SQL：如何使用带 case 表达式的外部别名进行子查询

Question

如何创建一个使用 case 表达式的子查询的外列值的子查询？

SQL 示例：

SELECT
    name,
    average,
    crnt,
    goodFactor
FROM
(
 "Crispy Taco" AS name,
  ( SELECT avg(x)
    FROM (
     SELECT COUNT() x
     FROM tacos
     WHERE orderDate > DATEADD(DAY,  -1, GETDATE()))   <--- THE AVERAGE PRIOR TO TODAY
  ) AS average,
  COUNT(*) AS crnt,
  (
    SELECT
    CASE WHEN crnt < average    <---- PROBLEM HERE : Missing columns 'crnt' and 'average'
     THEN "not good"
     ELSE "pretty good"
    END
  ) AS goodFactor
)
from tacos

问题：

'

crnt

' 和 '

average

' 是 'goodFactor' 子查询的 '

missing columns

'

样本数据：（数据库条目）

+-------------+---------+---------+
|    Name     | Ordered |  Date   |
+-------------+---------+---------+
| Crispy Taco |       1 | 8/29/24 |
| Crispy Taco |       1 | 8/29/24 |
| Crispy Taco |       1 | 8/29/24 |
| Crispy Taco |       1 | 8/29/24 |
| Crispy Taco |       1 | 8/29/24 |
| Crispy Taco |       1 | 8/30/24 |
| Crispy Taco |       1 | 8/30/24 |
| Crispy Taco |       1 | 8/30/24 |
| Crispy Taco |       1 | 8/30/24 |
| Crispy Taco |       1 | 8/30/24 |
| Crispy Taco |      12 | 8/31/24 | <---- ASSUME QUERY INVOKED THIS DAY PRODUCING THE TABLE BELOW
+-------------+---------+---------+

预期结果：

如果“

crnt

”值大于计算的“

average

”
然后返回“

相当好”，否则返回“不好”

预期结果

+-------------+---------------+---------+--------------------+
|    Name     | Daily Average | Current | GoodFactor         |
+-------------+---------------+---------+--------------------+
| Crispy Taco |       5       |      12 | pretty Good        |
+-------------+---------------+---------+--------------------+

Answer 1

这应该适用于所有标准 DBMS - 今天，即 2024 年 8 月 31 日，否则您 必须将所有地方的

CURRENT_DATE

 替换为 2024 年 8 月 31 日，并使用该样本数据：

WITH                                                                                                                                                                           
-- your input , don't use in final query
indata(name,ordered,date) AS (
          SELECT 'Crispy Taco', 1,DATE '08/29/24'
UNION ALL SELECT 'Crispy Taco', 1,DATE '08/29/24'
UNION ALL SELECT 'Crispy Taco', 1,DATE '08/29/24'
UNION ALL SELECT 'Crispy Taco', 1,DATE '08/29/24'
UNION ALL SELECT 'Crispy Taco', 1,DATE '08/29/24'
UNION ALL SELECT 'Crispy Taco', 1,DATE '08/30/24'
UNION ALL SELECT 'Crispy Taco', 1,DATE '08/30/24'
UNION ALL SELECT 'Crispy Taco', 1,DATE '08/30/24'
UNION ALL SELECT 'Crispy Taco', 1,DATE '08/30/24'
UNION ALL SELECT 'Crispy Taco', 1,DATE '08/30/24'
UNION ALL SELECT 'Crispy Taco',12,DATE '08/31/24'
)
-- real query starts here, replace following comma with "WITH"
,
-- need a query grouping by day to get the daily sums
-- before today
daily AS (
  SELECT 
    name
  , date
  , SUM(ordered) AS per_day
  FROM indata
  WHERE date < CURRENT_DATE
  GROUP BY  
    name
  , date
)
,
-- and one summing the current day
current AS (
  SELECT
    name
  , date
  , SUM(ordered) AS current
  FROM indata
  WHERE date=CURRENT_DATE
  GROUP BY
    name
  , date
)
SELECT
  daily.name
, AVG(per_day) AS "Daily Average"
, MAX(current) AS current
, CASE WHEN MAX(current) < AVG(per_day)
    THEN 'not good'
    ELSE 'pretty good'
  END AS good_factor
FROM daily
JOIN current ON daily.name = current.name
GROUP BY  
  daily.name;

名字每日平均当前好因素脆玉米饼512还不错