Bewerken

Delen via


Functions on Sequences - id

Applies to: SQL Server

Returns the sequence of element nodes with xs:ID values that match the values of one or more of the xs:IDREF values supplied in $arg.

Syntax

  
fn:id($arg as xs:IDREF*) as element()*  

Arguments

$arg
One or more xs:IDREF values.

Remarks

The result of the function is a sequence of elements in the XML instance, in document order, that has an xs:ID value equal to one or more of the xs:IDREFs in the list of candidate xs:IDREFs.

If the xs:IDREF value does not match any element, the function returns the empty sequence.

Examples

This topic provides XQuery examples against XML instances that are stored in various xml type columns in the AdventureWorks2022 database.

A. Retrieving elements based on the IDREF attribute value

The following example uses fn:id to retrieve the <employee> elements, based on the IDREF manager attribute. In this example, the manager attribute is an IDREF type attribute and the eid attribute is an ID type attribute.

For a specific manager attribute value, the id() function finds the <employee> element whose ID type attribute value matches the input IDREF value. In other words, for a specific employee, the id() function returns employee manager.

This is what happens in the example:

  • An XML schema collection is created.

  • A typed xml variable is created by using the XML schema collection.

  • The query retrieves the element that has an ID attribute value referenced by the manager IDREF attribute of the <employee> element.

-- If exists, drop the XML schema collection (SC).  
-- drop xml schema collection SC  
-- go  
  
create xml schema collection SC as  
'<schema xmlns="http://www.w3.org/2001/XMLSchema" xmlns:e="emp" targetNamespace="emp">  
            <element name="employees" type="e:EmployeesType"/>  
            <complexType name="EmployeesType">  
                 <sequence>  
                      <element name="employee" type="e:EmployeeType" minOccurs="0" maxOccurs="unbounded" />  
                 </sequence>  
            </complexType>    
  
            <complexType name="EmployeeType">  
                        <attribute name="eid" type="ID" />  
                        <attribute name="name" type="string" />  
                        <attribute name="manager" type="IDREF" />  
            </complexType>         
</schema>'  
go  
declare @x xml(SC)  
set @x='<e:employees xmlns:e="emp">  
<employee eid="e1" name="Joe" manager="e10" />  
<employee eid="e2" name="Bob" manager="e10" />  
<employee eid="e10" name="Dave" manager="e10" />  
</e:employees>'  
  
select @x.value(' declare namespace e="emp";   
 (fn:id(e:employees/employee[@name="Joe"]/@manager)/@name)[1]', 'varchar(50)')   
Go  

The query returns "Dave" as the value. This indicates that Dave is Joe's manager.

B. Retrieving elements based on the OrderList IDREFS attribute value

In the following example, the OrderList attribute of the <Customer> element is an IDREFS type attribute. It lists the order ids for that specific customer. For each order id, there is an <Order> element child under the <Customer> providing the order value.

The query expression, data(CustOrders:Customers/Customer[1]/@OrderList)[1], retrieves the first value from the IDRES list for the first customer. This value is then passed to the id() function. The function then finds the <Order> element whose OrderID attribute value matches the input to the id() function.

drop xml schema collection SC  
go  
create xml schema collection SC as  
'<schema xmlns="http://www.w3.org/2001/XMLSchema" xmlns:Customers="Customers" targetNamespace="Customers">  
            <element name="Customers" type="Customers:CustomersType"/>  
            <complexType name="CustomersType">  
                        <sequence>  
                            <element name="Customer" type="Customers:CustomerType" minOccurs="0" maxOccurs="unbounded" />  
                        </sequence>  
            </complexType>  
             <complexType name="OrderType">  
                <sequence minOccurs="0" maxOccurs="unbounded">  
                            <choice>  
                                <element name="OrderValue" type="integer" minOccurs="0" maxOccurs="unbounded"/>  
                            </choice>  
                </sequence>                                             
                <attribute name="OrderID" type="ID" />  
            </complexType>  
  
            <complexType name="CustomerType">  
                <sequence minOccurs="0" maxOccurs="unbounded">  
                            <choice>  
                                <element name="spouse" type="string" minOccurs="0" maxOccurs="unbounded"/>  
                                <element name="Order" type="Customers:OrderType" minOccurs="0" maxOccurs="unbounded"/>  
                            </choice>  
                </sequence>                                             
                <attribute name="CustomerID" type="string" />  
                <attribute name="OrderList" type="IDREFS" />  
            </complexType>  
 </schema>'  
go  
declare @x xml(SC)  
set @x='<CustOrders:Customers xmlns:CustOrders="Customers">  
                <Customer CustomerID="C1" OrderList="OrderA OrderB"  >  
                              <spouse>Jenny</spouse>  
                                <Order OrderID="OrderA"><OrderValue>11</OrderValue></Order>  
                                <Order OrderID="OrderB"><OrderValue>22</OrderValue></Order>  
  
                </Customer>  
                <Customer CustomerID="C2" OrderList="OrderC OrderD" >  
                                <spouse>John</spouse>  
                                <Order OrderID="OrderC"><OrderValue>33</OrderValue></Order>  
                                <Order OrderID="OrderD"><OrderValue>44</OrderValue></Order>  
  
                        </Customer>  
                <Customer CustomerID="C3"  OrderList="OrderE OrderF" >  
                                <spouse>Jane</spouse>  
                                <Order OrderID="OrderE"><OrderValue>55</OrderValue></Order>  
                                <Order OrderID="OrderF"><OrderValue>55</OrderValue></Order>  
                </Customer>  
                <Customer CustomerID="C4"  OrderList="OrderG"  >  
                                <spouse>Tim</spouse>  
                                <Order OrderID="OrderG"><OrderValue>66</OrderValue></Order>  
                        </Customer>  
                <Customer CustomerID="C5"  >  
                </Customer>  
                <Customer CustomerID="C6" >  
                </Customer>  
                <Customer CustomerID="C7"  >  
                </Customer>  
</CustOrders:Customers>'  
select @x.query('declare namespace CustOrders="Customers";  
  id(data(CustOrders:Customers/Customer[1]/@OrderList)[1])')  
  
-- result  
<Order OrderID="OrderA">  
  <OrderValue>11</OrderValue>  
</Order>  

Implementation Limitations

These are the limitations:

  • SQL Server does not support the two-argument version of id().

  • SQL Server requires the argument type of id() to be a subtype of xs:IDREF*.

See Also

Functions on Sequences