计数日期范围内的不同，运行 grass_date

Question

我有一个 table 这样的：

日期	名字
1/1	一个
1/2	b
1/3	c
1/2	一个
1/3	d

并且需要计算每行中日期后 7 天内可用的不同名称。结果应该是：

日期	计数	解释
1/1	1	一个
1/2	2	a,b
1/3	4	a,b,c,d

我试过这些 count+if、count+case when、滞后函数：

COUNT(DISTINCT(IF(date_1 在日期之间 - INTERVAL '7' DAY 和日期，名称)))
lag(count(distinct name), 7) 超过（按日期排序）

没有任何效果，结果是只计算每天的名字。

请指教，万分感谢！

Answer 1

嗯。 . .这是否符合您的要求？

select min_date, 
       sum(count(*)) over (order by date) as cnt,
       array_agg(array_agg(name)) over (order by date) as names
from (select name, min(date) as min_date
      from t
      group by name
     ) t
group by min_date;

编辑：

假设每个日期一行，添加一个 window 框架规范：

select min_date, 
       sum(count(*)) over (order by date rows between 6 preceding and current row) as cnt,
       array_agg(array_agg(name)) over (order by date rows between 6 preceding and current row) as names
from (select name, min(date) as min_date
      from t
      group by name
     ) t
group by min_date

计数日期范围内的不同，运行 grass_date

Count distinct in a date range, running grass_date

sql

presto