.nettransactionstransactionscopesystem.transactions

what is practical use of System.Transactions?

I have seen System.Transactions namespace, and wondered, can I actually make a RDMBS with this namespace usage?

But when I saw some examples, I do not understand how System.Transactions does anything beyond simple try catch and getting us success/failure result?

This is the example on MSDN's website, I know it may be very simple but I am unable to understand the benefit in this sample, can someone tell me what is difference between simple try/catch and Transaction scope in this following sample.

If I am supposed to make a RDBMS (create my own RDMBS), I understand we have to write lots of logs to disk of the operations we execute and at the end we undo those operations in the case of rollback, but here there is nothing about undoing anything.

// This function takes arguments for 2 connection strings and commands to create a transaction 
// involving two SQL Servers. It returns a value > 0 if the transaction is committed, 0 if the 
// transaction is rolled back. To test this code, you can connect to two different databases 
// on the same server by altering the connection string, or to another 3rd party RDBMS by 
// altering the code in the connection2 code block.
static public int CreateTransactionScope(
    string connectString1, string connectString2,
    string commandText1, string commandText2)
{
    // Initialize the return value to zero and create a StringWriter to display results.
    int returnValue = 0;
    System.IO.StringWriter writer = new System.IO.StringWriter();

    try
    {
        // Create the TransactionScope to execute the commands, guaranteeing
        // that both commands can commit or roll back as a single unit of work.
        using (TransactionScope scope = new TransactionScope())
        {
            using (SqlConnection connection1 = new SqlConnection(connectString1))
            {
                // Opening the connection automatically enlists it in the 
                // TransactionScope as a lightweight transaction.
                connection1.Open();

                // Create the SqlCommand object and execute the first command.
                SqlCommand command1 = new SqlCommand(commandText1, connection1);
                returnValue = command1.ExecuteNonQuery();
                writer.WriteLine("Rows to be affected by command1: {0}", returnValue);

                // If you get here, this means that command1 succeeded. By nesting
                // the using block for connection2 inside that of connection1, you
                // conserve server and network resources as connection2 is opened
                // only when there is a chance that the transaction can commit.   
                using (SqlConnection connection2 = new SqlConnection(connectString2))
                {
                    // The transaction is escalated to a full distributed
                    // transaction when connection2 is opened.
                    connection2.Open();

                    // Execute the second command in the second database.
                    returnValue = 0;
                    SqlCommand command2 = new SqlCommand(commandText2, connection2);
                    returnValue = command2.ExecuteNonQuery();
                    writer.WriteLine("Rows to be affected by command2: {0}", returnValue);
                }
            }

            // The Complete method commits the transaction. If an exception has been thrown,
            // Complete is not  called and the transaction is rolled back.
            scope.Complete();

        }

    }
    catch (TransactionAbortedException ex)
    {
        writer.WriteLine("TransactionAbortedException Message: {0}", ex.Message);
    }
    catch (ApplicationException ex)
    {
        writer.WriteLine("ApplicationException Message: {0}", ex.Message);
    }

    // Display messages.
    Console.WriteLine(writer.ToString());

    return returnValue;
}

In above example what are we committing? I guess SQL Client library will do everything right? Does this mean that System.IO.StringWriter will either contain all success text or all failure text? or is there any locking between scope of TransactionScope?

Solution

  • First of all TransactionScope is not the same as try/catch. TransactionScope is by the name scope of a transaction. Transaction in scope has to be explicitly commited by calling Complete on the scope. Any other case (including exception raised in scope) results in finishing using block which disposes the scope and implicitly rollback the incomplete transaction but it will not handle the exception.

    In basic scenarios transaction from System.Transactions behaves same as db client transaction. System.Transactions provides following additional features: