Muokkaa

Jaa


Use a format file to map table columns to data-file fields (SQL Server)

Applies to: SQL Server Azure SQL Database Azure SQL Managed Instance Azure Synapse Analytics Analytics Platform System (PDW)

A data file can contain fields arranged in a different order from the corresponding columns in the table. This article presents both non-XML and XML format files that are modified to accommodate a data file whose fields are arranged in a different order from the table columns. The modified format file maps the data fields to their corresponding table columns. Review Create a format file with bcp (SQL Server) for additional information.

Either a non-XML or XML format file can be used to bulk import a data file into the table, using a bcp utility command, BULK INSERT statement, or INSERT ... SELECT * FROM OPENROWSET statement. For more information, see Use a format file to bulk import data (SQL Server).

Note

This syntax, including bulk insert, isn't supported in Azure Synapse Analytics. In Azure Synapse Analytics and other cloud database platform integrations, accomplish data movement via the COPY statement in Azure Data Factory, or by using T-SQL statements such as COPY INTO and PolyBase.

Example test conditions

The examples of modified format files in this article are based on the table and data file defined later in this article.

Sample table

The following Transact-SQL script creates a test database and a table named myRemap.

CREATE DATABASE TestDatabase;
GO

USE TestDatabase;

CREATE TABLE myRemap
(
    PersonID SMALLINT,
    FirstName VARCHAR (25),
    LastName VARCHAR (30),
    Gender CHAR (1)
);

Sample data file

The following data presents FirstName and LastName in the reverse order as presented in the table myRemap. Using a text editor such as Notepad, create an empty file D:\BCP\myRemap.bcp and insert the following data:

1,Grosse,Anthony,M
2,Fatnowna,Alica,F
3,Rosenhain,Stella,F

Create the format files

To bulk import data from myRemap.bcp into the myRemap table, the format file must do the following actions:

  • Map the first data field to the first column, PersonID.
  • Map the second data field to the third column, LastName.
  • Map the third data field to the second column, FirstName.
  • Map the fourth data field to the fourth column, Gender.

The simplest method to create the format file is by using the bcp utility. First, create a base format file from the existing table. Second, modify the base format file to reflect the actual data file.

Create a non-XML format file

Review Use Non-XML format files (SQL Server) for detailed information. The following command uses the bcp utility to generate a non-XML format file, myRemap.fmt, based on the schema of myRemap. In addition, the qualifier c is used to specify character data, t, is used to specify a comma as a field terminator, and T is used to specify a trusted connection using integrated security. At a command prompt, enter the following command:

bcp TestDatabase.dbo.myRemap format nul -c -f D:\BCP\myRemap.fmt -t, -T

Modify the non-XML format file

For information about terminology, see Structure of Non-XML Format Files. Open D:\BCP\myRemap.fmt in Notepad and perform the following modifications:

  1. Rearrange the order of the format-file rows so that the rows are in the same order as the data in myRemap.bcp.
  2. Ensure the host file field order values are sequential.
  3. Ensure there's a carriage return after the last format-file row.

Compare the changes:

Before

13.0
4
1       SQLCHAR    0       7       ","      1     PersonID               ""
2       SQLCHAR    0       25      ","      2     FirstName              SQL_Latin1_General_CP1_CI_AS
3       SQLCHAR    0       30      ","      3     LastName               SQL_Latin1_General_CP1_CI_AS
4       SQLCHAR    0       1       "\r\n"   4     Gender                 SQL_Latin1_General_CP1_CI_AS

After

13.0
4
1       SQLCHAR    0       7       ","      1     PersonID               ""
2       SQLCHAR    0       30      ","      3     LastName               SQL_Latin1_General_CP1_CI_AS
3       SQLCHAR    0       25      ","      2     FirstName              SQL_Latin1_General_CP1_CI_AS
4       SQLCHAR    0       1       "\r\n"   4     Gender                 SQL_Latin1_General_CP1_CI_AS

The modified format file now reflects:

  • The first data field in myRemap.bcp is mapped to the first column, myRemap.. PersonID
  • The second data field in myRemap.bcp is mapped to the third column, myRemap.. LastName
  • The third data field in myRemap.bcp is mapped to the second column, myRemap.. FirstName
  • The fourth data field in myRemap.bcp is mapped to the fourth column, myRemap.. Gender

Create an XML format file

Review XML Format Files (SQL Server) for detailed information. The following command uses the bcp utility to create an xml format file, myRemap.xml, based on the schema of myRemap. In addition, the qualifier c is used to specify character data, t, is used to specify a comma as a field terminator, and T is used to specify a trusted connection using integrated security. The x qualifier must be used to generate an XML-based format file. At a command prompt, enter the following command:

bcp TestDatabase.dbo.myRemap format nul -c -x -f D:\BCP\myRemap.xml -t, -T

Modify the XML format file

Review Schema syntax for XML format files for terminology. Open D:\BCP\myRemap.xml in Notepad and perform the following modifications:

  1. The order in which the <FIELD> elements are declared in the format file, is the order in which those fields appear in the data file. Thus, reverse the order for the <FIELD> elements with ID attributes 2 and 3.

  2. Ensure the <FIELD> ID attribute values are sequential.

  3. The order of the <COLUMN> elements in the <ROW> element defines the order in which the bulk operation sends them to the target. The XML format file assigns each <COLUMN> element a local name that has no relationship to the column in the target table of a bulk import operation. The order of the <COLUMN> elements is independent of the order of <FIELD> elements in a <RECORD> definition. Each <COLUMN> element corresponds to a <FIELD> element (whose ID is specified in the SOURCE attribute of the <COLUMN> element). Thus, the values for <COLUMN> SOURCE are the only attributes that require revision. Reverse the order for <COLUMN> SOURCE attributes 2 and 3.

Compare the changes:

Before

<?xml version="1.0"?>
<BCPFORMAT xmlns="http://schemas.microsoft.com/sqlserver/2004/bulkload/format" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">
<RECORD>
  <FIELD ID="1" xsi:type="CharTerm" TERMINATOR="," MAX_LENGTH="7"/>
  <FIELD ID="2" xsi:type="CharTerm" TERMINATOR="," MAX_LENGTH="25" COLLATION="SQL_Latin1_General_CP1_CI_AS"/>
  <FIELD ID="3" xsi:type="CharTerm" TERMINATOR="," MAX_LENGTH="30" COLLATION="SQL_Latin1_General_CP1_CI_AS"/>
  <FIELD ID="4" xsi:type="CharTerm" TERMINATOR="\r\n" MAX_LENGTH="1" COLLATION="SQL_Latin1_General_CP1_CI_AS"/>
</RECORD>
<ROW>
  <COLUMN SOURCE="1" NAME="PersonID" xsi:type="SQLSMALLINT"/>
  <COLUMN SOURCE="2" NAME="FirstName" xsi:type="SQLVARYCHAR"/>
  <COLUMN SOURCE="3" NAME="LastName" xsi:type="SQLVARYCHAR"/>
  <COLUMN SOURCE="4" NAME="Gender" xsi:type="SQLCHAR"/>
</ROW>
</BCPFORMAT>

After

<?xml version="1.0"?>
<BCPFORMAT xmlns="http://schemas.microsoft.com/sqlserver/2004/bulkload/format" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">
<RECORD>
  <FIELD ID="1" xsi:type="CharTerm" TERMINATOR="," MAX_LENGTH="7"/>
  <FIELD ID="2" xsi:type="CharTerm" TERMINATOR="," MAX_LENGTH="30" COLLATION="SQL_Latin1_General_CP1_CI_AS"/>
  <FIELD ID="3" xsi:type="CharTerm" TERMINATOR="," MAX_LENGTH="25" COLLATION="SQL_Latin1_General_CP1_CI_AS"/>
  <FIELD ID="4" xsi:type="CharTerm" TERMINATOR="\r\n" MAX_LENGTH="1" COLLATION="SQL_Latin1_General_CP1_CI_AS"/>
</RECORD>
<ROW>
  <COLUMN SOURCE="1" NAME="PersonID" xsi:type="SQLSMALLINT"/>
  <COLUMN SOURCE="3" NAME="FirstName" xsi:type="SQLVARYCHAR"/>
  <COLUMN SOURCE="2" NAME="LastName" xsi:type="SQLVARYCHAR"/>
  <COLUMN SOURCE="4" NAME="Gender" xsi:type="SQLCHAR"/>
</ROW>
</BCPFORMAT>

The modified format file now reflects:

  • FIELD 1, which corresponds to COLUMN 1, is mapped to the first table column, myRemap.. PersonID
  • FIELD 2, which corresponds to COLUMN 2, is remapped to the third table column, myRemap.. LastName
  • FIELD 3, which corresponds to COLUMN 3, is remapped to the second table column, myRemap.. FirstName
  • FIELD 4, which corresponds to COLUMN 4, is mapped to the fourth table column, myRemap.. Gender

Import data with a format file to map table columns to data-file field

The following examples use the database, datafile, and format files created previously.

Use bcp and non-XML format file

At a command prompt, enter the following command.

bcp TestDatabase.dbo.myRemap IN D:\BCP\myRemap.bcp -f D:\BCP\myRemap.fmt -T

Use bcp and XML format file

At a command prompt, enter the following command.

bcp TestDatabase.dbo.myRemap IN D:\BCP\myRemap.bcp -f D:\BCP\myRemap.xml -T

Use BULK INSERT and non-XML format file

Run the following Transact-SQL script.

USE TestDatabase;
GO

TRUNCATE TABLE myRemap;

BULK INSERT dbo.myRemap FROM 'D:\BCP\myRemap.bcp'
    WITH (FORMATFILE = 'D:\BCP\myRemap.fmt');
GO

-- review results
SELECT *
FROM TestDatabase.dbo.myRemap;

Use BULK INSERT and XML format file

Run the following Transact-SQL script.

USE TestDatabase;
GO

TRUNCATE TABLE myRemap;

BULK INSERT dbo.myRemap FROM 'D:\BCP\myRemap.bcp'
    WITH (FORMATFILE = 'D:\BCP\myRemap.xml');
GO

-- review results
SELECT *
FROM TestDatabase.dbo.myRemap;

Use OPENROWSET and non-XML format file

Run the following Transact-SQL script.

USE TestDatabase;
GO

TRUNCATE TABLE myRemap;

INSERT INTO dbo.myRemap
SELECT *
FROM OPENROWSET (
    BULK 'D:\BCP\myRemap.bcp',
    FORMATFILE = 'D:\BCP\myRemap.fmt'
) AS t1;
GO

-- review results
SELECT *
FROM TestDatabase.dbo.myRemap;

Use OPENROWSET and XML format file

Run the following Transact-SQL script.

USE TestDatabase;
GO

TRUNCATE TABLE myRemap;

INSERT INTO dbo.myRemap
SELECT *
FROM OPENROWSET (
    BULK 'D:\BCP\myRemap.bcp',
    FORMATFILE = 'D:\BCP\myRemap.xml'
) AS t1;
GO

-- review results
SELECT *
FROM TestDatabase.dbo.myRemap;