I came upon this question. Perhaps some will find it useful.

select *
  from thetable
    join (
      select thedate, count( thedate ) as cnt
        from thetable
        group by thedate
    ) as counts
    using( thedate )
  order by thedate

The above query will select everything with an extra field cnt containing the number of records having the same date. Then it is trivial to print something like:

some date, 2 records for this date
col1, col2, col3, col4
col1, col2, col3, col4

some other date, 3 records for this date
col1, col2, col3, col4
col1, col2, col3, col4
col1, col2, col3, col4

yet some other date, 1 record for this date
col1, col2, col3, col4

0
votes
SELECT ...
FROM yourtable
GROUP BY DATE(datefield)
ORDER BY COUNT(DATE(datefield)) DESC

note that I'm using the DATE() function, in case your date fields are actually date-time. If you group on date time, it'd group by the fully yyyy-mm-dd hh:mm:ss, not just the yyyy-mm-dd and you'd get totally different results.

That'll get you the core results. Doing the output as you want it will need some post-processing in your script, but isn't too hard. Just buffer the found rows until the date changes, then output the buffer with a row count.

0
votes
SELECT *, COUNT(thedate) daycount 
FROM `table` 
GROUP BY thedate
ORDER BY thedate DESC
0
votes
SELECT thedate, COUNT(id)
FROM table
WHERE 1
GROUP BY thedate
ORDER BY thedate