我想要 table 的最新日期和价格，在大查询中该 ID 有多个不同日期的价格。下面是我的查询

Question

我面临的问题是我不想按价格列分组。我只想按 id 分组，这样我就可以获得最新的日期和价格。但是大查询不允许我运行这个查询没有按价格分组。

这是我的查询：

SELECT ID, 最大（创建日期）作为最新日期，价格来自 table name 按id分组

Answer 1

我有一个解决办法给你。它不是最干净的，但它可以工作。

SELECT id, "created date" as "latest date", price 
FROM (SELECT id, 
FIRST_VALUE("created date") OVER (PARTITION BY id ORDER BY "created date" DESC) as "created date", 
FIRST_VALUE("price") OVER (PARTITION BY id ORDER BY "created date" DESC) as "price" 
FROM "table name")
GROUP BY id

我觉得这个应该也可以

SELECT UNIQUE id, 
FIRST_VALUE("created date") OVER (PARTITION BY id ORDER BY "created date" DESC) as "created date", 
FIRST_VALUE("price") OVER (PARTITION BY id ORDER BY "created date" DESC) as "price" 
FROM "table name"

Answer 2

使用以下方法

select as value 
  array_agg(t order by created_date desc limit 1)[offset(0)] 
from your_table t
group by id

Answer 3

在这种情况下，我使用了聚合函数ARRAY_AGG，它returns 一个按日期排序的数组，每个ID 只获取一条记录。还，我使用 OFFSET 命令从零位置读取数组。

WITH Example AS
(SELECT 1 as ID, 32.5 as price, PARSE_DATE("%x", "12/25/22")  as sdate UNION ALL
SELECT 1 as ID, 3 as price,  PARSE_DATE("%x",'01/13/22') as sdate UNION ALL
SELECT 2 as ID, 2.5 as price,  PARSE_DATE("%x",'01/30/22')  as sdate UNION ALL
SELECT 3 as ID, 5 as price,  PARSE_DATE("%x",'01/20/22')  as sdate UNION ALL
SELECT 4 as ID, 50 as price,  PARSE_DATE("%x",'01/10/22') as sdate UNION ALL
SELECT 4 as ID, 1.13 as price,  PARSE_DATE("%x",'01/08/22') as sdate UNION ALL
SELECT 5 as ID, 150 as price,  PARSE_DATE("%x",'01/05/22') as sdate UNION ALL
SELECT 6 as ID, 30 as price,  PARSE_DATE("%x",'01/05/22') as sdate UNION ALL
SELECT 7 as ID, 5.2 as price,  PARSE_DATE("%x",'01/03/22') as sdate UNION ALL
SELECT 8 as ID, 200 as price,  PARSE_DATE("%x",'02/01/22') as sdate
)
SELECT AS VALUE ARRAY_AGG(table ORDER BY sdate DESC LIMIT 1)[OFFSET(0)]
FROM Example table
group by id

我想要 table 的最新日期和价格，在大查询中该 ID 有多个不同日期的价格。下面是我的查询

I want the latest date and price from the table which has multiple prices with different date on that ID in big query. here is my query below

google-bigquery