Is there a way to select all elements from xmlvalue.value method instead of specifying [1] - sql

Question

I want to access all instances and not specify in XMLvalues.value method like 1
I have joined the nodes using outer apply but I need to specify instances number
SELECT
Id ,
XmlValues2.value('(date1)[1]', 'DateTime') AS date1 ,
XmlValues.value('(name)[1]', 'varchar(1299)') AS bank ,
XmlValues.value('(country)[1]', 'varchar(1299)') AS Country ,
FROM
Temp_board I OUTER APPLY I.board.nodes('/Report/basicInfo/report') AS xmlTableInner(XmlValues2)
OUTER APPLY XmlValues2.nodes('/Report/basicInfo/bank') AS xmlTable(XmlValues)
WHERE
Id ='235908235'
So I want the result of all nodes elements and not just the first instant
Attached screenshot of xml.I want to access all instances of transactions.XML example

Answer 1

For your next question please try to set up a MCVE. This is a stand along self-running example to reproduce your issue. And please do not provide code or data as pictures. This means, that somebody else has to type this in...
It is probleamtic, that the XML's picture does not inlcude the root tag and eventually existing namespaces. Furthermore, the targeted <Transaction> elements are completely empty in your picture... Furthermore, the code you provide in your question has nothing to do with the XML you show...
I think you can understand, that everything must be pure guessing and my magic crystall ball is out for cleaning at the moment ;-)
This time I do the mcve for you. A lot is just guessing to show the principles. Try this out and try to adopt it for your actual issues:
Declare a mockup-XML:
DECLARE #xml XML=
N'<SomeRootNode>
<accountList>
<Transaction SomeAttribute="blah1">
<SomeSubElement subAttr="sub1">element1</SomeSubElement>
</Transaction>
<Transaction SomeAttribute="blah2">
<SomeSubElement subAttr="sub2">element2</SomeSubElement>
</Transaction>
<Transaction SomeAttribute="blah3">
<SomeSubElement subAttr="sub3">element3</SomeSubElement>
</Transaction>
</accountList>
</SomeRootNode>';
--The query
SELECT tr.value('#SomeAttribute','varchar(100)') TheAttributeInTransactionElement
,tr.value('(SomeSubElement/#subAttr)[1]','varchar(100)') TheAttributeInTheSubElement
,tr.value('(SomeSubElement/text())[1]','varchar(100)') TheSubElementsContent
FROM #xml.nodes('/SomeRootNode/accountList/Transaction') A(tr);
The idea in short:
- .nodes() will dive into <SomeRootNode>, deeper into <accountList> and will return all <Transaction> elements found at this level
- The result of .nodes() is a derived table called A with one single XML-typed column tr, representing each Transaction in a single row.
- The .value() can pick attributes directly
- The .value() can use any kind of XPath/XQuery to fetch values from within the structure.
UPDATE Your XML has one more level Account
Try this
DECLARE #xml XML=
N'<SomeRootNode>
<accountList>
<Account>
<number>1</number>
<Transaction SomeAttribute="blah1">
<SomeSubElement subAttr="sub1">element1</SomeSubElement>
</Transaction>
<Transaction SomeAttribute="blah2">
<SomeSubElement subAttr="sub2">element2</SomeSubElement>
</Transaction>
<Transaction SomeAttribute="blah3">
<SomeSubElement subAttr="sub3">element3</SomeSubElement>
</Transaction>
</Account>
<Account>
<number>2</number>
<Transaction SomeAttribute="blah2-1">
<SomeSubElement subAttr="sub2-1">element2-1</SomeSubElement>
</Transaction>
<Transaction SomeAttribute="blah2-2">
<SomeSubElement subAttr="sub2-2">element2-2</SomeSubElement>
</Transaction>
<Transaction SomeAttribute="blah2-3">
<SomeSubElement subAttr="sub2-3">element2-3</SomeSubElement>
</Transaction>
</Account>
</accountList>
</SomeRootNode>';
SELECT acc.value('(number/text())[1]','int') AccountNumber
,tr.value('#SomeAttribute','varchar(100)') TheAttributeInTransactionElement
,tr.value('(SomeSubElement/#subAttr)[1]','varchar(100)') TheAttributeInTheSubElement
,tr.value('(SomeSubElement/text())[1]','varchar(100)') TheSubElementsContent
FROM #xml.nodes('/SomeRootNode/accountList/Account') A(acc)
OUTER APPLY A.acc.nodes('Transaction') B(tr);
Now we can use two calls to .nodes(). The first will return all <Account> elements within <accountList>, the second will return the nested <Transaction> elements as it is working with A.acc as input.