Question

我有一个运行良好的代码-仅按电话使用日期

。

select  date,
        data,
        SUM(data) OVER (ORDER BY date asc
                            ROWS between 30 PRECEDING and current row) data_30,
        texts,
        SUM(tests) OVER (ORDER BY date asc
                ROWS between 30 PRECEDING and current row) text_30,
        voice,
        SUM(voice) OVER (ORDER BY date asc
                ROWS between 30 PRECEDING and current row) voice_30,
        wifi,
        SUM(wifi) 
                OVER (ORDER BY date asc
                ROWS between 30 PRECEDING and current row) wifi_30
    FROM Table

我只是想知道如何使用前30天的总和，但是是否有可能再加上第二个变量，例如，我想查看日期，并按费率计划使用这些用法？

类似

select  date,
        plan, b, c, d, 
        data,
        SUM(data) OVER (ORDER BY date asc
                            ROWS between 30 PRECEDING and current row) data_30,
        texts,
        SUM(tests) OVER (ORDER BY date asc
                ROWS between 30 PRECEDING and current row) text_30,
        voice,
        SUM(voice) OVER (ORDER BY date asc
                ROWS between 30 PRECEDING and current row) voice_30,
        wifi,
        SUM(wifi) 
                OVER (ORDER BY date asc
                ROWS between 30 PRECEDING and current row) wifi_30
    FROM Table
    group by date, plan, b, c, d

公正日期的结果

date  usage  last30sum
1/1   1       1
1/2   1       2
....
1/20  1       30

如果我的源数据为

date line     rateplan        usage  
1/1  phone1   10gbplan        1
1/1  phone2   unlimited       2                
1/2  phone3   10gbplan        1                
....
1/30 phone200 10gbplan        1

我想看看整理

date plan        totalusage   rolling_30
1/2  10gbplan   1            4+ sum(28 days before 1/2)

能否按日期将其分组，以获取最近的30天，即a，b，c仅表示其他分组依据，可能类似于设备型号或区域。

Answer 1

我通过添加分区来弄清楚

select  date,
        plan, b, c, d, 
        data,
        SUM(data) OVER (partition by plan, b, c, d
                        ORDER BY date asc
                            ROWS between 30 PRECEDING and current row) data_30,
        texts,
        SUM(tests) OVER (partition by plan, b, c, d
                ORDER BY date asc
                ROWS between 30 PRECEDING and current row) text_30,
        voice,
        SUM(voice) OVER (partition by plan, b, c, d
                ORDER BY date asc
                ROWS between 30 PRECEDING and current row) voice_30,
        wifi,
        SUM(wifi) 
                OVER (partition by plan, b, c, d
                ORDER BY date asc
                ROWS between 30 PRECEDING and current row) wifi_30
    FROM Table

总和（）与分组依据和前置

1 个答案: