[Note] In ADO 2.x, we used the term recordset to refer to a collection of records returned from a query. In ADO.NET we use the term rowset. [End Note]
If we wanted to use a recordset object in ADO 2.x to store the results of a query and loop through these results, we could do so by calling its open and movenext methods. We could also check whether there were any records beyond our cursor by checking its eof variable, like this:
|objRS.ActiveConnection = objConn |
objRS.LockType = 1 'adLockReadOnly
objRS.CursorType = 0 'adOpenForwardOnly
objRS.Open "SELECT au_fname + ' ' + au_lname as name FROM authors"
while not objRS.EOF
Response.Write objRS("name") & "<br>"
In ADO.NET, the OleDbDataReader and SqlDataReader classes can be used to hold rowsets. Both of these classes cannot be directly instantiated, and can only be created as a result of a call to the ExecuteReader method, like this:
|string strQuery = "SELECT au_fname + ' ' + au_lname as name FROM authors"; |
SqlCommand objCmd = new SqlCommand(strQuery, objConn);
objCmd.CommandType = CommandType.Text;
SqlDataReader objDR = objCmd.ExecuteReader();
Response.Write(objDR["name"] + "<br>");
The Read() method of the SqlDataReader in our example above advances the rowsets cursor and gives us access to the next row. The SqlDataReader class has an overloaded indexer to accept either the name or index of the field to return. We could just as easily specify a field index rather that a name, like this:
Response.Write(objDR + "
The return type from the SqlDataReaders indexer is actually an object, which, in our example above is implicitly converted to a string. If we knew that a field was going to be a specific data type, then we could specify an explicit case using C style casting:
// Perform an explicit case from object to int32
int intAge = (int)objDR;
[Note] When using both the OleDbDataReader and SqlDataReader classes, field indexes start at 0 and not 1. [End Note]