Question

我使用以下示例行的SQLite数据库，列为{id，temperature，description，time_stamp：

1   45  Clear   2016-01-13 10:24:17
2   45  Clear   2016-01-13 10:24:41
3   45  Clear   2016-01-13 10:24:41
4   45  Rain    2016-01-13 10:24:41
5   46  Clear   2016-01-13 10:38:29
6   46  Clear   2016-01-13 10:38:53
7   46  Clear   2016-01-13 10:39:08
8   46  Clear   2016-01-13 10:39:08

我对它们运行此查询，按时间间隔和天气描述对它们进行分组：

SELECT AVG(current_temperatures) AS temp_avg,  
            CASE WHEN  strftime('%M',  time_stamp) < '30' 
            THEN strftime('%H', time_stamp)  
            ELSE strftime('%H', time_stamp, '+1 hours') END as hour, 
            current_weather_description, 
            count(*) as counter
            FROM weather_events
            GROUP BY strftime('%H',  time_stamp, '+30 minutes'), current_weather_description
            order by hour desc

使用以下结果{row_num，temp_avg，hour，current_weather_description，counter}：

"46.0"  "11"    "Clear" "4"
"45.0"  "10"    "Clear" "3"
"45.0"  "10"    "Rain"  "1"

我的问题是如何通过最大计数器子查询每个小时和组。所以最终我想得到结果：

"46.0"  "11"    "Clear" "4"
"45.0"  "10"    "Clear" "3"

我是SQL和SQLite的新手。所有数据都来自同一个表。此外，由于查询的温度是平均值，它如何选择我要求它选择的其余列？例如，如果您同时选择time_stamp，您将看到它是数据库中的特定time_stamp。是否随机选择要选择的列？

Answer 1

您的sql查询输出正确。您已按half hour time intervals和weather description对行进行了分组。 10:00 time interval有两种不同的天气：＆＃34;清除＆＃34;和＆＃34; Rain＆＃34;和10:30 time interval的一个天气：＆＃34;清除＆＃34; 。如果您要从GROUP BY子句中删除weather_description列，则只能获得两行。

使用common table expressions的解决方案，仅显示在时间间隔内最常出现的天气描述。

;WITH temp_data AS(
    SELECT AVG(current_temperatures) AS temp_avg,  
            CASE WHEN  strftime('%M',  time_stamp) < '30' 
            THEN strftime('%H', time_stamp)  
            ELSE strftime('%H', time_stamp, '+1 hours') END as hour, 
            current_weather_description,
            count(*) as counter
    FROM weather_events
    GROUP BY strftime('%H',  time_stamp, '+30 minutes'),
             current_weather_description
)

SELECT T1.* FROM temp_data AS T1
WHERE T1.counter = (SELECT MAX(counter) FROM temp_data AS T2
                    WHERE T2.hour = T1.hour)

SQLite：子查询按最大计数分组

1 个答案: