[AccessD] Making SQL Server NewID() work with a dataset in .net

Hi Charlotte

I would be both.
Is this a SQL Server issue only or would I face it with other SQL
server engines too?

Is this a limitation of all datagrids or do third-party grids manage to
deal with it?

/gustav

>>> cfoust at infostatsystems.com 23-02-2005 17:30:25 >>>
Yes, it is.  The problem only occurs with a SQL Server backend.  An
Access backend doesn't experience the problem.  One of the nice things
about .Net is that you can create separate data providers to handle
SQL
Server and Access, so the details are not dealt with by the UI
programmer, only by the data tier programmer.  Of course, if you
happen
to be both ... <g>

Charlotte Foust


-----Original Message-----
From: Gustav Brock [mailto:Gustav at cactus.dk] 
Sent: Wednesday, February 23, 2005 1:09 AM
To: accessd at databaseadvisors.com 
Subject: [AccessD] Making SQL Server NewID() work with a dataset in
.net


Hi all using dotnet.

Snipped this from the ElementK Tips. 
I have not used the datagrid from dotnet and have to ask if this
really
is the way to handle this task? From an Access view it seems like
moving
to the stone age.

Or would I just use another more clever grid?

/gustav


Making SQL Server primary key NEWID() default values work with a
Visual
Basic .NET DataSet (Visual Basic .NET/SQL 2000)

Designing a SQL Server table to use the uniqueidentifier datatype for
its primary key column and then giving it a default with the NEWID()
function is a great way to manage your data; however, there can be a
slight snag if you use ADO.NET's DataSet in Visual Basic .NET with
such
a table. In this scenario, adding records in a DataGrid in Visual
Basic
.NET will cause an error. The primary key field doesn't allow NULL
values. We know that the key field will be defaulted on the SQL Server
side, but Visual Basic .NET enforces the NULL before the data is
pushed
to SQL Server. The solution is easy, since we have access to the
dataset
XML. First, we can delete the key from the dataset table by
right-clicking on the table and selecting the Delete Key. The field
isn't deleted, just the key indicator for the field. Now the field
won't
require a unique value, but we still need to allow the value to be
NULL.
This is done by adding minOccurs="0" to the <xs:element> code for this
field, as in the example below:

	<xs:element
		name="KeyFieldName"
		msdata:DataType="System.Guid, mscorlib,
version=1.0.3300.0,
Culture=neutral, PublicKeyToken=b77a5c561934e089"
		type="xs:string"
		minOccurs="0" />

Now records can be added to the grid without raising any errors, and
SQL
Server will use the NEWID() default for primary key column values once
the table is updated.