FOR Clause (Transact-SQL)

FOR clause is used to specify either the BROWSE or the XML option. BROWSE and XML are unrelated options.

Important

The XMLDATA directive to the FOR XML option is deprecated. Use XSD generation in the case of RAW and AUTO modes. There is no replacement for the XMLDATA directive in EXPLICIT mode. This feature will be removed in a future version of Microsoft SQL Server. Avoid using this feature in new development work, and plan to modify applications that currently use this feature.

Topic link iconTransact-SQL Syntax Conventions

Syntax

[ FOR { BROWSE | <XML> } ]
<XML> ::=
XML 
{ 
    { RAW [ ( 'ElementName' ) ] | AUTO } 
    [ 
        <CommonDirectives> 
        [ , { XMLDATA | XMLSCHEMA [ ( 'TargetNameSpaceURI' ) ] } ] 
        [ , ELEMENTS [ XSINIL | ABSENT ] 
    ]
  | EXPLICIT 
    [ 
        <CommonDirectives> 
        [ , XMLDATA ] 
    ]
  | PATH [ ( 'ElementName' ) ] 
    [
        <CommonDirectives> 
        [ , ELEMENTS [ XSINIL | ABSENT ] ]
    ]
} 

<CommonDirectives> ::= 
[ , BINARY BASE64 ]
[ , TYPE ]
[ , ROOT [ ( 'RootName' ) ] ]

Arguments

  • BROWSE
    Specifies that updates be allowed while viewing the data in a DB-Library browse mode cursor. A table can be browsed in an application if the table includes a timestamp column, the table has a unique index, and the FOR BROWSE option is at the end of the SELECT statements sent to an instance of SQL Server.

    Note

    You cannot use the <lock_hint> HOLDLOCK in a SELECT statement that includes the FOR BROWSE option.

    FOR BROWSE cannot appear in SELECT statements that are joined by the UNION operator.

    Note

    When the unique index key columns of a table are nullable, and the table is on the inner side of an outer join, the index is not supported by browse mode.

    The browse mode lets you scan the rows in your SQL Server table and update the data in your table one row at a time. To access a SQL Server table in your application in the browse mode, you must use one of the following two options:

    • The SELECT statement that you use to access the data from your SQL Server table must end with the keywords FOR BROWSE. When you turn on the FOR BROWSE option to use browse mode, temporary tables are created.

    • You must run the following Transact-SQL statement to turn on the browse mode by using the NO_BROWSETABLE option:

      SET NO_BROWSETABLE ON
      

      When you turn on the NO_BROWSETABLE option, all the SELECT statements behave as if the FOR BROWSE option is appended to the statements. However, the NO_BROWSETABLE option does not create the temporary tables that the FOR BROWSE option generally uses to send the results to your application.

    When you try to access the data from SQL Server tables in browse mode by using a SELECT query that involves an outer join statement, and when a unique index is defined on the table that is present on the inner side of an outer join statement, the browse mode does not support the unique index. The browse mode supports the unique index only when all the unique index key columns can accept null values. The browse mode does not support the unique index if the following conditions are true:

    • You try to access the data from SQL Server tables in browse mode by using a SELECT query that involves an outer join statement.

    • A unique index is defined on the table that is present on the inner side of an outer join statement.

    To reproduce this behavior in the browse mode, follow these steps:

    1. In SQL Server Management Studio, create a database, named SampleDB.

    2. In the SampleDB database, create a tleft table and a tright table that both contain a single column that is named c1. Define a unique index on the c1 column in the tleft table, and set the column to accept null values. To do this, run the following Transact-SQL statements in an appropriate query window:

      CREATE TABLE tleft(c1 INT NULL UNIQUE) ;
      GO 
      CREATE TABLE tright(c1 INT NULL) ;
      GO
      
    3. Insert several values in the tleft table and the tright table. Make sure that you insert a null value in the tleft table. To do this, run the following Transact-SQL statements in the query window:

      INSERT INTO tleft VALUES(2) ;
      INSERT INTO tleft VALUES(NULL) ;
      INSERT INTO tright VALUES(1) ;
      INSERT INTO tright VALUES(3) ;
      INSERT INTO tright VALUES(NULL) ;
      GO
      
    4. Turn on the NO_BROWSETABLE option. To do this, run the following Transact-SQL statements in the query window:

      SET NO_BROWSETABLE ON ;
      GO
      
    5. Access the data in the tleft table and the tright table by using an outer join statement in the SELECT query. Make sure that the tleft table is on the inner side of the outer join statement. To do this, run the following Transact-SQL statements in the query window:

      SELECT tleft.c1 
      FROM tleft 
      RIGHT JOIN tright 
      ON tleft.c1 = tright.c1 
      WHERE tright.c1 <> 2 ;
      

      Notice the following output in the Results pane:

      c1

      ----

      NULL

      NULL

    After you run the SELECT query to access the tables in the browse mode, the result set of the SELECT query contains two null values for the c1 column in the tleft table because of the definition of the right outer join statement. Therefore, in the result set, you cannot distinguish between the null values that came from the table and the null values that the right outer join statement introduced. You might receive incorrect results if you must ignore the null values from the result set.

    Note

    If the columns that are included in the unique index do not accept null values, all the null values in the result set were introduced by the right outer join statement.

  • XML
    Specifies that the results of a query are to be returned as an XML document. One of the following XML modes must be specified: RAW, AUTO, EXPLICIT. For more information about XML data and SQL Server, see Constructing XML Using FOR XML.

  • RAW [ ('ElementName') ]
    Takes the query result and transforms each row in the result set into an XML element with a generic identifier <row /> as the element tag. You can optionally specify a name for the row element. The resulting XML output uses the specified ElementName as the row element generated for each row. For more information, see Using RAW Mode and Using RAW Mode.

  • AUTO
    Returns query results in a simple, nested XML tree. Each table in the FROM clause, for which at least one column is listed in the SELECT clause, is represented as an XML element. The columns listed in the SELECT clause are mapped to the appropriate element attributes. For more information, see Using AUTO Mode.

  • EXPLICIT
    Specifies that the shape of the resulting XML tree is defined explicitly. Using this mode, queries must be written in a particular way so that additional information about the desired nesting is specified explicitly. For more information, see Using EXPLICIT Mode.

  • XMLDATA
    Returns inline XDR schema, but does not add the root element to the result. If XMLDATA is specified, XDR schema is appended to the document.

  • XMLSCHEMA [ ('TargetNameSpaceURI') ]
    Returns inline XSD schema. You can optionally specify a target namespace URI when you specify this directive, which returns the specified namespace in the schema. For more information, see Inline XSD Schema Generation.

  • ELEMENTS
    Specifies that the columns are returned as subelements. Otherwise, they are mapped to XML attributes. This option is supported in RAW, AUTO and PATH modes only. For more information, see Using RAW Mode.

  • XSINIL
    Specifies that an element with xsi:nil attribute set to True be created for NULL column values. This option can only be specified with ELEMENTS directive. For more information, see Generating Elements for NULL Values Using the XSINIL Parameter.

  • ABSENT
    Indicates that for null column values, corresponding XML elements will not be added in the XML result. Specify this option only with ELEMENTS.

  • PATH [ ('ElementName') ]
    Generates a <row> element wrapper for each row in the result set. You can optionally specify an element name for the <row> element wrapper. If an empty string is provided, such as FOR XML PATH ('') ), a wrapper element is not generated. Using PATH may provide a simpler alternative to queries written using the EXPLICIT directive. For more information, see Using PATH Mode.

  • BINARY BASE64
    Specifies that the query returns the binary data in binary base64-encoded format. When you retrieve binary data by using RAW and EXPLICIT mode, this option must be specified. This is the default in AUTO mode.

  • TYPE
    Specifies that the query returns results as xml type. For more information, see TYPE Directive in FOR XML Queries.

  • ROOT [ ('RootName') ]
    Specifies that a single top-level element be added to the resulting XML. You can optionally specify the root element name to generate. If the optional root name is not specified, the default <root> element is added.

Examples

The following example specifies FOR XML AUTO with the TYPE and XMLSCHEMA options. Because of the TYPE option, the result set is returned to the client as an xml type. The XMLSCHEMA option specifies that the inline XSD schema is included in the XML data returned, and the ELEMENTS option specifies that the XML result is element-centric.

USE AdventureWorks2008R2;
GO
SELECT p.BusinessEntityID, FirstName, LastName, PhoneNumber AS Phone
FROM Person.Person AS p
Join Person.PersonPhone AS pph ON p.BusinessEntityID  = pph.BusinessEntityID
WHERE LastName LIKE 'G%'
ORDER BY LastName, FirstName 
FOR XML AUTO, TYPE, XMLSCHEMA, ELEMENTS XSINIL;