Question

我有一个查询，按照our_price升序按以下结构顺序返回结果

/^[MLVH]/.test(myString)

输出

SELECT our_price,
       mrp,
       marketplace_id,
       test_id
FROM   marketplace_test_mapping mtm
       INNER JOIN marketplace m
               ON m.id = mtm.marketplace_id
WHERE  m.city_id = 1
       AND mtm.test_id IN ( 36, 23, 43, 107,
                            121, 82 )
ORDER  BY our_price ASC;

我想获取每个test_id的最低价格，但是当我对our_price | mrp | marketplace_id | test_id ----------+-----+----------------+-------- 50 90 3 23 51 70 2 23 52 88 1 23 53 80 3 24 54 90 2 24 55 90 4 23 56 90 1 25 57 90 2 25 58 90 1 24进行group_by时，它也会对test_id进行排序并返回以下内容

marketplace_id

实际输出

SELECT *
FROM   (SELECT our_price,
               mrp,
               marketplace_id,
               test_id
        FROM   marketplace_test_mapping mtm
               INNER JOIN marketplace m
                       ON m.id = mtm.marketplace_id
        WHERE  m.city_id = 1
               AND mtm.test_id IN ( 36, 23, 43, 107,
                                    121, 82 )
        ORDER  BY our_price ASC) AS temp_marketplace_test_mapping
GROUP  BY test_id;

预期产量

our_price | mrp | marketplace_id | test_id
----------+-----+----------------+--------
52         88     1                23
58         90     1                24
56         90     1                25

在our_price | mrp | marketplace_id | test_id ----------+-----+----------------+-------- 50 90 3 23 53 90 3 24 56 90 1 25上进行分组时，两个结果的不同之处在于actual output也在对marketplace_id进行排序。

Answer 1

“我想获取每个test_id的最低价格”

select test_id, min(our_price) as our_price_min -- changed
from marketplace_test_mapping  mtm
inner join marketplace m on m.id=mtm.marketplace_id
where m.city_id=1 and mtm.test_id in (36,23,43,107,121,82)
group by test_id -- added
order by our_price asc;

Answer 2

WITH cte (our_price,mrp,marketplace_id,test_id) AS
(SELECT our_price,mrp,marketplace_id,test_id
FROM   marketplace_test_mapping mtm
       INNER JOIN marketplace m
       ON m.id = mtm.marketplace_id
WHERE  m.city_id = 1 AND mtm.test_id IN ( 36, 23, 43, 107,121, 82 )
)
select t1.our_price, t1.mrp, t1.marketplace_id, t1.test_id from cte t1 
join (select test_id, min(our_price) our_price from cte group by test_id) t2
on t1.test_id = t2.test_id and t1.our_price = t2.our_price;

Answer 3

找到每个test_id的最低价格，然后获取MARKETPLACE_ID和MRP。

with transactions as (
  -- get the transactions list, execute filters
  SELECT test_id, marketplace_id, mrp, our_price       
  FROM   marketplace_test_mapping mtm
  INNER JOIN marketplace m ON m.id = mtm.marketplace_id
  WHERE  m.city_id = 1
  AND mtm.test_id IN (23,36,43,82,107,121) --it's a good practice to sorta data in IN clause
)

-- if there are few marketplaces with minimum price on each test_id
-- then each of then will be shown
select a.*, b.marketplace_id, b.mrp
from (
  -- minimum price for each test_id
  select test_id, min(our_price) as our_price_min
  from transactions
  group by test_id
) a
left join transactions b on a.test_id = b.test_id and a.our_price_min = b.our_price
order by a.test_id, b.marketplace_id

通过test_id选择最小的our_price组

3 个答案: