Examples of Bulk Importing and Exporting XML Documents
You can bulk import XML documents into a SQL Server database or bulk export them from a SQL Server database. This topic provides examples of both.
To bulk import data from a data file into a SQL Server table or non-partitioned view, you can use the following:
You can also use the bcp utility to export data from anywhere in a SQL Server database that a SELECT statement works, including partitioned views.
INSERT ... SELECT * FROM OPENROWSET(BULK...)
For more information, see Importing and Exporting Bulk Data by Using the bcp Utility and Importing Bulk Data by Using BULK INSERT or OPENROWSET(BULK...).
The examples are the following:
A. Bulk importing XML data as a binary byte stream
When you bulk import XML data from a file that contains an encoding declaration that you want to apply, specify the SINGLE_BLOB option in the OPENROWSET(BULK…) clause. The SINGLE_BLOB option makes sure that the XML parser in SQL Server imports the data according to the encoding scheme specified in the XML declaration.
To test example A, you must create sample table T.
USE tempdb CREATE TABLE T (IntCol int, XmlCol xml) GO
Sample Data File
Before you can run example A, you must create a UTF-8 encoded file (C:\SampleFolder\SampleData3.txt) that contains the following sample instance that specifies the UTF-8 encoding scheme.
<?xml version="1.0" encoding="UTF-8"?> <Root> <ProductDescription ProductModelID="5"> <Summary>Some Text</Summary> </ProductDescription> </Root>
This example uses the SINGLE_BLOB option in an INSERT ... SELECT * FROM OPENROWSET(BULK...) statement to import data from a file named SampleData3.txt and insert an XML instance in the single-column table, sample table T.
INSERT INTO T(XmlCol) SELECT * FROM OPENROWSET( BULK 'c:\SampleFolder\SampleData3.txt', SINGLE_BLOB) AS x
By using SINGLE_BLOB in this case, you can avoid a mismatch between the encoding of the XML document (as specified by the XML encoding declaration) and the string codepage implied by the server.
If you use NCLOB or CLOB data types and run into a codepage or encoding conflict, you must do one of the following:
Remove the XML declaration to successfully import the contents of the XML data file.
Specify a code page in the CODEPAGE option of the query that matches the encoding scheme that is used in the XML declaration.
Match, or resolve, the database collation settings with a non-Unicode XML encoding scheme.
B. Bulk importing XML data in an existing row
This example uses the OPENROWSET bulk rowset provider to add an XML instance to an existing row or rows in sample table T.
To run this example, you must first complete the test script provided in example A. That example creates the tempdb.dbo.T table and bulk imports data from SampleData3.txt.
Sample Data File
Example B uses a modified version of the SampleData3.txt sample data file from the preceding example. To run this example, modify the content of this file as follows:
<Root> <ProductDescription ProductModelID="10"> <Summary>Some New Text</Summary> </ProductDescription> </Root>
C. Bulk importing XML data from a file that contains a DTD
We recommended that you not enable support for Document Type Definitions (DTDs) if it is not required in your XML environment. Turning on DTD support increases the attackable surface area of your server, and may expose it to a denial-of-service attack. If you must enable DTD support, you can reduce this security risk by processing only trusted XML documents.
During an attempt to use a bcp command to import XML data from a file that contains a DTD, an error similar to the following can occur:
"SQLState = 42000, NativeError = 6359"
"Error = [Microsoft][SQL Server Native Client][ SQL Server]Parsing XML with internal subset DTDs not allowed. Use CONVERT with style option 2 to enable limited internal subset DTD support."
"BCP copy %s failed"
To work around this problem, you can import XML data from a data file that contains a DTD by using the OPENROWSET(BULK...) function and then specifying the CONVERT option in the SELECT clause of the command. The basic syntax for the command is:
INSERT ... SELECT CONVERT(…) FROM OPENROWSET(BULK...)
Sample Data File
Before you can test this bulk import example, create a file (C:\temp\Dtdfile.xml) that contains the following sample instance:
<!DOCTYPE DOC [<!ATTLIST elem1 attr1 CDATA "defVal1">]><elem1>January</elem1>
Example C uses the T1 sample table that is created by the following CREATE TABLE statement:
USE tempdb CREATE TABLE T1(XmlCol xml); GO
This example uses OPENROWSET(BULK...) and specifies the CONVERT option in the SELECT clause to import the XML data from Dtdfile.xml into sample table T1.
INSERT T1 SELECT CONVERT(xml, BulkColumn, 2) FROM OPENROWSET(Bulk 'c:\temp\Dtdfile.xml', SINGLE_BLOB) [rowsetresults]
After the INSERT statement executes, the DTD is stripped from the XML and stored in the T1 table.
D. Specifying the field terminator explicitly using a format file
The following example shows how to bulk import the following XML document, Xmltable.dat.
Sample Data File
The document in Xmltable.dat contains two XML values, one for each row. The first XML value is encoded with UTF-16, and the second value is encoded with UTF-8.
The contents of this data file are shown in the following Hex dump:
FF FE 3C 00 3F 00 78 00-6D 00 6C 00 20 00 76 00 *..<.?.x.m.l. .v.* 65 00 72 00 73 00 69 00-6F 00 6E 00 3D 00 22 00 *e.r.s.i.o.n.=.".* 31 00 2E 00 30 00 22 00-20 00 65 00 6E 00 63 00 *1...0.". .e.n.c.* 6F 00 64 00 69 00 6E 00-67 00 3D 00 22 00 75 00 *o.d.i.n.g.=.".u.* 74 00 66 00 2D 00 31 00-36 00 22 00 3F 00 3E 00 *t.f.-.1.6.".?.>.* 3C 00 72 00 6F 00 6F 00-74 00 3E 00 A2 4F 9C 76 *<.r.o.o.t.>..O.v* 0C FA 77 E4 80 00 89 00-00 06 90 06 91 2E 9B 2E *..w.............* 99 34 A2 34 86 00 83 02-92 20 7F 02 4E C5 E4 A3 *.4.4..... ..N...* 34 B2 B7 B3 B7 FE F8 FF-F8 00 3C 00 2F 00 72 00 *4.........<./.r.* 6F 00 6F 00 74 00 3E 00-00 00 00 00 7A EF BB BF *o.o.t.>.....z...* 3C 3F 78 6D 6C 20 76 65-72 73 69 6F 6E 3D 22 31 *<?xml version="1* 2E 30 22 20 65 6E 63 6F-64 69 6E 67 3D 22 75 74 *.0" encoding="ut* 66 2D 38 22 3F 3E 3C 72-6F 6F 74 3E E4 BE A2 E7 *f-8"?><root>....* 9A 9C EF A8 8C EE 91 B7-C2 80 C2 89 D8 80 DA 90 *................* E2 BA 91 E2 BA 9B E3 92-99 E3 92 A2 C2 86 CA 83 *................* E2 82 92 C9 BF EC 95 8E-EA 8F A4 EB 88 B4 EB 8E *................* B7 EF BA B7 EF BF B8 C3-B8 3C 2F 72 6F 6F 74 3E *.........</root>* 00 00 00 00 7A *....z*
When you bulk import or export an XML document, you should use a field terminator that cannot possibly appear in any of the documents; for example, a series of four nulls (\0) followed by the letter z: \0\0\0\0z.
This example shows how to use this field terminator for the xTable sample table. To create this sample table, use the following CREATE TABLE statement:
USE tempdb CREATE TABLE xTable (xCol xml); GO
Sample Format File
The field terminator must be specified in the format file. Example D uses a non-XML format file named Xmltable.fmt that contains the following:
9.0 1 1 SQLBINARY 0 0 "\0\0\0\0z" 1 xCol ""
You can use this format file to bulk import XML documents into the xTable table by using a bcp command or a BULK INSERT or INSERT ... SELECT * FROM OPENROWSET(BULK...) statement.
E. Bulk exporting XML data
The following example uses bcp to bulk export XML data from the table that is created in the preceding example by using the same XML format file. In the following bcp command, <server_name> and <instance_name> represent placeholders that must be replaced with appropriate values:
bcp bulktest..xTable out a-wn.out -N -T -S<server_name>\<instance_name>
SQL Server does not save the XML encoding when XML data is persisted in the database. Therefore, the original encoding of XML fields is not available when XML data is exported. SQL Server uses UTF-16 encoding when exporting XML data.