MySQL GROUP BY每组返回多行

Question

Given a sample table:

| id | locationID |      Date_Time      | temp |
|----|------------|---------------------|------|
| 1  | L001       | 2018-09-04 11:25:00 | 52.6 |
| 2  | L002       | 2018-09-04 11:35:00 | 66.1 |
| 3  | L003       | 2018-09-04 03:30:00 | 41.2 |
| 4  | L003       | 2018-09-05 10:22:00 | 71.8 |
| 5  | L003       | 2018-09-06 14:21:00 | 63.4 |
| 6  | L003       | 2018-09-06 18:18:00 | 50.1 |

I would like to return the latest N number of records for each group as below:

Expected output:

| id | locationID |      Date_Time      | temp |
|----|------------|---------------------|------|
| 1  | L001       | 2018-09-04 11:25:00 | 52.6 |
| 2  | L002       | 2018-09-04 11:35:00 | 66.1 |
| 4  | L003       | 2018-09-05 10:22:00 | 71.8 |
| 5  | L003       | 2018-09-06 14:21:00 | 63.4 |
| 6  | L003       | 2018-09-06 18:18:00 | 50.1 |

I have this query but it only returns the latest row for each group? I would like to return more than one row (N number of rows) for each group?

SELECT *
FROM HealthStatus
WHERE Date_Time IN (
    SELECT MAX(Date_Time)
    FROM HealthStatus
    GROUP BY LocationID
)

Would really appreciate some help on how I can achieve my desired output. Thanks in advance.

Answer 1

is there any way that I can get the latest N rows for each group?

Prior to the availability of row_number() you can use variables to mimic that function

SELECT
*
FROM (
    SELECT
          @row_num :=IF(@prev_value=locationID,@row_num+1,1)AS RowNumber
        , id , locationID , Date_Time,temp
        , @prev_value := locationID
    FROM HealthStatus
    CROSS JOIN (SELECT @row_num :=1,  @prev_value :='') vars
    ORDER BY
          locationID , Date_Time DESC
    ) derived
WHERE RowNumber < 4

Answer 2

If you upgraded to MySQL 8.0, then this would solve it:

SELECT * FROM (
  SELECT *, ROW_NUMBER() OVER (PARTITION BY locationID ORDER BY Date_Time DESC) AS r
  FROM HealthStatus
) T
WHERE r <= 3;

On 5.7... I can't think of a way. :(