我要求在条件下扣除产品数量。它似乎有点复杂，不知道如何使用SQL查询。以下是它的概念：产品在这里意味着原材料。出于生产目的，我们必须从库存中扣除原材料。几条规则可以遵循：
表格 - ProductEntry：
i）使用PO（采购订单）和供应商的发票编号购买产品。在这种情况下有一个条件。假设已经购买了100个产品ID为1001的产品，它分为以下两个部分：
Id - ProductId - PO - Invoice no - Quantity - Price - EntryDate
1st section: 1 - 1001 - PO-102 - Inv-122 - 20 - 200 - 2017-07-10 10:00:00
2nd section: 2 - 1001 - PO-102 - Inv-122 - 80 - 800 - 2017-07-10 11:00:00
3 - 1002 - PO-102 - Inv-122 - 20 - 400 - 2017-07-10 10:00:00
这里开始游戏。在许多情况下，原材料或产品可能分为多个部分或完全一次，我的意思是总共100件。
ii）现在购买之后，它必须进入商店，还有另一个程序。每个购买的产品应分别收到IP（进口许可证）号码，如下所示：
表格 - IpEntry：
Id - ProductId - Invoice no - IP - AnotherProductId
1 - 1001 - Inv-122 - IP2244 - 2
2 - 1001 - Inv-122 - IP2244 - 2
3 - 1002 - Inv-122 - IP2244 - 4
iii）收到产品后，应将其用于生产目的，即消费。在消费中，应使用首次输入的产品或原材料。这意味着，如果必须扣除产品ID 1001，则应根据 EntryDate ＆＃39; EntryDate ＆＃39;因为它已进入最低点。时间是同一天的10:00:00。因此，对于扣除或消费，应该遵循：
表 - 消费：
Id - 消费否 - AnotherProductId - 数量
1 - Con-122 - 2 - 10
3 - Con-122 - 4 - 10
所以最终输出如下：
Id - AnotherProductId - Stock - Quantity Used - Remaining Balance
1 - 2 - 10 - 10 - 100
2 - 4 - 10 - 10 - 200
我不在这里分享sql查询，因为使用返回以下内容的INNER JOIN和MIN函数应该不够准确和简单：
Id - AnotherProductId - Stock - Quantity Used - Remaining Balance
1 - 2 - 10 - 10 - 100
2 - 2 - 10 - 10 - 100 //It returns **AnotherProductId or ProductId - 1001 or 2** twice as it should only return once
3 - 4 - 10 - 10 - 200
我不知道如何处理上述情况，特别是同一产品，数量不同而且有点混淆。
以下是更好理解的脚本：
USE [Demo]
GO
/****** Object: Table [dbo].[ProductEntry] Script Date: 07/19/2017 20:37:41 ******/
SET ANSI_NULLS ON
GO
SET QUOTED_IDENTIFIER ON
GO
CREATE TABLE [dbo].[ProductEntry](
[Id] [int] IDENTITY(1,1) NOT NULL,
[ProductId] [int] NULL,
[PO] [nvarchar](60) NULL,
[Invoice No] [nvarchar](60) NULL,
[Quantity] [float] NULL,
[Price] [float] NULL,
[EntryDate] [datetime] NULL,
CONSTRAINT [PK_ProductEntry] PRIMARY KEY CLUSTERED
(
[Id] ASC
)WITH (PAD_INDEX = OFF, STATISTICS_NORECOMPUTE = OFF, IGNORE_DUP_KEY = OFF, ALLOW_ROW_LOCKS = ON, ALLOW_PAGE_LOCKS = ON) ON [PRIMARY]
) ON [PRIMARY]
GO
SET IDENTITY_INSERT [dbo].[ProductEntry] ON
INSERT [dbo].[ProductEntry] ([Id], [ProductId], [PO], [Invoice No], [Quantity], [Price], [EntryDate]) VALUES (1, 1001, N'PO-102', N'Inv-122', 20, 200, CAST(0x0000A7AC00A4CB80 AS DateTime))
INSERT [dbo].[ProductEntry] ([Id], [ProductId], [PO], [Invoice No], [Quantity], [Price], [EntryDate]) VALUES (2, 1001, N'PO-102', N'Inv-122', 80, 800, CAST(0x0000A7AC00B54640 AS DateTime))
INSERT [dbo].[ProductEntry] ([Id], [ProductId], [PO], [Invoice No], [Quantity], [Price], [EntryDate]) VALUES (3, 1002, N'PO-102', N'Inv-122', 20, 400, CAST(0x0000A7AC00A4CB80 AS DateTime))
SET IDENTITY_INSERT [dbo].[ProductEntry] OFF
/****** Object: Table [dbo].[IpEntry] Script Date: 07/19/2017 20:37:41 ******/
SET ANSI_NULLS ON
GO
SET QUOTED_IDENTIFIER ON
GO
CREATE TABLE [dbo].[IpEntry](
[Id] [int] IDENTITY(1,1) NOT NULL,
[ProductId] [int] NULL,
[Invoice No] [nvarchar](60) NULL,
[IP] [nvarchar](60) NULL,
[AnotherProductId] [int] NULL,
CONSTRAINT [PK_IpEntry] PRIMARY KEY CLUSTERED
(
[Id] ASC
)WITH (PAD_INDEX = OFF, STATISTICS_NORECOMPUTE = OFF, IGNORE_DUP_KEY = OFF, ALLOW_ROW_LOCKS = ON, ALLOW_PAGE_LOCKS = ON) ON [PRIMARY]
) ON [PRIMARY]
GO
SET IDENTITY_INSERT [dbo].[IpEntry] ON
INSERT [dbo].[IpEntry] ([Id], [ProductId], [Invoice No], [IP], [AnotherProductId]) VALUES (1, 1001, N'Inv-122', N'IP2244', 2)
INSERT [dbo].[IpEntry] ([Id], [ProductId], [Invoice No], [IP], [AnotherProductId]) VALUES (2, 1001, N'Inv-122', N'IP2244', 2)
INSERT [dbo].[IpEntry] ([Id], [ProductId], [Invoice No], [IP], [AnotherProductId]) VALUES (3, 1002, N'Inv-122', N'IP2244', 4)
SET IDENTITY_INSERT [dbo].[IpEntry] OFF
/****** Object: Table [dbo].[Consumption] Script Date: 07/19/2017 20:37:41 ******/
SET ANSI_NULLS ON
GO
SET QUOTED_IDENTIFIER ON
GO
CREATE TABLE [dbo].[Consumption](
[Id] [int] IDENTITY(1,1) NOT NULL,
[Consumption no] [nvarchar](40) NULL,
[AnotherProductId] [int] NULL,
[Quantity] [float] NULL,
CONSTRAINT [PK_Consumption] PRIMARY KEY CLUSTERED
(
[Id] ASC
)WITH (PAD_INDEX = OFF, STATISTICS_NORECOMPUTE = OFF, IGNORE_DUP_KEY = OFF, ALLOW_ROW_LOCKS = ON, ALLOW_PAGE_LOCKS = ON) ON [PRIMARY]
) ON [PRIMARY]
GO
SET IDENTITY_INSERT [dbo].[Consumption] ON
INSERT [dbo].[Consumption] ([Id], [Consumption no], [AnotherProductId], [Quantity]) VALUES (1, N'Con-122 ', 2, 10)
INSERT [dbo].[Consumption] ([Id], [Consumption no], [AnotherProductId], [Quantity]) VALUES (2, N'Con-122 ', 4, 10)
SET IDENTITY_INSERT [dbo].[Consumption] OFF
答案 0 :(得分：1)
这应该给你预期的结果。请试试。
{{1}}
答案 1 :(得分：1)
预计这将涵盖所有情景。
SELECT DISTINCT P.ProductID,P.Quantity,-1 Flag,C.[Quantity] Balance
INTO #TMP
FROM [ProductEntry] P
INNER JOIN [IpEntry] I ON I.ProductID=P.ProductId
INNER JOIN (SELECT [AnotherProductId],SUM([Quantity]) [Quantity] FROM [Consumption] GROUP BY [AnotherProductId])C ON C.AnotherProductId=I.AnotherProductId
DECLARE @Counter INT=1
WHILE((SELECT TOP 1 1 FROM #TMP WHERE Flag=-1 )=1)
BEGIN
UPDATE T SET T.Balance = T.Balance-T.Quantity,
T.Quantity = CASE WHEN T.Quantity-T.Balance>=0 THEN T.Quantity-T.Balance ELSE 0 END,
T.Flag = CASE WHEN T.Quantity-T.Balance>=0 THEN 0 ELSE 1 END
FROM (SELECT ProductId,Quantity,row_number() over (partition by ProductId order by Quantity)RID FROM [ProductEntry])P
INNER JOIN [IpEntry] I ON I.ProductID=P.ProductId and P.RID=@Counter
INNER JOIN (SELECT ProductId,Quantity,Flag,Balance,row_number() over (partition by ProductId order by Quantity)RID FROM #TMP ) T ON T.ProductID=P.ProductID and T.RID=@Counter
INNER JOIN (SELECT [AnotherProductId],SUM([Quantity]) [Quantity] FROM [Consumption] GROUP BY [AnotherProductId])C ON C.AnotherProductId=I.AnotherProductId
UPDATE T1 SET Balance=T2.Balance
FROM #TMP T1 INNER JOIN #TMP T2 ON T1.ProductId=T2.ProductId
WHERE T2.Flag IN (0,1)
UPDATE T1 SET Flag= (SELECT T2.Flag FROM #TMP T2 WHERE T1.ProductId=T2.ProductId AND T2.Flag=0)
FROM #TMP T1
WHERE Flag=0
SET @Counter=@Counter+1
SELECT * FROM #TMP
END
SELECT ProductId,Quantity FROM #TMP --You can add more details by joining with other tables as per your requirement
drop table #TMP