Question

我有一个txt文件，其中包含1600行和82列逗号描述的数据，我试图将其导入到表中。我在最后一个字段的每一行都收到以下错误：

Msg 4864，Level 16，State 1，Line 1 第1行第81列（DB252D20C8）的批量加载数据转换错误（指定代码页的类型不匹配或无效字符）。

导入声明是

BULK
INSERT  [ENERGY].[dbo].[READINGS1]
       from 'c:\readings2.txt'
with
(
DATAFILETYPE='widechar',
FIELDTERMINATOR = ',',
ROWTERMINATOR = '\n'
)
GO

表结构如下，脚本的顶部和底部：

USE [ENERGY]
GO
/****** Object:  Table [dbo].[READINGS1]    Script Date: 05/13/2013 20:00:30 ******/
SET ANSI_NULLS ON
GO
SET QUOTED_IDENTIFIER ON
GO
CREATE TABLE [dbo].[READINGS1](
[DateAndTime] [datetime] NOT NULL,
[DB240D4C7] [float] NULL,
[DB240D8C7] [float] NULL,
[DB240D12C7] [float] NULL,
[DB240D16C7] [float] NULL, 



[DB252D12C8] [float] NULL,
[DB252D16C8] [float] NULL,
[DB252D20C8] [float] NULL,
 CONSTRAINT [READINGS1DataTimeStamp] PRIMARY KEY CLUSTERED 
(
[DateAndTime] ASC
)WITH (PAD_INDEX  = OFF, IGNORE_DUP_KEY = OFF) ON [PRIMARY]
) ON [PRIMARY]

文本文件如下：

2013-02-19 00:00:00.000,6,945,1886,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,22,2040,6299,0,0,6,567,1248,0,0,251,8859,8655,0,0,10,316,1786,0,0,7,180,1206,0,0,1,16,56,0,0,368,18953,36949,0,0,NULL,NULL
2013-02-19 01:00:00.000,6,147,1886,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,22,1516,6299,0,0,3,115,1248,0,0,250,5077,8655,0,0,9,219,1786,0,0,5,147,1206,0,0,1,15,56,0,0,362,8907,36949,0,0,NULL,NULL

Answer 1

好的，你需要做的就是改变你的语句，以便在文件结束后使用KEEPNULLS。这会通知SQL服务器您希望保留空值。目前，它正在尝试将NULL转换为您的FLOAT COLUMN字符串。改变你的观点，看起来像这样。

BULK
INSERT  [ENERGY].[dbo].[READINGS1]
       from 'c:\readings2.txt'
with
(
DATAFILETYPE='widechar',
FIELDTERMINATOR = ',',
ROWTERMINATOR = '\n',
KEEPNULLS
)
GO

BOL上有一篇关于此事的文章。

否则，您始终可以构建Integration Services包来处理此问题。这是从平面文件源导入信息的一种简单快捷的方法。

Answer 2

事实证明，表格的输入文本文件中有太多字段。

SQL Server 2005 - 批量插入失败

2 个答案: