我有一个像这样的表（交易）
ID | Name | Unit |Quantity_in|Quantity_out|transaction_date|
-----|-------| --------|-----------|------------|----------------|
100 | aaaa | packet | 100 | 0 | 02/07/2017 |
99 | cccc | packet | 70 | 0 | 20/06/2017 |
99 | cccc | packet | 0 | 20 | 22/07/2017 |
100 | aaaa | strip | 0 | 30 | 05/07/2017 |
102 | bbbb | packet | 50 | 0 | 29/06/2017 |
我想进行查询以获取两天之间的交易和当前余额，同时在WHERE条件中的日期之前获得先前的余额，因此我进行以下查询
从事务tr中选择d.ID，d.Name，d。单位，（选择sum（tr.Quantity_in） - sum（tr.Quantity_out），其中tr.transaction_date＆lt; 01/07 / 2017年和d.ID = tr.ID）作为Prev_bal，sum（d.Quantity_in），sum（d.Quantity_out），Prev_bal +（sum（d.Quantity_in） - sum（d.Quantity_out））as current_balance来自交易d 在2017年7月1日至2017年7月30日期间d.transaction_date 按d.ID分组，d.Name，d。单位
但查询下面的结果不包含ID 102，因为Where条件中的日期之间没有交易
ID | Name | Unit | Prev_bal | Quantity_in | Quantity_out |Current_balance|
----|------| ------ |----------|-------------|---------------|---------------|
100 |aaaa |packet | 0 | 100 | 30 | 70
----|----- |--------|----------|-------------|---------------|---------------|
99 |cccc |packet | 70 | 0 | 20 | 50 |
任何人都可以帮助我进行查询，即使WHERE条件中的日期之间没有事务处理（如表中的ID 102），也会获得所有具有余额的ID的结果。 我需要获得当前余额（Current_balance），其中2个日期之间的transaction_date和之前的余额（Prev_bal）用于具有balance_date＆lt;的余额的ID。在哪里条件的日期 在此先感谢您的帮助。 结果应该像下面那样
102 | bbbb |包| 50 | 0 | 0 | 50 |
答案 0 :(得分：0)
所以，这就是我的想法。
无论您提供的日期范围如何，您仍希望每个ID都包含在结果集中。因此无需过滤掉行。所以问题是现在只是在一天结束时找出剩余的余额。因此，对于日期范围内的项目，余额是当前余额，对于日期范围之前的项目，它是之前的余额（基本上是最后一个当前余额）。 以下是使用Oracle中的分析函数的查询。
WITH dat
AS (SELECT 100 AS id,
'aaaa' AS name,
'packet' AS unit,
100 AS quantity_in,
0 AS quantity_out,
TO_DATE ('02/07/2017', 'DD/MM/YYYY') AS transaction_date
FROM DUAL
UNION
SELECT 99 AS id,
'cccc' AS name,
'packet' AS unit,
70 AS quantity_in,
0 AS quantity_out,
TO_DATE ('20/06/2017', 'DD/MM/YYYY') AS transaction_date
FROM DUAL
UNION
SELECT 99 AS id,
'cccc' AS name,
'packet' AS unit,
0 AS quantity_in,
20 AS quantity_out,
TO_DATE ('22/07/2017', 'DD/MM/YYYY') AS transaction_date
FROM DUAL
UNION
SELECT 100 AS id,
'aaaa' AS name,
'strip' AS unit,
0 AS quantity_in,
30 AS quantity_out,
TO_DATE ('05/07/2017', 'DD/MM/YYYY') AS transaction_date
FROM DUAL
UNION
SELECT 102 AS id,
'bbbb' AS name,
'packet' AS unit,
50 AS quantity_in,
0 AS quantity_out,
TO_DATE ('29/06/2017', 'DD/MM/YYYY') AS transaction_date
FROM DUAL)
SELECT id,
name,
unit,
quantity_in,
quantity_out,
transaction_date,
remaining_balance
FROM (SELECT id,
name,
unit,
quantity_in,
quantity_out,
balance,
transaction_date,
SUM (balance)
OVER (PARTITION BY id ORDER BY transaction_date ROWS UNBOUNDED PRECEDING)
AS remaining_balance,
RANK () OVER (PARTITION BY id ORDER BY transaction_date DESC) AS rn
FROM (SELECT id,
name,
unit,
quantity_in,
quantity_out,
quantity_in
- quantity_out
AS balance,
transaction_date
FROM dat))
WHERE rn = 1;
它产生了这个结果。
ID |NAME|UNIT | QUANTITY_IN|QUANTITY_OUT|TRANSACTION_DATE|REMAINING_BALANCE
99 |cccc|packet | 0| 20| 7/22/2017| 50
100|aaaa|strip | 0| 30| 7/5/2017| 70
102|bbbb|packet | 50| 0| 6/29/2017| 50