Failed to enable constraints. One or more rows contain values violating non-null, unique, or for

Hello all, I just want to make a quick post on the “Failed

to enable constraints. One or more rows contain values violating non-null,

unique, or foreign-key constraints.”

I am working on a small system at the moment and I have a

CODES table with columns CODE and DESC. When I first set up the database I set CODE

to varchar(MAX) and same with DESC, Then when I put all the codes and

descriptions in and tried to run fill I got the error.

Now weather this will help anyone I don’t know, I found out

that it was the Colum DESC in the DataSet that was causing the problem, if you

enter your dataset designer and select one of your columns, Right click, properties,

there is a field called MAXLENGTH, When I checked it was set to 50 even though I

had set varchar(MAX), so I knew that some of my descriptions where bigger than

50 so I upped it to 100 and everything was working fine. I don’t know if this

will be applicable to everyone or just me for being stupid but hey, it worked

;)

Moved to SQL server - seems off topic for the VS debugger.|||

I got the same error message because my dataset had columns with minoccur property as null but there were no records in the database. Some of these columns were nullable in the database but not marked as 0 in dataset minoccur property.

Solution - For every column thats marked as nullable in database, I set the minoccur property to 0. This makes dataser expect null values

Failed to enable constraints. One or more rows contain values violating non-null, unique, or for

Hello all, I just want to make a quick post on the “Failed to enable constraints. One or more rows contain values violating non-null, unique, or foreign-key constraints.”

I am working on a small system at the moment and I have a CODES table with columns CODE and DESC. When I first set up the database I set CODE to varchar(MAX) and same with DESC, Then when I put all the codes and descriptions in and tried to run fill I got the error.

Now weather this will help anyone I don’t know, I found out that it was the Colum DESC in the DataSet that was causing the problem, if you enter your dataset designer and select one of your columns, Right click, properties, there is a field called MAXLENGTH, When I checked it was set to 50 even though I had set varchar(MAX), so I knew that some of my descriptions where bigger than 50 so I upped it to 100 and everything was working fine. I don’t know if this will be applicable to everyone or just me for being stupid but hey, it worked ;)

Moved to SQL server - seems off topic for the VS debugger.|||

I got the same error message because my dataset had columns with minoccur property as null but there were no records in the database. Some of these columns were nullable in the database but not marked as 0 in dataset minoccur property.

Solution - For every column thats marked as nullable in database, I set the minoccur property to 0. This makes dataser expect null values

|||hello I'm new using c# I would like to know how to put the minoccur property to 0. Im having the same error
thanks

Failed to enable constraints. One or more rows contain values violating non-null, unique, or

Hello all, I just want to make a quick post on the “Failed

to enable constraints. One or more rows contain values violating non-null,

unique, or foreign-key constraints.”

I am working on a small system at the moment and I have a

CODES table with columns CODE and DESC. When I first set up the database I set CODE

to varchar(MAX) and same with DESC, Then when I put all the codes and

descriptions in and tried to run fill I got the error.

Now weather this will help anyone I don’t know, I found out

that it was the Colum DESC in the DataSet that was causing the problem, if you

enter your dataset designer and select one of your columns, Right click, properties,

there is a field called MAXLENGTH, When I checked it was set to 50 even though I

had set varchar(MAX), so I knew that some of my descriptions where bigger than

50 so I upped it to 100 and everything was working fine. I don’t know if this

will be applicable to everyone or just me for being stupid but hey, it worked

;)

Moved to SQL server - seems off topic for the VS debugger.|||

I got the same error message because my dataset had columns with minoccur property as null but there were no records in the database. Some of these columns were nullable in the database but not marked as 0 in dataset minoccur property.

Solution - For every column thats marked as nullable in database, I set the minoccur property to 0. This makes dataser expect null values

Extreme slow SQL Server 2005 and high CPU usage - Casting of values

Hello sql and .net gurus :-)
I have a problem with my websitewww.eventguide.it. It's completly developed under .NET 2 and SQL Server 2005 Express. My problem is the folowing:
The server is a Intel 3Ghz HT processor with 1GB Ram. No other page on the running system is a CPU consuming site. We optimized the SQL statements, the code, the caching and many other parts of the website (pooling on SQL access), but the SQL Server uses about 50% to100% of the CPU and about 400MB RAM all the time. The whole site seems to be very, very slow. In fact there are many of SQL operations on every page request, but we cache a lot of them in different ways (page output caching, application caching). So I don't understand we have so much performance problems. Any suggestions for optimised code in general? I read nearly all of the MS .NET performance papers - but real world experience is the missing part :-)

It is better to cast the values of a SQL reader like this
Dim String1 as String = Ctype(DataReader.item(0), String)
Dim Integer1 as Intger = Ctype(DataReader.item(1), Integer)
or like this
Dim String1 as String = DataReader.item(0)
Dim Integer1 as Intger = DataReader.item(1)

Thanks a lot for your help!
FOX

I visited your site, and honestly I cant answer your question. Your SQL CPU usage seems way off, but the problem appears to be bandwidth entirely.

I pinged your site, and traced it with good solid pathways and quick responses. But going into the Foto's (photos) section resulted in pictures taking an extremely long time to load even after the page was rendered. Why type of up-pipe is that server pushing?

|||What do you mean with: "Why type of up-pipe is that server pushing?"|||

Sorry, horribly formed question:

How much bandwidth is available to the server over your connection, dedicated to "upload"?

|||The upload of the server is about 10Mbit - enough to deliver requests in a fast way.
Well - look now the speed of the entire site. Today I changed the SQL database index of a 2.000.000 rows counting table (counter for images and other things) and from one moment to another I reduced CPU usage dramatically! Now I try to optimize also other table indexes to increase the speed further.
Any suggestions to handle indexes in a good way?
Thanks for your time!!!|||I had close to the same issue. I added the port to the IP addresss in the connection string. Don't ask me why this fixed the issue, but it did. For the RAM issue, I also had this. The server would go up to almost 500-600MB of ram.. and it barely had anything in it!!

So I made a batch script to restart the server, and setup a windows scheduled task to run it every other day.

Here is my batch:
net stop MSSQLSERVER
echo y
net start MSSQLSERVER

net stop hmailserver
net start hmailserver

Basically, this will restart your SQL server. I do this at about 3AM. It keeps the server running top notch as far as I know.

I had to add hmailserver in there because if you don't restart it also, it for some reason will fail to verify accounts. If you don't have hmail, remove it.|||Because if you don't specify the port, it first tries to contact the SQL Server on one of the standard ports, and if it's firewalled, or blocked, then it takes a bit for it to realize it and then switch over. Also, if you a fairly large latency between the web and sql server, it can reduce the time significantly because there is one less connection that needs to be established, and with a normal TCP/IP connection that does a 3-way handshake to start up, it's many times the latency between the two machines.

Extracting the parameter values

I have a report application where I am using Microsoft AxBrowser and url acess to view the reports. I have a report with multivalued parameter. I need to extract the user selected parameter value for that rpeort for some other purpose. How to extract the parameter value?

Please help

Do you have access to the cookies collection of the AxBrowser control? If so, then you can try to get the RS session ID and call the SOAP API GetExecutionInfo() on that session. It will return the effective parameters for the current session. Assuming you haven't changed them then it should be the parameters that were used to render the report.

Question: Why aren't you using the viewer control?

|||Thanks for the reply John.
I am passing the default parameters to the report and getting it.
I needed to extract the parameter values when the user inputs (selects) some other value.
I could able to retrieve the parameter values by capturing the url after navigation and used some string functions. I used the Navigate2 event of AXBroweser. This works for me becuase I had to extract values for only two reports and I know the names of the parameters
we have SQL2000 and Reporting Services2000. I guess we can use report viewr control only with Reporting Services 2005. Correct me if I am wrong.
Thanks,
Siri..
|||

Yes the viewer control only works with the 2005 endpoints.

Your solution will only work in the case where parameters are passed on the URL string, which in general is not the case. I would recommend trying to get the sessionId from the AxBrowser and calling GetExecutionInfo() as it should work regardless of how the parameters are set.

extracting the duplicate values

hi all
i want to extract only the duplicate values from a table
can any one know the sql syntax for that
thnking u
jagselect foo from yourtable
group by foo having count(*) > 1|||thnk u rudy thak u very much
i got it!!!!!!!!!!!!!!!

Extracting Parameter values from a Custom Data Processing Extension

I've created a custom Data Processing Extension and I've implemented the IDBCommandAnalysis interface so that my reports can enter parameters and pass them to my Data Processing Extension.

My question is, how do I extract the value from the Parameters coming from the report? Where do the parameters get passed off from the report? I can query the Parameters collection and my report gets prompted in Preview mode to enter something for the parameter but I can't find the spot where it gets passed for processing.

The IDBCommand interface contains a method CreateParameter(...).

For every query parameter, RS will call CreateParameter to add the parameter to the Parameters collection.

See also: http://msdn.microsoft.com/library/default.asp?url=/library/en-us/RSPROG/htm/rsp_ref_clr_dataproc_2a4a.asp

-- Robert

|||Where is the proper place to initialize the Parameter collection? I keep getting a "Collection was modified" when I try to tell my Command object was parameters it has.|||

Robert Bruckner MSFT wrote:

For every query parameter, RS will call CreateParameter to add the parameter to the Parameters collection.

Sorry for pullig out this old thread, but it's exactly the problem I have right now.
At the moment my CreateParameter()-Function looks like this (like in the FSI example)

Code Snippet

Return CType(New DataParameter, IDataParameter)

nothing more. As an extra, I am searching for parameters everytime the command-text is set. I am pullig out the parameters from the text with a regExp and add them to the collection.
Is this the right way?

In technet it says: "The CreateParameter method is called by the report server to set the name and value for each query parameter in the command text. The parameter is then added to the Parameters collection prior to the ExecuteReader method being called."

Right now, I am doing all of that in my cde manually:
set command-text? then get parameters from string -> add all of them to the collection.

without this extra-step after setting the command-text, the collection gets called later and would be empty. There is no call from report-server to CreateParameter. (I put traces in every function/sub)|||Ok, I worked it out:

"Public ReadOnly Property Parameters() As IDataParameterCollection Implements IDbCommand.Parameters"
This one is used by the server to access and build up the parameterCollection.

"Public Function GetParameters() As IDataParameterCollection Implements IDbCommandAnalysis.GetParameters"
This one has to parse out all parameters from the query, so the dataset preview can list them.
http://technet.microsoft.com/en-us/library/aa256673(SQL.80).aspx

Extracting Parameter values from a Custom Data Processing Extension

I've created a custom Data Processing Extension and I've implemented the IDBCommandAnalysis interface so that my reports can enter parameters and pass them to my Data Processing Extension.

My question is, how do I extract the value from the Parameters coming from the report? Where do the parameters get passed off from the report? I can query the Parameters collection and my report gets prompted in Preview mode to enter something for the parameter but I can't find the spot where it gets passed for processing.

The IDBCommand interface contains a method CreateParameter(...).

For every query parameter, RS will call CreateParameter to add the parameter to the Parameters collection.

See also: http://msdn.microsoft.com/library/default.asp?url=/library/en-us/RSPROG/htm/rsp_ref_clr_dataproc_2a4a.asp

-- Robert

|||Where is the proper place to initialize the Parameter collection? I keep getting a "Collection was modified" when I try to tell my Command object was parameters it has.
|||

Robert Bruckner MSFT wrote:

For every query parameter, RS will call CreateParameter to add the parameter to the Parameters collection.

Sorry for pullig out this old thread, but it's exactly the problem I have right now.
At the moment my CreateParameter()-Function looks like this (like in the FSI example)

Code Snippet

Return CType(New DataParameter, IDataParameter)

nothing more. As an extra, I am searching for parameters everytime the command-text is set. I am pullig out the parameters from the text with a regExp and add them to the collection.
Is this the right way?

In technet it says: "The CreateParameter method is called by the report server to set the name and value for each query parameter in the command text. The parameter is then added to the Parameters collection prior to the ExecuteReader method being called."

Right now, I am doing all of that in my cde manually:
set command-text? then get parameters from string -> add all of them to the collection.

without this extra-step after setting the command-text, the collection gets called later and would be empty. There is no call from report-server to CreateParameter. (I put traces in every function/sub)
|||Ok, I worked it out:

"Public ReadOnly Property Parameters() As IDataParameterCollection Implements IDbCommand.Parameters"
This one is used by the server to access and build up the parameterCollection.

"Public Function GetParameters() As IDataParameterCollection Implements IDbCommandAnalysis.GetParameters"
This one has to parse out all parameters from the query, so the dataset preview can list them.
http://technet.microsoft.com/en-us/library/aa256673(SQL.80).aspx

Extracting multiple values as comma seperated

Hi all,

This is my table :

WorkstationNo UserID

101 a1

102 a2

103 a3

101 a2

and there are many other fields too. The output I need is something like this

WorsktationNo OccupiedBy

101 a1,a2

102 a2

103 a3

In the similar fashion I would also require to retrieve the values based on the UserID something like this

UserID Workstations

a1 101

a2 101,102

a3 103

Could someone tell me how to write the query for this.

Here ya go

Code Snippet

create table #t (WorkstationNo int, UserID varchar(10) )

insert into #t

select 101, 'a1'

union all select 102, 'a2'

union all select 103, 'a3'

union all select 101, 'a2'

select WorkstationNo,

substring( ( SELECT ', ' + UserID AS [text()]

FROM #t t2

WHERE t1.WorkstationNo = t2.WorkstationNo

FOR XML path(''), elements ), 2, 1000) as OccupiedBy

from #t t1

group by WorkstationNo

select UserID,

substring( ( SELECT ', ' + convert( varchar(10), WorkstationNo ) AS [text()]

FROM #t t2

WHERE t1.UserID = t2.UserID

FOR XML path(''), elements ), 2, 1000) as OccupiedBy

from #t t1

group by UserID

|||

Thanks DaleJ..

worked like a charm. I guess i will take time to understand the code...

extracting data from a varchar variable?

Guys.

I have some data like this one:

'Uploadfiles\CompanyID\ProjectCode\Files\File1.doc'

P.S: the values may vary each time.

I need to programmatically extract from that kind of string only file1.doc

Ho can I dynamically do it? any function? statement? etc? please help.

The idea is getting only the string after the last "\"

Thanks
Frdeclare @.t varchar(256)

select @.t = 'Uploadfiles\CompanyID\ProjectCode\Files\File1.doc'

select reverse(substring(reverse(@.t),1,patindex('%\%',reverse(@.t))-1))

extracting csv string and using in NOTIN clause

Hi,
I'm trying to pass a stored proc a list of values in a comma seperated strin
g:
("3125,3126,3129...") The values represent an integer column in a table.
Then I want to use the string to exclude the values from a SELECT as in:
CREATE PROCEDURE myProc
@.transactionIDs varchar(255)
AS
SELECT TransactionID FROM SalesTransactions
WHERE TransactionID NOT IN (@.transactionIDs)
GO
SQL Server complains that it can't cast a varchar to an int
Any suggestions on how to do this?
Can someone provide an example?
ThanksWhile someone might suggest you use dynamic SQL for this, I
wouldn't. Instead, consider either this not-so efficient solution:
select TransactionID
from SalesTransactions
where
charindex(
',' + rtrim(TransactionID) + ',',
',' + @.transactionIDs + ','
) = 0
-- be sure @.transactionIDs has no spaces
or, for greater efficiency, the more involved solution
that uses a table-valued function to convert @.transactionIDs
into a table of id values:
http://groups.google.co.uk/groups?q...wierzbicki+kass
You can then write:
select TransactionID
from SalesTransactions
where TransactionID not in (
select Item from ListTable(@.transactionIDs)
)
Steve Kass
Drew University
Opa wrote:

>Hi,
>I'm trying to pass a stored proc a list of values in a comma seperated stri
ng:
>("3125,3126,3129...") The values represent an integer column in a table.
>Then I want to use the string to exclude the values from a SELECT as in:
>CREATE PROCEDURE myProc
> @.transactionIDs varchar(255)
>AS
>SELECT TransactionID FROM SalesTransactions
>WHERE TransactionID NOT IN (@.transactionIDs)
>GO
>SQL Server complains that it can't cast a varchar to an int
>
>Any suggestions on how to do this?
>Can someone provide an example?
>Thanks
>
>
>

Extracting Attribute Values from XML Using XQuery

I am serializing a .NET object as XML and passing it as a parameter to a stored procedure. I am new to XQuery and have tried other places to find the answer to this question but have so far been unsuccessful.

How can I get the value of an attribute out of an XML document as a standard SQL Server data type? For instance, if my XML document were:

<RootNode>
<Object ID = "1234">
<AnElement>Some Other Value</AnElement>
</Object>
</RootNode>

What would be the correct XQuery to get the value of the ID attribute in the Object element?

Thanks in advance.

Kevin J Lambert

create table #xml_temp (

xml_col xml

)

insert into #xml_temp values('<RootNode>

<Object ID = "1234">

<AnElement>Some Other Value</AnElement>

</Object>

</RootNode>')

--With query() function

select xml_col.query('data(/RootNode/Object/@.ID)') from #xml_temp

--With value() function

select xml_col.value('data(/RootNode/Object/@.ID)[1]','int') from #xml_temp

Extract values from XML in a Strored Proc

I pass an xml to a Stored Proc similar to below.
How do I extaract the id and value from below XML?
DECLARE @.piMnemonicIDs xml
SET @.piMnemonicIDs =
'<MnemonicIDs><id>1</id><value>Costs</value><id>2</id><value>MarketPowerInfluence</value><value>IndustryStanding</value><value>SupplierRelations</value></MnemonicIDs>'
SELECT T.MnemonicIDs.value('/value', 'VARCHAR(50)'),
T.MnemonicIDs.value('/id', 'VARCHAR(50)')
FROM @.piMnemonicIDs.nodes('/MnemonicIDs') AS T(MnemonicIDs)
You probably want:
DECLARE @.piMnemonicIDs xml
SET @.piMnemonicIDs =
'<MnemonicIDs>
<id>1</id><value>Costs</value>
<id>2</id><value>MarketPowerInfluence</value><value>IndustryStanding</value>
<value>SupplierRelations</value>
</MnemonicIDs>'
SELECT T.MnemonicIDs.value('.', 'VARCHAR(50)'),
T.MnemonicIDs.value('(for $v in . return ../id[. << $v])[last()]',
'VARCHAR(50)')
FROM @.piMnemonicIDs.nodes('/MnemonicIDs/value') AS T(MnemonicIDs)
However, if you can change the XML structure to something like:
DECLARE @.piMnemonicIDs xml
SET @.piMnemonicIDs =
'<MnemonicIDs>
<id>1</id><value>Costs</value>
</MnemonicIDs>
<MnemonicIDs>
<id>2</id><value>MarketPowerInfluence</value>
</MnemonicIDs>
<MnemonicIDs>
<id>2</id><value>IndustryStanding</value>
</MnemonicIDs>
<MnemonicIDs>
<id>2</id><value>SupplierRelations</value>
</MnemonicIDs>'
then the query becomes simpler:
SELECT T.MnemonicIDs.value('value[1]', 'VARCHAR(50)'),
T.MnemonicIDs.value('id[1]', 'VARCHAR(50)')
FROM @.piMnemonicIDs.nodes('/MnemonicIDs') AS T(MnemonicIDs)
Best regards
Michael
"C" <C@.discussions.microsoft.com> wrote in message
news:BCFF9712-A442-42E9-9F7F-E686F65137D0@.microsoft.com...
>I pass an xml to a Stored Proc similar to below.
> How do I extaract the id and value from below XML?
>
> DECLARE @.piMnemonicIDs xml
> SET @.piMnemonicIDs =
> '<MnemonicIDs><id>1</id><value>Costs</value><id>2</id><value>MarketPowerInfluence</value><value>IndustryStanding</value><value>SupplierRelations</value></MnemonicIDs>'
> SELECT T.MnemonicIDs.value('/value', 'VARCHAR(50)'),
> T.MnemonicIDs.value('/id', 'VARCHAR(50)')
> FROM @.piMnemonicIDs.nodes('/MnemonicIDs') AS T(MnemonicIDs)
>

Extract values from XML in a Strored Proc

I pass an xml to a Stored Proc similar to below.
How do I extaract the id and value from below XML?
DECLARE @.piMnemonicIDs xml
SET @.piMnemonicIDs =
'<MnemonicIDs><id>1</id><value>Costs</value><id>2</id><value>MarketPowerInfl
uence</value><value>IndustryStanding</value><value>SupplierRelations</value>
</MnemonicIDs>'
SELECT T.MnemonicIDs.value('/value', 'VARCHAR(50)'),
T.MnemonicIDs.value('/id', 'VARCHAR(50)')
FROM @.piMnemonicIDs.nodes('/MnemonicIDs') AS T(MnemonicIDs)You probably want:
DECLARE @.piMnemonicIDs xml
SET @.piMnemonicIDs =
'<MnemonicIDs>
<id>1</id><value>Costs</value>
<id>2</id><value>MarketPowerInfluence</value><value>IndustryStanding</value>
<value>SupplierRelations</value>
</MnemonicIDs>'
SELECT T.MnemonicIDs.value('.', 'VARCHAR(50)'),
T.MnemonicIDs.value('(for $v in . return ../id[. << $v])[last()]',
'VARCHAR(50)')
FROM @.piMnemonicIDs.nodes('/MnemonicIDs/value') AS T(MnemonicIDs)
However, if you can change the XML structure to something like:
DECLARE @.piMnemonicIDs xml
SET @.piMnemonicIDs =
'<MnemonicIDs>
<id>1</id><value>Costs</value>
</MnemonicIDs>
<MnemonicIDs>
<id>2</id><value>MarketPowerInfluence</value>
</MnemonicIDs>
<MnemonicIDs>
<id>2</id><value>IndustryStanding</value>
</MnemonicIDs>
<MnemonicIDs>
<id>2</id><value>SupplierRelations</value>
</MnemonicIDs>'
then the query becomes simpler:
SELECT T.MnemonicIDs.value('value[1]', 'VARCHAR(50)'),
T.MnemonicIDs.value('id[1]', 'VARCHAR(50)')
FROM @.piMnemonicIDs.nodes('/MnemonicIDs') AS T(MnemonicIDs)
Best regards
Michael
"C" <C@.discussions.microsoft.com> wrote in message
news:BCFF9712-A442-42E9-9F7F-E686F65137D0@.microsoft.com...
>I pass an xml to a Stored Proc similar to below.
> How do I extaract the id and value from below XML?
>
> DECLARE @.piMnemonicIDs xml
> SET @.piMnemonicIDs =
> '<MnemonicIDs><id>1</id><value>Costs</value><id>2</id><value>MarketPowerIn
fluence</value><value>IndustryStanding</value><value>SupplierRelations</valu
e></MnemonicIDs>'
> SELECT T.MnemonicIDs.value('/value', 'VARCHAR(50)'),
> T.MnemonicIDs.value('/id', 'VARCHAR(50)')
> FROM @.piMnemonicIDs.nodes('/MnemonicIDs') AS T(MnemonicIDs)
>

Extract values from XML column (rows) to a rowset

Hi,
I have a question regarding extracting XML data as relational data. From
the looks of it looks like the OPENXML statement is not relevant since the
data I need to extract resides in XML columns (in a table).
So I have this query
SELECT [MyXmlCol] FROM [MyTable] WHERE [XmlTypeName]='Schema1'
MyXmlCol – is an XML column, and the WHERE clause assures me that I know h
ow
to access this XML data (using xpath)
Let’s say the data in XML column looks like this
<MyStuff>
<Product>Windows Vista</Product>
<Version>1</Version>
</MyStuff>
(were in each row the value of the Product & Version nodes are different).
I want a table/rowset with two columns Product (xpath
/MyStuff/Product/text()) and Version (xpath /MyStuff/Version/text())
I want the rowset/table to look like this:
Product Version
-- --
WindowsVista 1
WindowsHome 2
WindowsXP 3
How do I accomplish this ?
Asher.Asher
Take a look at this example to resturn Cities
declare @.tAddress table (Address xml DEFAULT '<Address Record />' )
INSERT INTO @.tAddress (Address)
VALUES ('<AddressType name="Shipping">
<AddressRecord StreetLine1="111 Main"
City="Houston"
State="TX"
ZipCode = "11111" />
</AddressType> ');
INSERT INTO @.tAddress (Address)
VALUES ('<AddressType name="Shipping">
<AddressRecord StreetLine1="222 Main"
City="SanDiego"
State="CA"
ZipCode = "2222" />
</AddressType> ');
SELECT address.value('(/AddressType/AddressRecord/@.City)[1]',
'varchar(50)' )
FROM @.tAddress
"Asher F." <asherfoa@.community.nospam> wrote in message
news:407A573D-01B4-4AEE-A10B-FA2CB5AA0A17@.microsoft.com...
> Hi,
> I have a question regarding extracting XML data as relational data. From
> the looks of it looks like the OPENXML statement is not relevant since the
> data I need to extract resides in XML columns (in a table).
> So I have this query
> SELECT [MyXmlCol] FROM [MyTable] WHERE [XmlTypeName]='Schema1'
> MyXmlCol is an XML column, and the WHERE clause assures me that I know
> how
> to access this XML data (using xpath)
> Lets say the data in XML column looks like this
> <MyStuff>
> <Product>Windows Vista</Product>
> <Version>1</Version>
> </MyStuff>
> (were in each row the value of the Product & Version nodes are different).
> I want a table/rowset with two columns Product (xpath
> /MyStuff/Product/text()) and Version (xpath /MyStuff/Version/text())
> I want the rowset/table to look like this:
> Product Version
> -- --
> WindowsVista 1
> WindowsHome 2
> WindowsXP 3
> How do I accomplish this ?
>
> --
> Asher.|||Hello Asher,
I understand that you'd like to insert a table from a xml type value. You
may want to use openxml to do the job:
CREATE TABLE MyStuff( product varchar(20), version int)
go
DECLARE @.docHandle int
declare @.xmlDocument xml
set @.xmlDocument = N'
<MyStuff>
<Product>Windows Vista</Product>
<Version>1</Version>
</MyStuff>
'
EXEC sp_xml_preparedocument @.docHandle OUTPUT, @.xmlDocument
-- Use OPENXML to provide rowset consisting of customer data.
INSERT Mystuff
SELECT *
FROM OPENXML(@.docHandle, N'/MyStuff', 2)
WITH (proudct varchar(20) '/MyStuff/Product',
version int '/MyStuff/Version')
EXEC sp_xml_removedocument @.docHandle
select * from mystuff
If you have any comments or feedback, please feel free to let's know. Thank
you.
Best Regards,
Peter Yang
MCSE2000/2003, MCSA, MCDBA
Microsoft Online Community Support
========================================
==========
Get notification to my posts through email? Please refer to
http://msdn.microsoft.com/subscript...ault.aspx#notif
ications
<http://msdn.microsoft.com/subscript...ps/default.aspx>.
Note: The MSDN Managed Newsgroup support offering is for non-urgent issues
where an initial response from the community or a Microsoft Support
Engineer within 1 business day is acceptable. Please note that each follow
up response may take approximately 2 business days as the support
professional working with you may need further investigation to reach the
most efficient resolution. The offering is not appropriate for situations
that require urgent, real-time or phone-based interactions or complex
project analysis and dump analysis issues. Issues of this nature are best
handled working with a dedicated Microsoft Support Engineer by contacting
Microsoft Customer Support Services (CSS) at
<http://msdn.microsoft.com/subscript...rt/default.aspx>.
========================================
==========
This posting is provided "AS IS" with no warranties, and confers no rights.|||Hi Uri,
Thanks for the quick response, I understand the code you suggested and it
works for me 80% of times. I would very much appreciate your help in this
follow-up.
To extend your example, let’s say I have this:
declare @.tAddress table (Address xml DEFAULT '<Address Record />' )
INSERT INTO @.tAddress (Address)
VALUES ('<AddressType name="Shipping">
<AddressRecord StreetLine1="111 Main"
City="Houston"
State="TX"
ZipCode = "11111" />
<AddressRecord StreetLine1="222 Left"
City="New York"
State="NY"
ZipCode = "2222" />
</AddressType> ');
INSERT INTO @.tAddress (Address)
VALUES ('<AddressType name="Shipping">
<AddressRecord StreetLine1="222 Main"
City="SanDiego"
State="CA"
ZipCode = "2222" />
</AddressType> ');
Note that the XML in the first row has two AddressRecord elements.
How do I make my result return a rowset with 3 rows representing all the
values:
SanDiego
New York
Houston
Thanks in advance!
--
Asher.
"Uri Dimant" wrote:

> Asher
> Take a look at this example to resturn Cities
> declare @.tAddress table (Address xml DEFAULT '<Address Record />' )
> INSERT INTO @.tAddress (Address)
> VALUES ('<AddressType name="Shipping">
> <AddressRecord StreetLine1="111 Main"
> City="Houston"
> State="TX"
> ZipCode = "11111" />
> </AddressType> ');
> INSERT INTO @.tAddress (Address)
> VALUES ('<AddressType name="Shipping">
> <AddressRecord StreetLine1="222 Main"
> City="SanDiego"
> State="CA"
> ZipCode = "2222" />
> </AddressType> ');
> SELECT address.value('(/AddressType/AddressRecord/@.City)[1]',
> 'varchar(50)' )
> FROM @.tAddress
>
> "Asher F." <asherfoa@.community.nospam> wrote in message
> news:407A573D-01B4-4AEE-A10B-FA2CB5AA0A17@.microsoft.com...
>
>|||Hello Peter,
Thank you for your response.
What I am trying to do is to extract data from an XML column I have in one
of my tables. I am trying to make the data in the XML column appear as part
of the normal rowset.
I will run a SQL SELECT statement from my C# code expecting to receive the
data back in the form of a rowset (so I can read it as primitive types using
IDataReader).
Anyway, Uri's response works for me in most of the cases, it doesn't cover
(and I tried several things myself) the case where I need to extract multipl
e
values (i.e. rows) from a single XML document (stored in the xml column in a
given row in my table).
Thanks for the help
Asher.
"Peter Yang [MSFT]" wrote:

> Hello Asher,
> I understand that you'd like to insert a table from a xml type value. You
> may want to use openxml to do the job:
> CREATE TABLE MyStuff( product varchar(20), version int)
> go
> DECLARE @.docHandle int
> declare @.xmlDocument xml
> set @.xmlDocument = N'
> <MyStuff>
> <Product>Windows Vista</Product>
> <Version>1</Version>
> </MyStuff>
> '
> EXEC sp_xml_preparedocument @.docHandle OUTPUT, @.xmlDocument
> -- Use OPENXML to provide rowset consisting of customer data.
> INSERT Mystuff
> SELECT *
> FROM OPENXML(@.docHandle, N'/MyStuff', 2)
> WITH (proudct varchar(20) '/MyStuff/Product',
> version int '/MyStuff/Version')
> EXEC sp_xml_removedocument @.docHandle
> select * from mystuff
>
> If you have any comments or feedback, please feel free to let's know. Than
k
> you.
> Best Regards,
> Peter Yang
> MCSE2000/2003, MCSA, MCDBA
> Microsoft Online Community Support
> ========================================
==========
> Get notification to my posts through email? Please refer to
> l]
> ications
> <[url]http://msdn.microsoft.com/subscriptions/managednewsgroups/default.aspx" target="_blank">http://msdn.microsoft.com/subscript...ps/default.aspx>.
> Note: The MSDN Managed Newsgroup support offering is for non-urgent issues
> where an initial response from the community or a Microsoft Support
> Engineer within 1 business day is acceptable. Please note that each follow
> up response may take approximately 2 business days as the support
> professional working with you may need further investigation to reach the
> most efficient resolution. The offering is not appropriate for situations
> that require urgent, real-time or phone-based interactions or complex
> project analysis and dump analysis issues. Issues of this nature are best
> handled working with a dedicated Microsoft Support Engineer by contacting
> Microsoft Customer Support Services (CSS) at
> <http://msdn.microsoft.com/subscript...rt/default.aspx>.
> ========================================
==========
> This posting is provided "AS IS" with no warranties, and confers no rights
.
>|||Asher
I see what you mean
DECLARE @.idoc int
DECLARE @.doc varchar(1000)
SET @.doc ='
<AddressType name="Shipping">
<AddressRecord StreetLine1="111 Main"
City="Houston"
State="TX"
ZipCode = "11111" />
<AddressRecord StreetLine1="111 Left"
City="New_York"
State="NY"
ZipCode = "3333" />
</AddressType> '
EXEC sp_xml_preparedocument @.idoc OUTPUT, @.doc
SELECT *
FROM OPENXML (@.idoc, '/AddressType/AddressRecord',1)
WITH (City varchar(10))
"Asher F." <asherfoa@.community.nospam> wrote in message
news:F678FA42-605F-4425-BED3-EBEF75A47219@.microsoft.com...[vbcol=seagreen]
> Hi Uri,
> Thanks for the quick response, I understand the code you suggested and it
> works for me 80% of times. I would very much appreciate your help in this
> follow-up.
> To extend your example, lets say I have this:
> declare @.tAddress table (Address xml DEFAULT '<Address Record />' )
> INSERT INTO @.tAddress (Address)
> VALUES ('<AddressType name="Shipping">
> <AddressRecord StreetLine1="111 Main"
> City="Houston"
> State="TX"
> ZipCode = "11111" />
> <AddressRecord StreetLine1="222 Left"
> City="New York"
> State="NY"
> ZipCode = "2222" />
> </AddressType> ');
> INSERT INTO @.tAddress (Address)
> VALUES ('<AddressType name="Shipping">
> <AddressRecord StreetLine1="222 Main"
> City="SanDiego"
> State="CA"
> ZipCode = "2222" />
> </AddressType> ');
> Note that the XML in the first row has two AddressRecord elements.
> How do I make my result return a rowset with 3 rows representing all the
> values:
> SanDiego
> New York
> Houston
> Thanks in advance!
> --
> Asher.
>
> "Uri Dimant" wrote:
>|||Or
SELECT address.value('(/AddressType/AddressRecord/@.City)[1]', 'varchar(5
0)')
FROM @.tAddress
UNION ALL
SELECT address.value('(/AddressType/AddressRecord/@.City)[2]', 'varchar(5
0)')
FROM @.tAddress
"Uri Dimant" <urid@.iscar.co.il> wrote in message
news:O9Qv3C1THHA.5108@.TK2MSFTNGP06.phx.gbl...
> Asher
> I see what you mean
> DECLARE @.idoc int
> DECLARE @.doc varchar(1000)
> SET @.doc ='
> <AddressType name="Shipping">
> <AddressRecord StreetLine1="111 Main"
> City="Houston"
> State="TX"
> ZipCode = "11111" />
> <AddressRecord StreetLine1="111 Left"
> City="New_York"
> State="NY"
> ZipCode = "3333" />
> </AddressType> '
> EXEC sp_xml_preparedocument @.idoc OUTPUT, @.doc
> SELECT *
> FROM OPENXML (@.idoc, '/AddressType/AddressRecord',1)
> WITH (City varchar(10))
>
>
>
> "Asher F." <asherfoa@.community.nospam> wrote in message
> news:F678FA42-605F-4425-BED3-EBEF75A47219@.microsoft.com...
>|||Hi Uri,
The OPENXML is probably not the solution since I have the XML document
stored in a column.
As for your second solution, it looks like it is in the right direction,
however I don't see how I can scale it to the case I have N (or an unkown
number of) AddressRecord nodes in each XML document (i.e. one row's XML
column might contain zero AddressRecord elements and another row might
contain 10...)
How do I scale this solution ?
Asher.
"Uri Dimant" wrote:

> Or
> SELECT address.value('(/AddressType/AddressRecord/@.City)[1]', 'varchar
(50)')
> FROM @.tAddress
> UNION ALL
> SELECT address.value('(/AddressType/AddressRecord/@.City)[2]', 'varchar
(50)')
> FROM @.tAddress
>
>
>
> "Uri Dimant" <urid@.iscar.co.il> wrote in message
> news:O9Qv3C1THHA.5108@.TK2MSFTNGP06.phx.gbl...
>
>|||Check out the nodes method. It returns a table with one column of datatype X
ML. So if the XML
instance is a column inside a table, you use this with APPLY against the tab
le where the column
exists. Example, assuming table is temp table instead of table variable:
SELECT nc.value('@.City[1]', 'varchar(39)')
FROM #tAddress
CROSS APPLY Address.nodes('/AddressType/AddressRecord') AS nt(nc)
---
Houston
New York
SanDiego
--
Tibor Karaszi, SQL Server MVP
http://www.karaszi.com/sqlserver/default.asp
http://www.solidqualitylearning.com/
"Asher F." <asherfoa@.community.nospam> wrote in message
news:F678FA42-605F-4425-BED3-EBEF75A47219@.microsoft.com...[vbcol=seagreen]
> Hi Uri,
> Thanks for the quick response, I understand the code you suggested and it
> works for me 80% of times. I would very much appreciate your help in this
> follow-up.
> To extend your example, let's say I have this:
> declare @.tAddress table (Address xml DEFAULT '<Address Record />' )
> INSERT INTO @.tAddress (Address)
> VALUES ('<AddressType name="Shipping">
> <AddressRecord StreetLine1="111 Main"
> City="Houston"
> State="TX"
> ZipCode = "11111" />
> <AddressRecord StreetLine1="222 Left"
> City="New York"
> State="NY"
> ZipCode = "2222" />
> </AddressType> ');
> INSERT INTO @.tAddress (Address)
> VALUES ('<AddressType name="Shipping">
> <AddressRecord StreetLine1="222 Main"
> City="SanDiego"
> State="CA"
> ZipCode = "2222" />
> </AddressType> ');
> Note that the XML in the first row has two AddressRecord elements.
> How do I make my result return a rowset with 3 rows representing all the
> values:
> SanDiego
> New York
> Houston
> Thanks in advance!
> --
> Asher.
>
> "Uri Dimant" wrote:
>|||See Tibor's reply , it is what you wanted
"Asher F." <asherfoa@.community.nospam> wrote in message
news:38DFF454-06D5-4E00-B623-F5FB216D87EC@.microsoft.com...[vbcol=seagreen]
> Hi Uri,
> The OPENXML is probably not the solution since I have the XML document
> stored in a column.
> As for your second solution, it looks like it is in the right direction,
> however I don't see how I can scale it to the case I have N (or an unkown
> number of) AddressRecord nodes in each XML document (i.e. one row's XML
> column might contain zero AddressRecord elements and another row might
> contain 10...)
> How do I scale this solution ?
> --
> Asher.
>
> "Uri Dimant" wrote:
>

Extract values from XML column (rows) to a rowset

Hi,
I have a question regarding extracting XML data as relational data. From
the looks of it looks like the OPENXML statement is not relevant since the
data I need to extract resides in XML columns (in a table).
So I have this query
SELECT [MyXmlCol] FROM [MyTable] WHERE [XmlTypeName]='Schema1'
MyXmlCol â' is an XML column, and the WHERE clause assures me that I know how
to access this XML data (using xpath)
Letâ's say the data in XML column looks like this
<MyStuff>
<Product>Windows Vista</Product>
<Version>1</Version>
</MyStuff>
(were in each row the value of the Product & Version nodes are different).
I want a table/rowset with two columns Product (xpath
/MyStuff/Product/text()) and Version (xpath /MyStuff/Version/text())
I want the rowset/table to look like this:
Product Version
-- --
WindowsVista 1
WindowsHome 2
WindowsXP 3
How do I accomplish this ?
--
Asher.Asher
Take a look at this example to resturn Cities
declare @.tAddress table (Address xml DEFAULT '<Address Record />' )
INSERT INTO @.tAddress (Address)
VALUES ('<AddressType name="Shipping">
<AddressRecord StreetLine1="111 Main"
City="Houston"
State="TX"
ZipCode = "11111" />
</AddressType> ');
INSERT INTO @.tAddress (Address)
VALUES ('<AddressType name="Shipping">
<AddressRecord StreetLine1="222 Main"
City="SanDiego"
State="CA"
ZipCode = "2222" />
</AddressType> ');
SELECT address.value('(/AddressType/AddressRecord/@.City)[1]',
'varchar(50)' )
FROM @.tAddress
"Asher F." <asherfoa@.community.nospam> wrote in message
news:407A573D-01B4-4AEE-A10B-FA2CB5AA0A17@.microsoft.com...
> Hi,
> I have a question regarding extracting XML data as relational data. From
> the looks of it looks like the OPENXML statement is not relevant since the
> data I need to extract resides in XML columns (in a table).
> So I have this query
> SELECT [MyXmlCol] FROM [MyTable] WHERE [XmlTypeName]='Schema1'
> MyXmlCol ? is an XML column, and the WHERE clause assures me that I know
> how
> to access this XML data (using xpath)
> Let?s say the data in XML column looks like this
> <MyStuff>
> <Product>Windows Vista</Product>
> <Version>1</Version>
> </MyStuff>
> (were in each row the value of the Product & Version nodes are different).
> I want a table/rowset with two columns Product (xpath
> /MyStuff/Product/text()) and Version (xpath /MyStuff/Version/text())
> I want the rowset/table to look like this:
> Product Version
> -- --
> WindowsVista 1
> WindowsHome 2
> WindowsXP 3
> How do I accomplish this ?
>
> --
> Asher.|||Hello Asher,
I understand that you'd like to insert a table from a xml type value. You
may want to use openxml to do the job:
CREATE TABLE MyStuff( product varchar(20), version int)
go
DECLARE @.docHandle int
declare @.xmlDocument xml
set @.xmlDocument = N'
<MyStuff>
<Product>Windows Vista</Product>
<Version>1</Version>
</MyStuff>
'
EXEC sp_xml_preparedocument @.docHandle OUTPUT, @.xmlDocument
-- Use OPENXML to provide rowset consisting of customer data.
INSERT Mystuff
SELECT *
FROM OPENXML(@.docHandle, N'/MyStuff', 2)
WITH (proudct varchar(20) '/MyStuff/Product',
version int '/MyStuff/Version')
EXEC sp_xml_removedocument @.docHandle
select * from mystuff
If you have any comments or feedback, please feel free to let's know. Thank
you.
Best Regards,
Peter Yang
MCSE2000/2003, MCSA, MCDBA
Microsoft Online Community Support
==================================================Get notification to my posts through email? Please refer to
http://msdn.microsoft.com/subscriptions/managednewsgroups/default.aspx#notif
ications
<http://msdn.microsoft.com/subscriptions/managednewsgroups/default.aspx>.
Note: The MSDN Managed Newsgroup support offering is for non-urgent issues
where an initial response from the community or a Microsoft Support
Engineer within 1 business day is acceptable. Please note that each follow
up response may take approximately 2 business days as the support
professional working with you may need further investigation to reach the
most efficient resolution. The offering is not appropriate for situations
that require urgent, real-time or phone-based interactions or complex
project analysis and dump analysis issues. Issues of this nature are best
handled working with a dedicated Microsoft Support Engineer by contacting
Microsoft Customer Support Services (CSS) at
<http://msdn.microsoft.com/subscriptions/support/default.aspx>.
==================================================This posting is provided "AS IS" with no warranties, and confers no rights.|||Hi Uri,
Thanks for the quick response, I understand the code you suggested and it
works for me 80% of times. I would very much appreciate your help in this
follow-up.
To extend your example, letâ's say I have this:
declare @.tAddress table (Address xml DEFAULT '<Address Record />' )
INSERT INTO @.tAddress (Address)
VALUES ('<AddressType name="Shipping">
<AddressRecord StreetLine1="111 Main"
City="Houston"
State="TX"
ZipCode = "11111" />
<AddressRecord StreetLine1="222 Left"
City="New York"
State="NY"
ZipCode = "2222" />
</AddressType> ');
INSERT INTO @.tAddress (Address)
VALUES ('<AddressType name="Shipping">
<AddressRecord StreetLine1="222 Main"
City="SanDiego"
State="CA"
ZipCode = "2222" />
</AddressType> ');
Note that the XML in the first row has two AddressRecord elements.
How do I make my result return a rowset with 3 rows representing all the
values:
SanDiego
New York
Houston
Thanks in advance!
--
Asher.
"Uri Dimant" wrote:
> Asher
> Take a look at this example to resturn Cities
> declare @.tAddress table (Address xml DEFAULT '<Address Record />' )
> INSERT INTO @.tAddress (Address)
> VALUES ('<AddressType name="Shipping">
> <AddressRecord StreetLine1="111 Main"
> City="Houston"
> State="TX"
> ZipCode = "11111" />
> </AddressType> ');
> INSERT INTO @.tAddress (Address)
> VALUES ('<AddressType name="Shipping">
> <AddressRecord StreetLine1="222 Main"
> City="SanDiego"
> State="CA"
> ZipCode = "2222" />
> </AddressType> ');
> SELECT address.value('(/AddressType/AddressRecord/@.City)[1]',
> 'varchar(50)' )
> FROM @.tAddress
>
> "Asher F." <asherfoa@.community.nospam> wrote in message
> news:407A573D-01B4-4AEE-A10B-FA2CB5AA0A17@.microsoft.com...
> > Hi,
> > I have a question regarding extracting XML data as relational data. From
> > the looks of it looks like the OPENXML statement is not relevant since the
> > data I need to extract resides in XML columns (in a table).
> > So I have this query
> > SELECT [MyXmlCol] FROM [MyTable] WHERE [XmlTypeName]='Schema1'
> > MyXmlCol â' is an XML column, and the WHERE clause assures me that I know
> > how
> > to access this XML data (using xpath)
> > Letâ's say the data in XML column looks like this
> > <MyStuff>
> > <Product>Windows Vista</Product>
> > <Version>1</Version>
> > </MyStuff>
> > (were in each row the value of the Product & Version nodes are different).
> > I want a table/rowset with two columns Product (xpath
> > /MyStuff/Product/text()) and Version (xpath /MyStuff/Version/text())
> > I want the rowset/table to look like this:
> > Product Version
> > -- --
> > WindowsVista 1
> > WindowsHome 2
> > WindowsXP 3
> >
> > How do I accomplish this ?
> >
> >
> >
> > --
> > Asher.
>
>|||Hello Peter,
Thank you for your response.
What I am trying to do is to extract data from an XML column I have in one
of my tables. I am trying to make the data in the XML column appear as part
of the normal rowset.
I will run a SQL SELECT statement from my C# code expecting to receive the
data back in the form of a rowset (so I can read it as primitive types using
IDataReader).
Anyway, Uri's response works for me in most of the cases, it doesn't cover
(and I tried several things myself) the case where I need to extract multiple
values (i.e. rows) from a single XML document (stored in the xml column in a
given row in my table).
Thanks for the help
--
Asher.
"Peter Yang [MSFT]" wrote:
> Hello Asher,
> I understand that you'd like to insert a table from a xml type value. You
> may want to use openxml to do the job:
> CREATE TABLE MyStuff( product varchar(20), version int)
> go
> DECLARE @.docHandle int
> declare @.xmlDocument xml
> set @.xmlDocument = N'
> <MyStuff>
> <Product>Windows Vista</Product>
> <Version>1</Version>
> </MyStuff>
> '
> EXEC sp_xml_preparedocument @.docHandle OUTPUT, @.xmlDocument
> -- Use OPENXML to provide rowset consisting of customer data.
> INSERT Mystuff
> SELECT *
> FROM OPENXML(@.docHandle, N'/MyStuff', 2)
> WITH (proudct varchar(20) '/MyStuff/Product',
> version int '/MyStuff/Version')
> EXEC sp_xml_removedocument @.docHandle
> select * from mystuff
>
> If you have any comments or feedback, please feel free to let's know. Thank
> you.
> Best Regards,
> Peter Yang
> MCSE2000/2003, MCSA, MCDBA
> Microsoft Online Community Support
> ==================================================> Get notification to my posts through email? Please refer to
> http://msdn.microsoft.com/subscriptions/managednewsgroups/default.aspx#notif
> ications
> <http://msdn.microsoft.com/subscriptions/managednewsgroups/default.aspx>.
> Note: The MSDN Managed Newsgroup support offering is for non-urgent issues
> where an initial response from the community or a Microsoft Support
> Engineer within 1 business day is acceptable. Please note that each follow
> up response may take approximately 2 business days as the support
> professional working with you may need further investigation to reach the
> most efficient resolution. The offering is not appropriate for situations
> that require urgent, real-time or phone-based interactions or complex
> project analysis and dump analysis issues. Issues of this nature are best
> handled working with a dedicated Microsoft Support Engineer by contacting
> Microsoft Customer Support Services (CSS) at
> <http://msdn.microsoft.com/subscriptions/support/default.aspx>.
> ==================================================> This posting is provided "AS IS" with no warranties, and confers no rights.
>|||Asher
I see what you mean
DECLARE @.idoc int
DECLARE @.doc varchar(1000)
SET @.doc ='
<AddressType name="Shipping">
<AddressRecord StreetLine1="111 Main"
City="Houston"
State="TX"
ZipCode = "11111" />
<AddressRecord StreetLine1="111 Left"
City="New_York"
State="NY"
ZipCode = "3333" />
</AddressType> '
EXEC sp_xml_preparedocument @.idoc OUTPUT, @.doc
SELECT *
FROM OPENXML (@.idoc, '/AddressType/AddressRecord',1)
WITH (City varchar(10))
"Asher F." <asherfoa@.community.nospam> wrote in message
news:F678FA42-605F-4425-BED3-EBEF75A47219@.microsoft.com...
> Hi Uri,
> Thanks for the quick response, I understand the code you suggested and it
> works for me 80% of times. I would very much appreciate your help in this
> follow-up.
> To extend your example, let?s say I have this:
> declare @.tAddress table (Address xml DEFAULT '<Address Record />' )
> INSERT INTO @.tAddress (Address)
> VALUES ('<AddressType name="Shipping">
> <AddressRecord StreetLine1="111 Main"
> City="Houston"
> State="TX"
> ZipCode = "11111" />
> <AddressRecord StreetLine1="222 Left"
> City="New York"
> State="NY"
> ZipCode = "2222" />
> </AddressType> ');
> INSERT INTO @.tAddress (Address)
> VALUES ('<AddressType name="Shipping">
> <AddressRecord StreetLine1="222 Main"
> City="SanDiego"
> State="CA"
> ZipCode = "2222" />
> </AddressType> ');
> Note that the XML in the first row has two AddressRecord elements.
> How do I make my result return a rowset with 3 rows representing all the
> values:
> SanDiego
> New York
> Houston
> Thanks in advance!
> --
> Asher.
>
> "Uri Dimant" wrote:
>> Asher
>> Take a look at this example to resturn Cities
>> declare @.tAddress table (Address xml DEFAULT '<Address Record />' )
>> INSERT INTO @.tAddress (Address)
>> VALUES ('<AddressType name="Shipping">
>> <AddressRecord StreetLine1="111 Main"
>> City="Houston"
>> State="TX"
>> ZipCode = "11111" />
>> </AddressType> ');
>> INSERT INTO @.tAddress (Address)
>> VALUES ('<AddressType name="Shipping">
>> <AddressRecord StreetLine1="222 Main"
>> City="SanDiego"
>> State="CA"
>> ZipCode = "2222" />
>> </AddressType> ');
>> SELECT address.value('(/AddressType/AddressRecord/@.City)[1]',
>> 'varchar(50)' )
>> FROM @.tAddress
>>
>> "Asher F." <asherfoa@.community.nospam> wrote in message
>> news:407A573D-01B4-4AEE-A10B-FA2CB5AA0A17@.microsoft.com...
>> > Hi,
>> > I have a question regarding extracting XML data as relational data.
>> > From
>> > the looks of it looks like the OPENXML statement is not relevant since
>> > the
>> > data I need to extract resides in XML columns (in a table).
>> > So I have this query
>> > SELECT [MyXmlCol] FROM [MyTable] WHERE [XmlTypeName]='Schema1'
>> > MyXmlCol ? is an XML column, and the WHERE clause assures me that I
>> > know
>> > how
>> > to access this XML data (using xpath)
>> > Let?s say the data in XML column looks like this
>> > <MyStuff>
>> > <Product>Windows Vista</Product>
>> > <Version>1</Version>
>> > </MyStuff>
>> > (were in each row the value of the Product & Version nodes are
>> > different).
>> > I want a table/rowset with two columns Product (xpath
>> > /MyStuff/Product/text()) and Version (xpath /MyStuff/Version/text())
>> > I want the rowset/table to look like this:
>> > Product Version
>> > -- --
>> > WindowsVista 1
>> > WindowsHome 2
>> > WindowsXP 3
>> >
>> > How do I accomplish this ?
>> >
>> >
>> >
>> > --
>> > Asher.
>>|||Or
SELECT address.value('(/AddressType/AddressRecord/@.City)[1]', 'varchar(50)')
FROM @.tAddress
UNION ALL
SELECT address.value('(/AddressType/AddressRecord/@.City)[2]', 'varchar(50)')
FROM @.tAddress
"Uri Dimant" <urid@.iscar.co.il> wrote in message
news:O9Qv3C1THHA.5108@.TK2MSFTNGP06.phx.gbl...
> Asher
> I see what you mean
> DECLARE @.idoc int
> DECLARE @.doc varchar(1000)
> SET @.doc ='
> <AddressType name="Shipping">
> <AddressRecord StreetLine1="111 Main"
> City="Houston"
> State="TX"
> ZipCode = "11111" />
> <AddressRecord StreetLine1="111 Left"
> City="New_York"
> State="NY"
> ZipCode = "3333" />
> </AddressType> '
> EXEC sp_xml_preparedocument @.idoc OUTPUT, @.doc
> SELECT *
> FROM OPENXML (@.idoc, '/AddressType/AddressRecord',1)
> WITH (City varchar(10))
>
>
>
> "Asher F." <asherfoa@.community.nospam> wrote in message
> news:F678FA42-605F-4425-BED3-EBEF75A47219@.microsoft.com...
>> Hi Uri,
>> Thanks for the quick response, I understand the code you suggested and it
>> works for me 80% of times. I would very much appreciate your help in this
>> follow-up.
>> To extend your example, let?s say I have this:
>> declare @.tAddress table (Address xml DEFAULT '<Address Record />' )
>> INSERT INTO @.tAddress (Address)
>> VALUES ('<AddressType name="Shipping">
>> <AddressRecord StreetLine1="111 Main"
>> City="Houston"
>> State="TX"
>> ZipCode = "11111" />
>> <AddressRecord StreetLine1="222 Left"
>> City="New York"
>> State="NY"
>> ZipCode = "2222" />
>> </AddressType> ');
>> INSERT INTO @.tAddress (Address)
>> VALUES ('<AddressType name="Shipping">
>> <AddressRecord StreetLine1="222 Main"
>> City="SanDiego"
>> State="CA"
>> ZipCode = "2222" />
>> </AddressType> ');
>> Note that the XML in the first row has two AddressRecord elements.
>> How do I make my result return a rowset with 3 rows representing all the
>> values:
>> SanDiego
>> New York
>> Houston
>> Thanks in advance!
>> --
>> Asher.
>>
>> "Uri Dimant" wrote:
>> Asher
>> Take a look at this example to resturn Cities
>> declare @.tAddress table (Address xml DEFAULT '<Address Record />' )
>> INSERT INTO @.tAddress (Address)
>> VALUES ('<AddressType name="Shipping">
>> <AddressRecord StreetLine1="111 Main"
>> City="Houston"
>> State="TX"
>> ZipCode = "11111" />
>> </AddressType> ');
>> INSERT INTO @.tAddress (Address)
>> VALUES ('<AddressType name="Shipping">
>> <AddressRecord StreetLine1="222 Main"
>> City="SanDiego"
>> State="CA"
>> ZipCode = "2222" />
>> </AddressType> ');
>> SELECT address.value('(/AddressType/AddressRecord/@.City)[1]',
>> 'varchar(50)' )
>> FROM @.tAddress
>>
>> "Asher F." <asherfoa@.community.nospam> wrote in message
>> news:407A573D-01B4-4AEE-A10B-FA2CB5AA0A17@.microsoft.com...
>> > Hi,
>> > I have a question regarding extracting XML data as relational data.
>> > From
>> > the looks of it looks like the OPENXML statement is not relevant since
>> > the
>> > data I need to extract resides in XML columns (in a table).
>> > So I have this query
>> > SELECT [MyXmlCol] FROM [MyTable] WHERE [XmlTypeName]='Schema1'
>> > MyXmlCol ? is an XML column, and the WHERE clause assures me that I
>> > know
>> > how
>> > to access this XML data (using xpath)
>> > Let?s say the data in XML column looks like this
>> > <MyStuff>
>> > <Product>Windows Vista</Product>
>> > <Version>1</Version>
>> > </MyStuff>
>> > (were in each row the value of the Product & Version nodes are
>> > different).
>> > I want a table/rowset with two columns Product (xpath
>> > /MyStuff/Product/text()) and Version (xpath /MyStuff/Version/text())
>> > I want the rowset/table to look like this:
>> > Product Version
>> > -- --
>> > WindowsVista 1
>> > WindowsHome 2
>> > WindowsXP 3
>> >
>> > How do I accomplish this ?
>> >
>> >
>> >
>> > --
>> > Asher.
>>
>|||Hi Uri,
The OPENXML is probably not the solution since I have the XML document
stored in a column.
As for your second solution, it looks like it is in the right direction,
however I don't see how I can scale it to the case I have N (or an unkown
number of) AddressRecord nodes in each XML document (i.e. one row's XML
column might contain zero AddressRecord elements and another row might
contain 10...)
How do I scale this solution ?
--
Asher.
"Uri Dimant" wrote:
> Or
> SELECT address.value('(/AddressType/AddressRecord/@.City)[1]', 'varchar(50)')
> FROM @.tAddress
> UNION ALL
> SELECT address.value('(/AddressType/AddressRecord/@.City)[2]', 'varchar(50)')
> FROM @.tAddress
>
>
>
> "Uri Dimant" <urid@.iscar.co.il> wrote in message
> news:O9Qv3C1THHA.5108@.TK2MSFTNGP06.phx.gbl...
> > Asher
> > I see what you mean
> >
> > DECLARE @.idoc int
> >
> > DECLARE @.doc varchar(1000)
> >
> > SET @.doc ='
> >
> > <AddressType name="Shipping">
> >
> > <AddressRecord StreetLine1="111 Main"
> >
> > City="Houston"
> >
> > State="TX"
> >
> > ZipCode = "11111" />
> >
> > <AddressRecord StreetLine1="111 Left"
> >
> > City="New_York"
> >
> > State="NY"
> >
> > ZipCode = "3333" />
> >
> > </AddressType> '
> >
> > EXEC sp_xml_preparedocument @.idoc OUTPUT, @.doc
> >
> > SELECT *
> >
> > FROM OPENXML (@.idoc, '/AddressType/AddressRecord',1)
> >
> > WITH (City varchar(10))
> >
> >
> >
> >
> >
> >
> >
> > "Asher F." <asherfoa@.community.nospam> wrote in message
> > news:F678FA42-605F-4425-BED3-EBEF75A47219@.microsoft.com...
> >> Hi Uri,
> >> Thanks for the quick response, I understand the code you suggested and it
> >> works for me 80% of times. I would very much appreciate your help in this
> >> follow-up.
> >>
> >> To extend your example, letâ's say I have this:
> >>
> >> declare @.tAddress table (Address xml DEFAULT '<Address Record />' )
> >>
> >> INSERT INTO @.tAddress (Address)
> >>
> >> VALUES ('<AddressType name="Shipping">
> >>
> >> <AddressRecord StreetLine1="111 Main"
> >>
> >> City="Houston"
> >>
> >> State="TX"
> >>
> >> ZipCode = "11111" />
> >> <AddressRecord StreetLine1="222 Left"
> >>
> >> City="New York"
> >>
> >> State="NY"
> >>
> >> ZipCode = "2222" />
> >>
> >> </AddressType> ');
> >>
> >> INSERT INTO @.tAddress (Address)
> >>
> >> VALUES ('<AddressType name="Shipping">
> >>
> >> <AddressRecord StreetLine1="222 Main"
> >>
> >> City="SanDiego"
> >>
> >> State="CA"
> >>
> >> ZipCode = "2222" />
> >>
> >> </AddressType> ');
> >>
> >> Note that the XML in the first row has two AddressRecord elements.
> >> How do I make my result return a rowset with 3 rows representing all the
> >> values:
> >> SanDiego
> >> New York
> >> Houston
> >>
> >> Thanks in advance!
> >> --
> >> Asher.
> >>
> >>
> >> "Uri Dimant" wrote:
> >>
> >> Asher
> >> Take a look at this example to resturn Cities
> >>
> >> declare @.tAddress table (Address xml DEFAULT '<Address Record />' )
> >>
> >> INSERT INTO @.tAddress (Address)
> >>
> >> VALUES ('<AddressType name="Shipping">
> >>
> >> <AddressRecord StreetLine1="111 Main"
> >>
> >> City="Houston"
> >>
> >> State="TX"
> >>
> >> ZipCode = "11111" />
> >>
> >> </AddressType> ');
> >>
> >> INSERT INTO @.tAddress (Address)
> >>
> >> VALUES ('<AddressType name="Shipping">
> >>
> >> <AddressRecord StreetLine1="222 Main"
> >>
> >> City="SanDiego"
> >>
> >> State="CA"
> >>
> >> ZipCode = "2222" />
> >>
> >> </AddressType> ');
> >>
> >> SELECT address.value('(/AddressType/AddressRecord/@.City)[1]',
> >> 'varchar(50)' )
> >>
> >> FROM @.tAddress
> >>
> >>
> >>
> >> "Asher F." <asherfoa@.community.nospam> wrote in message
> >> news:407A573D-01B4-4AEE-A10B-FA2CB5AA0A17@.microsoft.com...
> >> > Hi,
> >> > I have a question regarding extracting XML data as relational data.
> >> > From
> >> > the looks of it looks like the OPENXML statement is not relevant since
> >> > the
> >> > data I need to extract resides in XML columns (in a table).
> >> > So I have this query
> >> > SELECT [MyXmlCol] FROM [MyTable] WHERE [XmlTypeName]='Schema1'
> >> > MyXmlCol â' is an XML column, and the WHERE clause assures me that I
> >> > know
> >> > how
> >> > to access this XML data (using xpath)
> >> > Letâ's say the data in XML column looks like this
> >> > <MyStuff>
> >> > <Product>Windows Vista</Product>
> >> > <Version>1</Version>
> >> > </MyStuff>
> >> > (were in each row the value of the Product & Version nodes are
> >> > different).
> >> > I want a table/rowset with two columns Product (xpath
> >> > /MyStuff/Product/text()) and Version (xpath /MyStuff/Version/text())
> >> > I want the rowset/table to look like this:
> >> > Product Version
> >> > -- --
> >> > WindowsVista 1
> >> > WindowsHome 2
> >> > WindowsXP 3
> >> >
> >> > How do I accomplish this ?
> >> >
> >> >
> >> >
> >> > --
> >> > Asher.
> >>
> >>
> >>
> >
> >
>
>|||Check out the nodes method. It returns a table with one column of datatype XML. So if the XML
instance is a column inside a table, you use this with APPLY against the table where the column
exists. Example, assuming table is temp table instead of table variable:
SELECT nc.value('@.City[1]', 'varchar(39)')
FROM #tAddress
CROSS APPLY Address.nodes('/AddressType/AddressRecord') AS nt(nc)
---
Houston
New York
SanDiego
--
Tibor Karaszi, SQL Server MVP
http://www.karaszi.com/sqlserver/default.asp
http://www.solidqualitylearning.com/
"Asher F." <asherfoa@.community.nospam> wrote in message
news:F678FA42-605F-4425-BED3-EBEF75A47219@.microsoft.com...
> Hi Uri,
> Thanks for the quick response, I understand the code you suggested and it
> works for me 80% of times. I would very much appreciate your help in this
> follow-up.
> To extend your example, let's say I have this:
> declare @.tAddress table (Address xml DEFAULT '<Address Record />' )
> INSERT INTO @.tAddress (Address)
> VALUES ('<AddressType name="Shipping">
> <AddressRecord StreetLine1="111 Main"
> City="Houston"
> State="TX"
> ZipCode = "11111" />
> <AddressRecord StreetLine1="222 Left"
> City="New York"
> State="NY"
> ZipCode = "2222" />
> </AddressType> ');
> INSERT INTO @.tAddress (Address)
> VALUES ('<AddressType name="Shipping">
> <AddressRecord StreetLine1="222 Main"
> City="SanDiego"
> State="CA"
> ZipCode = "2222" />
> </AddressType> ');
> Note that the XML in the first row has two AddressRecord elements.
> How do I make my result return a rowset with 3 rows representing all the
> values:
> SanDiego
> New York
> Houston
> Thanks in advance!
> --
> Asher.
>
> "Uri Dimant" wrote:
>> Asher
>> Take a look at this example to resturn Cities
>> declare @.tAddress table (Address xml DEFAULT '<Address Record />' )
>> INSERT INTO @.tAddress (Address)
>> VALUES ('<AddressType name="Shipping">
>> <AddressRecord StreetLine1="111 Main"
>> City="Houston"
>> State="TX"
>> ZipCode = "11111" />
>> </AddressType> ');
>> INSERT INTO @.tAddress (Address)
>> VALUES ('<AddressType name="Shipping">
>> <AddressRecord StreetLine1="222 Main"
>> City="SanDiego"
>> State="CA"
>> ZipCode = "2222" />
>> </AddressType> ');
>> SELECT address.value('(/AddressType/AddressRecord/@.City)[1]',
>> 'varchar(50)' )
>> FROM @.tAddress
>>
>> "Asher F." <asherfoa@.community.nospam> wrote in message
>> news:407A573D-01B4-4AEE-A10B-FA2CB5AA0A17@.microsoft.com...
>> > Hi,
>> > I have a question regarding extracting XML data as relational data. From
>> > the looks of it looks like the OPENXML statement is not relevant since the
>> > data I need to extract resides in XML columns (in a table).
>> > So I have this query
>> > SELECT [MyXmlCol] FROM [MyTable] WHERE [XmlTypeName]='Schema1'
>> > MyXmlCol - is an XML column, and the WHERE clause assures me that I know
>> > how
>> > to access this XML data (using xpath)
>> > Let's say the data in XML column looks like this
>> > <MyStuff>
>> > <Product>Windows Vista</Product>
>> > <Version>1</Version>
>> > </MyStuff>
>> > (were in each row the value of the Product & Version nodes are different).
>> > I want a table/rowset with two columns Product (xpath
>> > /MyStuff/Product/text()) and Version (xpath /MyStuff/Version/text())
>> > I want the rowset/table to look like this:
>> > Product Version
>> > -- --
>> > WindowsVista 1
>> > WindowsHome 2
>> > WindowsXP 3
>> >
>> > How do I accomplish this ?
>> >
>> >
>> >
>> > --
>> > Asher.
>>|||See Tibor's reply , it is what you wanted
"Asher F." <asherfoa@.community.nospam> wrote in message
news:38DFF454-06D5-4E00-B623-F5FB216D87EC@.microsoft.com...
> Hi Uri,
> The OPENXML is probably not the solution since I have the XML document
> stored in a column.
> As for your second solution, it looks like it is in the right direction,
> however I don't see how I can scale it to the case I have N (or an unkown
> number of) AddressRecord nodes in each XML document (i.e. one row's XML
> column might contain zero AddressRecord elements and another row might
> contain 10...)
> How do I scale this solution ?
> --
> Asher.
>
> "Uri Dimant" wrote:
>> Or
>> SELECT address.value('(/AddressType/AddressRecord/@.City)[1]',
>> 'varchar(50)')
>> FROM @.tAddress
>> UNION ALL
>> SELECT address.value('(/AddressType/AddressRecord/@.City)[2]',
>> 'varchar(50)')
>> FROM @.tAddress
>>
>>
>>
>> "Uri Dimant" <urid@.iscar.co.il> wrote in message
>> news:O9Qv3C1THHA.5108@.TK2MSFTNGP06.phx.gbl...
>> > Asher
>> > I see what you mean
>> >
>> > DECLARE @.idoc int
>> >
>> > DECLARE @.doc varchar(1000)
>> >
>> > SET @.doc ='
>> >
>> > <AddressType name="Shipping">
>> >
>> > <AddressRecord StreetLine1="111 Main"
>> >
>> > City="Houston"
>> >
>> > State="TX"
>> >
>> > ZipCode = "11111" />
>> >
>> > <AddressRecord StreetLine1="111 Left"
>> >
>> > City="New_York"
>> >
>> > State="NY"
>> >
>> > ZipCode = "3333" />
>> >
>> > </AddressType> '
>> >
>> > EXEC sp_xml_preparedocument @.idoc OUTPUT, @.doc
>> >
>> > SELECT *
>> >
>> > FROM OPENXML (@.idoc, '/AddressType/AddressRecord',1)
>> >
>> > WITH (City varchar(10))
>> >
>> >
>> >
>> >
>> >
>> >
>> >
>> > "Asher F." <asherfoa@.community.nospam> wrote in message
>> > news:F678FA42-605F-4425-BED3-EBEF75A47219@.microsoft.com...
>> >> Hi Uri,
>> >> Thanks for the quick response, I understand the code you suggested and
>> >> it
>> >> works for me 80% of times. I would very much appreciate your help in
>> >> this
>> >> follow-up.
>> >>
>> >> To extend your example, let?s say I have this:
>> >>
>> >> declare @.tAddress table (Address xml DEFAULT '<Address Record />' )
>> >>
>> >> INSERT INTO @.tAddress (Address)
>> >>
>> >> VALUES ('<AddressType name="Shipping">
>> >>
>> >> <AddressRecord StreetLine1="111 Main"
>> >>
>> >> City="Houston"
>> >>
>> >> State="TX"
>> >>
>> >> ZipCode = "11111" />
>> >> <AddressRecord StreetLine1="222 Left"
>> >>
>> >> City="New York"
>> >>
>> >> State="NY"
>> >>
>> >> ZipCode = "2222" />
>> >>
>> >> </AddressType> ');
>> >>
>> >> INSERT INTO @.tAddress (Address)
>> >>
>> >> VALUES ('<AddressType name="Shipping">
>> >>
>> >> <AddressRecord StreetLine1="222 Main"
>> >>
>> >> City="SanDiego"
>> >>
>> >> State="CA"
>> >>
>> >> ZipCode = "2222" />
>> >>
>> >> </AddressType> ');
>> >>
>> >> Note that the XML in the first row has two AddressRecord elements.
>> >> How do I make my result return a rowset with 3 rows representing all
>> >> the
>> >> values:
>> >> SanDiego
>> >> New York
>> >> Houston
>> >>
>> >> Thanks in advance!
>> >> --
>> >> Asher.
>> >>
>> >>
>> >> "Uri Dimant" wrote:
>> >>
>> >> Asher
>> >> Take a look at this example to resturn Cities
>> >>
>> >> declare @.tAddress table (Address xml DEFAULT '<Address Record />' )
>> >>
>> >> INSERT INTO @.tAddress (Address)
>> >>
>> >> VALUES ('<AddressType name="Shipping">
>> >>
>> >> <AddressRecord StreetLine1="111 Main"
>> >>
>> >> City="Houston"
>> >>
>> >> State="TX"
>> >>
>> >> ZipCode = "11111" />
>> >>
>> >> </AddressType> ');
>> >>
>> >> INSERT INTO @.tAddress (Address)
>> >>
>> >> VALUES ('<AddressType name="Shipping">
>> >>
>> >> <AddressRecord StreetLine1="222 Main"
>> >>
>> >> City="SanDiego"
>> >>
>> >> State="CA"
>> >>
>> >> ZipCode = "2222" />
>> >>
>> >> </AddressType> ');
>> >>
>> >> SELECT address.value('(/AddressType/AddressRecord/@.City)[1]',
>> >> 'varchar(50)' )
>> >>
>> >> FROM @.tAddress
>> >>
>> >>
>> >>
>> >> "Asher F." <asherfoa@.community.nospam> wrote in message
>> >> news:407A573D-01B4-4AEE-A10B-FA2CB5AA0A17@.microsoft.com...
>> >> > Hi,
>> >> > I have a question regarding extracting XML data as relational data.
>> >> > From
>> >> > the looks of it looks like the OPENXML statement is not relevant
>> >> > since
>> >> > the
>> >> > data I need to extract resides in XML columns (in a table).
>> >> > So I have this query
>> >> > SELECT [MyXmlCol] FROM [MyTable] WHERE [XmlTypeName]='Schema1'
>> >> > MyXmlCol ? is an XML column, and the WHERE clause assures me that I
>> >> > know
>> >> > how
>> >> > to access this XML data (using xpath)
>> >> > Let?s say the data in XML column looks like this
>> >> > <MyStuff>
>> >> > <Product>Windows Vista</Product>
>> >> > <Version>1</Version>
>> >> > </MyStuff>
>> >> > (were in each row the value of the Product & Version nodes are
>> >> > different).
>> >> > I want a table/rowset with two columns Product (xpath
>> >> > /MyStuff/Product/text()) and Version (xpath
>> >> > /MyStuff/Version/text())
>> >> > I want the rowset/table to look like this:
>> >> > Product Version
>> >> > -- --
>> >> > WindowsVista 1
>> >> > WindowsHome 2
>> >> > WindowsXP 3
>> >> >
>> >> > How do I accomplish this ?
>> >> >
>> >> >
>> >> >
>> >> > --
>> >> > Asher.
>> >>
>> >>
>> >>
>> >
>> >
>>|||Did you look at my suggestion? It produces exactly the result you asked for...
--
Tibor Karaszi, SQL Server MVP
http://www.karaszi.com/sqlserver/default.asp
http://sqlblog.com/blogs/tibor_karaszi
"Asher F." <asherfoa@.community.nospam> wrote in message
news:161F4EC4-7C19-430D-91EF-D2684D840D9D@.microsoft.com...
> Hello Peter,
> Thank you for your response.
> What I am trying to do is to extract data from an XML column I have in one
> of my tables. I am trying to make the data in the XML column appear as part
> of the normal rowset.
> I will run a SQL SELECT statement from my C# code expecting to receive the
> data back in the form of a rowset (so I can read it as primitive types using
> IDataReader).
> Anyway, Uri's response works for me in most of the cases, it doesn't cover
> (and I tried several things myself) the case where I need to extract multiple
> values (i.e. rows) from a single XML document (stored in the xml column in a
> given row in my table).
> Thanks for the help
> --
> Asher.
>
> "Peter Yang [MSFT]" wrote:
>> Hello Asher,
>> I understand that you'd like to insert a table from a xml type value. You
>> may want to use openxml to do the job:
>> CREATE TABLE MyStuff( product varchar(20), version int)
>> go
>> DECLARE @.docHandle int
>> declare @.xmlDocument xml
>> set @.xmlDocument = N'
>> <MyStuff>
>> <Product>Windows Vista</Product>
>> <Version>1</Version>
>> </MyStuff>
>> '
>> EXEC sp_xml_preparedocument @.docHandle OUTPUT, @.xmlDocument
>> -- Use OPENXML to provide rowset consisting of customer data.
>> INSERT Mystuff
>> SELECT *
>> FROM OPENXML(@.docHandle, N'/MyStuff', 2)
>> WITH (proudct varchar(20) '/MyStuff/Product',
>> version int '/MyStuff/Version')
>> EXEC sp_xml_removedocument @.docHandle
>> select * from mystuff
>>
>> If you have any comments or feedback, please feel free to let's know. Thank
>> you.
>> Best Regards,
>> Peter Yang
>> MCSE2000/2003, MCSA, MCDBA
>> Microsoft Online Community Support
>> ==================================================>> Get notification to my posts through email? Please refer to
>> http://msdn.microsoft.com/subscriptions/managednewsgroups/default.aspx#notif
>> ications
>> <http://msdn.microsoft.com/subscriptions/managednewsgroups/default.aspx>.
>> Note: The MSDN Managed Newsgroup support offering is for non-urgent issues
>> where an initial response from the community or a Microsoft Support
>> Engineer within 1 business day is acceptable. Please note that each follow
>> up response may take approximately 2 business days as the support
>> professional working with you may need further investigation to reach the
>> most efficient resolution. The offering is not appropriate for situations
>> that require urgent, real-time or phone-based interactions or complex
>> project analysis and dump analysis issues. Issues of this nature are best
>> handled working with a dedicated Microsoft Support Engineer by contacting
>> Microsoft Customer Support Services (CSS) at
>> <http://msdn.microsoft.com/subscriptions/support/default.aspx>.
>> ==================================================>> This posting is provided "AS IS" with no warranties, and confers no rights.
>>|||Thank you Tibor, it looks like exactlly what I need.
I am now reading the relevant documentation about the functions you used so
I can get a deeper understanding.
Thank you very much for your help!
--
Asher.
"Tibor Karaszi" wrote:
> Did you look at my suggestion? It produces exactly the result you asked for...
> --
> Tibor Karaszi, SQL Server MVP
> http://www.karaszi.com/sqlserver/default.asp
> http://sqlblog.com/blogs/tibor_karaszi
>
> "Asher F." <asherfoa@.community.nospam> wrote in message
> news:161F4EC4-7C19-430D-91EF-D2684D840D9D@.microsoft.com...
> > Hello Peter,
> > Thank you for your response.
> > What I am trying to do is to extract data from an XML column I have in one
> > of my tables. I am trying to make the data in the XML column appear as part
> > of the normal rowset.
> > I will run a SQL SELECT statement from my C# code expecting to receive the
> > data back in the form of a rowset (so I can read it as primitive types using
> > IDataReader).
> > Anyway, Uri's response works for me in most of the cases, it doesn't cover
> > (and I tried several things myself) the case where I need to extract multiple
> > values (i.e. rows) from a single XML document (stored in the xml column in a
> > given row in my table).
> >
> > Thanks for the help
> >
> > --
> > Asher.
> >
> >
> > "Peter Yang [MSFT]" wrote:
> >
> >> Hello Asher,
> >>
> >> I understand that you'd like to insert a table from a xml type value. You
> >> may want to use openxml to do the job:
> >>
> >> CREATE TABLE MyStuff( product varchar(20), version int)
> >> go
> >>
> >> DECLARE @.docHandle int
> >> declare @.xmlDocument xml
> >> set @.xmlDocument = N'
> >> <MyStuff>
> >> <Product>Windows Vista</Product>
> >> <Version>1</Version>
> >> </MyStuff>
> >> '
> >> EXEC sp_xml_preparedocument @.docHandle OUTPUT, @.xmlDocument
> >> -- Use OPENXML to provide rowset consisting of customer data.
> >> INSERT Mystuff
> >> SELECT *
> >> FROM OPENXML(@.docHandle, N'/MyStuff', 2)
> >> WITH (proudct varchar(20) '/MyStuff/Product',
> >> version int '/MyStuff/Version')
> >>
> >> EXEC sp_xml_removedocument @.docHandle
> >>
> >> select * from mystuff
> >>
> >>
> >> If you have any comments or feedback, please feel free to let's know. Thank
> >> you.
> >> Best Regards,
> >>
> >> Peter Yang
> >> MCSE2000/2003, MCSA, MCDBA
> >> Microsoft Online Community Support
> >> ==================================================> >> Get notification to my posts through email? Please refer to
> >> http://msdn.microsoft.com/subscriptions/managednewsgroups/default.aspx#notif
> >> ications
> >> <http://msdn.microsoft.com/subscriptions/managednewsgroups/default.aspx>.
> >> Note: The MSDN Managed Newsgroup support offering is for non-urgent issues
> >> where an initial response from the community or a Microsoft Support
> >> Engineer within 1 business day is acceptable. Please note that each follow
> >> up response may take approximately 2 business days as the support
> >> professional working with you may need further investigation to reach the
> >> most efficient resolution. The offering is not appropriate for situations
> >> that require urgent, real-time or phone-based interactions or complex
> >> project analysis and dump analysis issues. Issues of this nature are best
> >> handled working with a dedicated Microsoft Support Engineer by contacting
> >> Microsoft Customer Support Services (CSS) at
> >> <http://msdn.microsoft.com/subscriptions/support/default.aspx>.
> >> ==================================================> >> This posting is provided "AS IS" with no warranties, and confers no rights.
> >>
> >>
>

Extract values from XML column (rows) to a rowset

Hi,
I have a question regarding extracting XML data as relational data. From
the looks of it looks like the OPENXML statement is not relevant since the
data I need to extract resides in XML columns (in a table).
So I have this query
SELECT [MyXmlCol] FROM [MyTable] WHERE [XmlTypeName]='Schema1'
MyXmlCol – is an XML column, and the WHERE clause assures me that I know how
to access this XML data (using xpath)
Let’s say the data in XML column looks like this
<MyStuff>
<Product>Windows Vista</Product>
<Version>1</Version>
</MyStuff>
(were in each row the value of the Product & Version nodes are different).
I want a table/rowset with two columns Product (xpath
/MyStuff/Product/text()) and Version (xpath /MyStuff/Version/text())
I want the rowset/table to look like this:
Product Version
WindowsVista1
WindowsHome2
WindowsXP3
How do I accomplish this ?
Asher.
Asher
Take a look at this example to resturn Cities
declare @.tAddress table (Address xml DEFAULT '<Address Record />' )
INSERT INTO @.tAddress (Address)
VALUES ('<AddressType name="Shipping">
<AddressRecord StreetLine1="111 Main"
City="Houston"
State="TX"
ZipCode = "11111" />
</AddressType> ');
INSERT INTO @.tAddress (Address)
VALUES ('<AddressType name="Shipping">
<AddressRecord StreetLine1="222 Main"
City="SanDiego"
State="CA"
ZipCode = "2222" />
</AddressType> ');
SELECT address.value('(/AddressType/AddressRecord/@.City)[1]',
'varchar(50)' )
FROM @.tAddress
"Asher F." <asherfoa@.community.nospam> wrote in message
news:407A573D-01B4-4AEE-A10B-FA2CB5AA0A17@.microsoft.com...
> Hi,
> I have a question regarding extracting XML data as relational data. From
> the looks of it looks like the OPENXML statement is not relevant since the
> data I need to extract resides in XML columns (in a table).
> So I have this query
> SELECT [MyXmlCol] FROM [MyTable] WHERE [XmlTypeName]='Schema1'
> MyXmlCol is an XML column, and the WHERE clause assures me that I know
> how
> to access this XML data (using xpath)
> Lets say the data in XML column looks like this
> <MyStuff>
> <Product>Windows Vista</Product>
> <Version>1</Version>
> </MyStuff>
> (were in each row the value of the Product & Version nodes are different).
> I want a table/rowset with two columns Product (xpath
> /MyStuff/Product/text()) and Version (xpath /MyStuff/Version/text())
> I want the rowset/table to look like this:
> Product Version
> -- --
> WindowsVista 1
> WindowsHome 2
> WindowsXP 3
> How do I accomplish this ?
>
> --
> Asher.
|||Hello Asher,
I understand that you'd like to insert a table from a xml type value. You
may want to use openxml to do the job:
CREATE TABLE MyStuff( product varchar(20), version int)
go
DECLARE @.docHandle int
declare @.xmlDocument xml
set @.xmlDocument = N'
<MyStuff>
<Product>Windows Vista</Product>
<Version>1</Version>
</MyStuff>
'
EXEC sp_xml_preparedocument @.docHandle OUTPUT, @.xmlDocument
-- Use OPENXML to provide rowset consisting of customer data.
INSERT Mystuff
SELECT *
FROM OPENXML(@.docHandle, N'/MyStuff', 2)
WITH (proudct varchar(20) '/MyStuff/Product',
version int '/MyStuff/Version')
EXEC sp_xml_removedocument @.docHandle
select * from mystuff
If you have any comments or feedback, please feel free to let's know. Thank
you.
Best Regards,
Peter Yang
MCSE2000/2003, MCSA, MCDBA
Microsoft Online Community Support
==================================================
Get notification to my posts through email? Please refer to
http://msdn.microsoft.com/subscriptions/managednewsgroups/default.aspx#notif
ications
<http://msdn.microsoft.com/subscriptions/managednewsgroups/default.aspx>.
Note: The MSDN Managed Newsgroup support offering is for non-urgent issues
where an initial response from the community or a Microsoft Support
Engineer within 1 business day is acceptable. Please note that each follow
up response may take approximately 2 business days as the support
professional working with you may need further investigation to reach the
most efficient resolution. The offering is not appropriate for situations
that require urgent, real-time or phone-based interactions or complex
project analysis and dump analysis issues. Issues of this nature are best
handled working with a dedicated Microsoft Support Engineer by contacting
Microsoft Customer Support Services (CSS) at
<http://msdn.microsoft.com/subscriptions/support/default.aspx>.
==================================================
This posting is provided "AS IS" with no warranties, and confers no rights.
|||Hi Uri,
Thanks for the quick response, I understand the code you suggested and it
works for me 80% of times. I would very much appreciate your help in this
follow-up.
To extend your example, let’s say I have this:
declare @.tAddress table (Address xml DEFAULT '<Address Record />' )
INSERT INTO @.tAddress (Address)
VALUES ('<AddressType name="Shipping">
<AddressRecord StreetLine1="111 Main"
City="Houston"
State="TX"
ZipCode = "11111" />
<AddressRecord StreetLine1="222 Left"
City="New York"
State="NY"
ZipCode = "2222" />
</AddressType> ');
INSERT INTO @.tAddress (Address)
VALUES ('<AddressType name="Shipping">
<AddressRecord StreetLine1="222 Main"
City="SanDiego"
State="CA"
ZipCode = "2222" />
</AddressType> ');
Note that the XML in the first row has two AddressRecord elements.
How do I make my result return a rowset with 3 rows representing all the
values:
SanDiego
New York
Houston
Thanks in advance!
Asher.
"Uri Dimant" wrote:

> Asher
> Take a look at this example to resturn Cities
> declare @.tAddress table (Address xml DEFAULT '<Address Record />' )
> INSERT INTO @.tAddress (Address)
> VALUES ('<AddressType name="Shipping">
> <AddressRecord StreetLine1="111 Main"
> City="Houston"
> State="TX"
> ZipCode = "11111" />
> </AddressType> ');
> INSERT INTO @.tAddress (Address)
> VALUES ('<AddressType name="Shipping">
> <AddressRecord StreetLine1="222 Main"
> City="SanDiego"
> State="CA"
> ZipCode = "2222" />
> </AddressType> ');
> SELECT address.value('(/AddressType/AddressRecord/@.City)[1]',
> 'varchar(50)' )
> FROM @.tAddress
>
> "Asher F." <asherfoa@.community.nospam> wrote in message
> news:407A573D-01B4-4AEE-A10B-FA2CB5AA0A17@.microsoft.com...
>
>
|||Hello Peter,
Thank you for your response.
What I am trying to do is to extract data from an XML column I have in one
of my tables. I am trying to make the data in the XML column appear as part
of the normal rowset.
I will run a SQL SELECT statement from my C# code expecting to receive the
data back in the form of a rowset (so I can read it as primitive types using
IDataReader).
Anyway, Uri's response works for me in most of the cases, it doesn't cover
(and I tried several things myself) the case where I need to extract multiple
values (i.e. rows) from a single XML document (stored in the xml column in a
given row in my table).
Thanks for the help
Asher.
"Peter Yang [MSFT]" wrote:

> Hello Asher,
> I understand that you'd like to insert a table from a xml type value. You
> may want to use openxml to do the job:
> CREATE TABLE MyStuff( product varchar(20), version int)
> go
> DECLARE @.docHandle int
> declare @.xmlDocument xml
> set @.xmlDocument = N'
> <MyStuff>
> <Product>Windows Vista</Product>
> <Version>1</Version>
> </MyStuff>
> '
> EXEC sp_xml_preparedocument @.docHandle OUTPUT, @.xmlDocument
> -- Use OPENXML to provide rowset consisting of customer data.
> INSERT Mystuff
> SELECT *
> FROM OPENXML(@.docHandle, N'/MyStuff', 2)
> WITH (proudct varchar(20) '/MyStuff/Product',
> version int '/MyStuff/Version')
> EXEC sp_xml_removedocument @.docHandle
> select * from mystuff
>
> If you have any comments or feedback, please feel free to let's know. Thank
> you.
> Best Regards,
> Peter Yang
> MCSE2000/2003, MCSA, MCDBA
> Microsoft Online Community Support
> ==================================================
> Get notification to my posts through email? Please refer to
> http://msdn.microsoft.com/subscriptions/managednewsgroups/default.aspx#notif
> ications
> <http://msdn.microsoft.com/subscriptions/managednewsgroups/default.aspx>.
> Note: The MSDN Managed Newsgroup support offering is for non-urgent issues
> where an initial response from the community or a Microsoft Support
> Engineer within 1 business day is acceptable. Please note that each follow
> up response may take approximately 2 business days as the support
> professional working with you may need further investigation to reach the
> most efficient resolution. The offering is not appropriate for situations
> that require urgent, real-time or phone-based interactions or complex
> project analysis and dump analysis issues. Issues of this nature are best
> handled working with a dedicated Microsoft Support Engineer by contacting
> Microsoft Customer Support Services (CSS) at
> <http://msdn.microsoft.com/subscriptions/support/default.aspx>.
> ==================================================
> This posting is provided "AS IS" with no warranties, and confers no rights.
>
|||Asher
I see what you mean
DECLARE @.idoc int
DECLARE @.doc varchar(1000)
SET @.doc ='
<AddressType name="Shipping">
<AddressRecord StreetLine1="111 Main"
City="Houston"
State="TX"
ZipCode = "11111" />
<AddressRecord StreetLine1="111 Left"
City="New_York"
State="NY"
ZipCode = "3333" />
</AddressType> '
EXEC sp_xml_preparedocument @.idoc OUTPUT, @.doc
SELECT *
FROM OPENXML (@.idoc, '/AddressType/AddressRecord',1)
WITH (City varchar(10))
"Asher F." <asherfoa@.community.nospam> wrote in message
news:F678FA42-605F-4425-BED3-EBEF75A47219@.microsoft.com...[vbcol=seagreen]
> Hi Uri,
> Thanks for the quick response, I understand the code you suggested and it
> works for me 80% of times. I would very much appreciate your help in this
> follow-up.
> To extend your example, lets say I have this:
> declare @.tAddress table (Address xml DEFAULT '<Address Record />' )
> INSERT INTO @.tAddress (Address)
> VALUES ('<AddressType name="Shipping">
> <AddressRecord StreetLine1="111 Main"
> City="Houston"
> State="TX"
> ZipCode = "11111" />
> <AddressRecord StreetLine1="222 Left"
> City="New York"
> State="NY"
> ZipCode = "2222" />
> </AddressType> ');
> INSERT INTO @.tAddress (Address)
> VALUES ('<AddressType name="Shipping">
> <AddressRecord StreetLine1="222 Main"
> City="SanDiego"
> State="CA"
> ZipCode = "2222" />
> </AddressType> ');
> Note that the XML in the first row has two AddressRecord elements.
> How do I make my result return a rowset with 3 rows representing all the
> values:
> SanDiego
> New York
> Houston
> Thanks in advance!
> --
> Asher.
>
> "Uri Dimant" wrote:
|||Or
SELECT address.value('(/AddressType/AddressRecord/@.City)[1]', 'varchar(50)')
FROM @.tAddress
UNION ALL
SELECT address.value('(/AddressType/AddressRecord/@.City)[2]', 'varchar(50)')
FROM @.tAddress
"Uri Dimant" <urid@.iscar.co.il> wrote in message
news:O9Qv3C1THHA.5108@.TK2MSFTNGP06.phx.gbl...
> Asher
> I see what you mean
> DECLARE @.idoc int
> DECLARE @.doc varchar(1000)
> SET @.doc ='
> <AddressType name="Shipping">
> <AddressRecord StreetLine1="111 Main"
> City="Houston"
> State="TX"
> ZipCode = "11111" />
> <AddressRecord StreetLine1="111 Left"
> City="New_York"
> State="NY"
> ZipCode = "3333" />
> </AddressType> '
> EXEC sp_xml_preparedocument @.idoc OUTPUT, @.doc
> SELECT *
> FROM OPENXML (@.idoc, '/AddressType/AddressRecord',1)
> WITH (City varchar(10))
>
>
>
> "Asher F." <asherfoa@.community.nospam> wrote in message
> news:F678FA42-605F-4425-BED3-EBEF75A47219@.microsoft.com...
>
|||Hi Uri,
The OPENXML is probably not the solution since I have the XML document
stored in a column.
As for your second solution, it looks like it is in the right direction,
however I don't see how I can scale it to the case I have N (or an unkown
number of) AddressRecord nodes in each XML document (i.e. one row's XML
column might contain zero AddressRecord elements and another row might
contain 10...)
How do I scale this solution ?
Asher.
"Uri Dimant" wrote:

> Or
> SELECT address.value('(/AddressType/AddressRecord/@.City)[1]', 'varchar(50)')
> FROM @.tAddress
> UNION ALL
> SELECT address.value('(/AddressType/AddressRecord/@.City)[2]', 'varchar(50)')
> FROM @.tAddress
>
>
>
> "Uri Dimant" <urid@.iscar.co.il> wrote in message
> news:O9Qv3C1THHA.5108@.TK2MSFTNGP06.phx.gbl...
>
>
|||See Tibor's reply , it is what you wanted
"Asher F." <asherfoa@.community.nospam> wrote in message
news:38DFF454-06D5-4E00-B623-F5FB216D87EC@.microsoft.com...[vbcol=seagreen]
> Hi Uri,
> The OPENXML is probably not the solution since I have the XML document
> stored in a column.
> As for your second solution, it looks like it is in the right direction,
> however I don't see how I can scale it to the case I have N (or an unkown
> number of) AddressRecord nodes in each XML document (i.e. one row's XML
> column might contain zero AddressRecord elements and another row might
> contain 10...)
> How do I scale this solution ?
> --
> Asher.
>
> "Uri Dimant" wrote:
|||Thank you Tibor, it looks like exactlly what I need.
I am now reading the relevant documentation about the functions you used so
I can get a deeper understanding.
Thank you very much for your help!
Asher.
"Tibor Karaszi" wrote:

> Did you look at my suggestion? It produces exactly the result you asked for...
> --
> Tibor Karaszi, SQL Server MVP
> http://www.karaszi.com/sqlserver/default.asp
> http://sqlblog.com/blogs/tibor_karaszi
>
> "Asher F." <asherfoa@.community.nospam> wrote in message
> news:161F4EC4-7C19-430D-91EF-D2684D840D9D@.microsoft.com...
>