四个星期的数据 - 分组答案

【问题标题】：four Weeks data - group by四个星期的数据 - 分组
【发布时间】：2015-07-05 23:26:28
【问题描述】：

我有以下表格及其各自的数据：

FORECAST:
- ITEM
- FORECAST_WEEK_DATE
- FORECAST

SALES:
- ITEM
- SALES_WEEK_DATE
- SALES

WEEK_DATE 列包含每个周末日期的记录，例如： 2015 年 7 月 2 日 2015 年 2 月 15 日等等。

我想从两个表中获取结果，以便结果如下：

ITEM FORECAST_TOTAL SALES_TOTAL
001       23             45

最近 4 周的总数据。

我如何做到这一点？

【问题讨论】：

标签： sql oracle date group-by

【解决方案1】：

SQL Fiddle

Oracle 11g R2 架构设置：

CREATE TABLE FORECAST ( ITEM, FORECAST_WEEK_DATE, FORECAST) AS
          SELECT 1, DATE '2015-01-01',                     1 FROM DUAL
UNION ALL SELECT 1, DATE '2015-01-01' + INTERVAL  '7' DAY, 2 FROM DUAL
UNION ALL SELECT 1, DATE '2015-01-01' + INTERVAL '14' DAY, 3 FROM DUAL
UNION ALL SELECT 1, DATE '2015-01-01' + INTERVAL '21' DAY, 4 FROM DUAL
UNION ALL SELECT 1, DATE '2015-01-01' + INTERVAL '28' DAY, 5 FROM DUAL
UNION ALL SELECT 1, DATE '2015-01-01' + INTERVAL '35' DAY, 6 FROM DUAL
UNION ALL SELECT 1, DATE '2015-01-01' + INTERVAL '42' DAY, 7 FROM DUAL
UNION ALL SELECT 2, DATE '2015-01-01',                     3 FROM DUAL
UNION ALL SELECT 2, DATE '2015-01-01' + INTERVAL  '7' DAY, 1 FROM DUAL
UNION ALL SELECT 2, DATE '2015-01-01' + INTERVAL '14' DAY, 5 FROM DUAL
UNION ALL SELECT 2, DATE '2015-01-01' + INTERVAL '21' DAY, 8 FROM DUAL
UNION ALL SELECT 2, DATE '2015-01-01' + INTERVAL '28' DAY, 3 FROM DUAL
UNION ALL SELECT 2, DATE '2015-01-01' + INTERVAL '35' DAY, 2 FROM DUAL;

CREATE TABLE SALES (ITEM, SALES_WEEK_DATE, SALES) AS
          SELECT 1, DATE '2015-01-01',                     3 FROM DUAL
UNION ALL SELECT 1, DATE '2015-01-01' + INTERVAL  '7' DAY, 2 FROM DUAL
UNION ALL SELECT 1, DATE '2015-01-01' + INTERVAL '14' DAY, 1 FROM DUAL
UNION ALL SELECT 1, DATE '2015-01-01' + INTERVAL '21' DAY, 5 FROM DUAL
UNION ALL SELECT 1, DATE '2015-01-01' + INTERVAL '28' DAY, 3 FROM DUAL
UNION ALL SELECT 1, DATE '2015-01-01' + INTERVAL '35' DAY, 8 FROM DUAL
UNION ALL SELECT 1, DATE '2015-01-01' + INTERVAL '42' DAY, 11 FROM DUAL
UNION ALL SELECT 2, DATE '2015-01-01',                     7 FROM DUAL
UNION ALL SELECT 2, DATE '2015-01-01' + INTERVAL  '7' DAY, 1 FROM DUAL
UNION ALL SELECT 2, DATE '2015-01-01' + INTERVAL '14' DAY, 4 FROM DUAL
UNION ALL SELECT 2, DATE '2015-01-01' + INTERVAL '21' DAY, 3 FROM DUAL
UNION ALL SELECT 2, DATE '2015-01-01' + INTERVAL '28' DAY, 6 FROM DUAL
UNION ALL SELECT 2, DATE '2015-01-01' + INTERVAL '35' DAY, 8 FROM DUAL;

查询 1：

这将在表格中为每个项目获取最近 4 周的数据：

WITH ordered_week_data AS (
SELECT f.ITEM,
       ROW_NUMBER() OVER ( PARTITION BY f.ITEM ORDER BY FORECAST_WEEK_DATE DESC ) AS RN,
       f.FORECAST_WEEK_DATE AS WEEK_DATE,
       FORECAST,
       SALES
FROM   FORECAST f
       INNER JOIN
       SALES s
       ON ( f.ITEM = s.ITEM AND f.FORECAST_WEEK_DATE = s.SALES_WEEK_DATE )
)
SELECT ITEM,
       MAX( WEEK_DATE ) AS LAST_WEEK_DATE,
       SUM( FORECAST ) AS FORECAST_TOTAL,
       SUM( SALES ) AS SALES_TOTAL
FROM   ordered_week_data
WHERE  RN <= 4
GROUP BY
       ITEM

Results：

| ITEM |             LAST_WEEK_DATE | FORECAST_TOTAL | SALES_TOTAL |
|------|----------------------------|----------------|-------------|
|    1 | February, 12 2015 00:00:00 |             22 |          27 |
|    2 | February, 05 2015 00:00:00 |             18 |          21 |

【讨论】：