Question

I’m trying to load XML data into a SQL Server table.

The XML file is:

<root>
   <Company ID="183">
      <User UserName="Kim" />
      <User UserName = "Joe" />
   </Company>
 <Company ID="123">
      <User UserName="George" />
      <User UserName = "Sal" />
   </Company>
</root>

I would like to get a table of UserName that has both the company ID and the UserName.
for this example:

ID      UserName
---     --------
183     Kim
183     Joe
123     George
123     Sal

I’ve tried a few things, such as looping through all usernames of each Company, but this is quite complex especially when I have more than 2 layers to suit for this solution.

Another thing I tried is:

SELECT
t.c.query('./UserName').value('.','varchar(20)') AS A ,
(
    SELECT
    t1.c1.value('@ID', 'varchar(10)')
    FROM @x.nodes('/root/Company')AS t1(c1)
)
FROM @x.nodes('/root/Company/User')AS t(c)

but I get an error that the subquery has more than 1 value.

Any suggestions?
Thanks!

Answer 1

Assuming your say SQL but you really mean SQL Server – try something like this:

DECLARE @input XML = '<root>
   <Company ID="183">
      <User UserName="Kim" />
      <User UserName = "Joe" />
   </Company>
 <Company ID="123">
      <User UserName="George" />
      <User UserName = "Sal" />
   </Company>
</root>'

SELECT
    CompanyID = Comp.value('(@ID)[1]', 'int'),
    UserName = USerTbl.value('(@UserName)[1]', 'varchar(50)')
FROM 
    @input.nodes('/root/Company') T(Comp)
CROSS APPLY
    Comp.nodes('User') AS T2(UserTbl)

That gives me the output you’re looking for.