SQL Server：批量加载失败。数据文件中第 1 行第 1 列的列太长

Question

请有人帮助我。已经看了几个小时了，但没有任何结果。

我使用以下脚本在 SQL Express 2008 R2 中创建了一个表：

CREATE TABLE Features
(
ID int not null identity(1,1 ),
StopID varchar(10), 
Code int,
Name varchar(100),
Summary varchar(200),
Lat real,
Lon real,
street varchar(100),
city varchar(50),
region varchar(50),
postcode varchar(10),
country varchar(20),
zone_id varchar(20),
the_geom geography


 CONSTRAINT [PK_Features] 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]

然后我使用 bcp 工具创建了针对我的数据库表创建的以下格式文件：

10.0
12
1       SQLCHAR             2       100     ","    2     StopID               Latin1_General_CI_AS
2       SQLINT              1       4       ","    3     Code                 ""
3       SQLCHAR             2       100     ","    4     Name                 Latin1_General_CI_AS
4       SQLCHAR             2       200     ","    5     Summary              Latin1_General_CI_AS
5       SQLFLT4             1       4       ","    6     Lat                  ""
6       SQLFLT4             1       4       ","    7     Lon                  ""
7       SQLCHAR             2       100     ","    8     street               Latin1_General_CI_AS
8       SQLCHAR             2       50      ","    9     city                 Latin1_General_CI_AS
9       SQLCHAR             2       50      ","    10    region               Latin1_General_CI_AS
10      SQLCHAR             2       10      ","    11    postcode             Latin1_General_CI_AS
11      SQLCHAR             2       20      ","    12    country              Latin1_General_CI_AS
12      SQLCHAR             2       20      "\r\n"    13    zone_id              Latin1_General_CI_AS

此文件已被修改以删除 ID 和 the_geom 字段，因为这些字段不在我的数据文件中。

然后我尝试批量插入包含以下内容的 1 行 csv：

a,8,S,,45.439869,-75.695839,,,,,,

我得到的一切：

Msg 4866, Level 16, State 7, Line 35
The bulk load failed. The column is too long in the data file for row 1, column 1. Verify that the field terminator and row terminator are specified correctly.
Msg 7399, Level 16, State 1, Line 35
The OLE DB provider "BULK" for linked server "(null)" reported an error. The provider did not give any information about the error.
Msg 7330, Level 16, State 2, Line 35
Cannot fetch a row from OLE DB provider "BULK" for linked server "(null)".

任何指示都会有所帮助，因为我无法弄清楚这一点。

Answer 1

问题是由我的格式文件中的默认前缀长度设置引起的。我导入的数据文件不是使用 bcp 创建的，因此我必须将所有字段的前缀长度设置为 0，如下所示：

0.0
12
1       SQLCHAR             0       100     ","    2     StopID               Latin1_General_CI_AS
2       SQLINT              0       4       ","    3     Code                 ""
3       SQLCHAR             0       100     ","    4     Name                 Latin1_General_CI_AS
4       SQLCHAR             0       200     ","    5     Summary              Latin1_General_CI_AS
5       SQLFLT4             0       4       ","    6     Lat                  ""
6       SQLFLT4             0       4       ","    7     Lon                  ""
7       SQLCHAR             0       100     ","    8     street               Latin1_General_CI_AS
8       SQLCHAR             0       50      ","    9     city                 Latin1_General_CI_AS
9       SQLCHAR             0       50      ","    10    region               Latin1_General_CI_AS
10      SQLCHAR             0       10      ","    11    postcode             Latin1_General_CI_AS
11      SQLCHAR             0       20      ","    12    country              Latin1_General_CI_AS
12      SQLCHAR             0       20      "\r\n"    13    zone_id              Latin1_General_CI_AS

这样修改后，导入成功了。

Answer 2

试试这个，

ROWTERMINATOR = '0x0a'

Answer 3

我今天遇到了这个问题，但仅限于文本值超过 8000 个字符的特定 ROWS 和 COLUMNS。不管我的 FMT 文件是否为 SQLCHAR 0 0（表示最大值），但在管道的某个地方，最大值为 8000

我正在使用 AZURE SQL 并尝试读取 Azure Blob 容器中的 CSV。

Answer 4

无论如何，我遇到了同样的问题，因为 CSV 中日期字段的预期格式和实际格式之间存在冲突。我更改了 CSV 中的日期格式，并且成功了。

Answer 5

如果您使用

bcp

生成格式文件，请注意 -n 和 -c 标志。

我的问题是我指定了 -n （本机）标志，而我的数据文件仅与 -c （字符）标志兼容。

格式文件生成

bcp <DATABASE>.<SCHEMA>.<TABLE> format nul -t ; -T -c -S <IP> -f filename.fmt

数据文件生成

bcp "select * from table" queryout "filename.dat" -c -C 65001 -t ; -S <IP> -d <DATABASE> -T

批量导入

BULK INSERT dbo.table 
FROM 'filename.dat'
WITH (FORMATFILE = 'filename.fmt');