Beginning C# 2005 Databases From Novice to Professional phần 8 doc

You specify the command for a stored procedure call, but you don’t specify the stored procedure to call // Specify that a stored procedure is to be executedcmd.CommandType = CommandType.

10. Run the program with Ctrl+F5 Click the ADO.NET Exception-1 button, and you’llsee the message box shown in Figure 13-3 Click OK.

11. When the message box in Figure 13-4 appears, click OK, then close the window

How It Works

It would be highly unusual to miss setting the CommandTextproperty However, this is an

expedient way to cause an ADO.NET exception You specify the command for a stored

procedure call, but you don’t specify the stored procedure to call

// Specify that a stored procedure is to be executedcmd.CommandType = CommandType.StoredProcedure;

// Deliberately fail to specify the procedure// cmd.CommandText = "sp_Select_AllEmployees";

so when you call the ExecuteReadermethod, you get an exception, as shown in

Figure 13-2 Though it’s an unhandled exception, it still gives you an accurate diagnostic

ExecuteReader: CommandText property has not been intiailized

and it even gives you the option to continue or quit However, leaving this decision to

users isn’t a very good idea

After seeing what happens without handling the exception, you place the call in

atryblock:

Figure 13-3.Handled exception message

Figure 13-4.Message from finally block

try{// Open connectionconn.Open();

// Create data readerSqlDataReader dr = cmd.ExecuteReader();

// Close readerdr.Close();

}

To handle the exception yourself, you code two catchclauses:

catch (System.Data.SqlClient.SqlException ex){

string str;

str = "Source:" + ex.Source;

str += "\n" + "Exception Message:" + ex.Message;

MessageBox.Show (str, "Database Exception");

}catch (System.Exception ex){

string str;

str = "Source:" + ex.Source;

str += "\n" + "Exception Message:" + ex.Message;

MessageBox.Show (str, "Non-Database Exception");

}

In the first catchclause, you specify a database exception type The second catchclause, which produces the message box in Figure 13-3, is a generic block that catchesall types of exceptions Note the caption of the message box in this catchblock It says

Non-Database Exception Although you may think that a failure to specify a command

string is a database exception, it’s actually an ADO.NET exception; in other words, thiserror is trapped before it gets to the database server

When the button is clicked, since the CommandTextproperty isn’t specified, an tion is thrown and caught by the second catchclause Even though a catchclause forSqlExceptionis provided, the exception is a System.InvalidOperationException—a com-mon exception thrown by the CLR, not a database exception

excep-The exception message indicates where the problem occurred: in the ExecuteReadermethod The finallyblock checks if the connection is open and, if it is, closes it and gives

a message to that effect Note that in handling the exception, you don’t terminate theapplication:

{

if (conn.State == ConnectionState.Open){

MessageBox.Show ("Finally block closing the connection", "Finally");

conn.Close();

}}

Try It Out: Handling an ADO.NET Exception (Part 2)

Let’s try another example of an ADO.NET exception You’ll execute a stored procedure

and then reference a nonexistent column in the returned dataset This will throw an

ADO.NET exception This time, you’ll code a specific catchclause to handle the

exception:

1. Use the sp_Select_All_Employeesstored procedure you created in Chapter 12

If you haven’t created it already, please go to Chapter 12 and follow the steps in

“Try It Out: Creating and Executing a Trivial Stored Procedure.”

2. Insert the code in Listing 13-3 into the body of the button2_Clickmethod

Listing 13-3.button2_Click()

// Create connectionSqlConnection conn = new SqlConnection(@"

data source = \sqlexpress;

integrated security = true;

database = northwind

");

// Create commandSqlCommand cmd = conn.CreateCommand();

// Specify that a stored procedure is to be executedcmd.CommandType = CommandType.StoredProcedure;

cmd.CommandText = "sp_Select_All_Employees";

try{// Open connectionconn.Open();

// Create data readerSqlDataReader dr = cmd.ExecuteReader();

// Access nonexistent columnstring str = dr.GetValue(20).ToString();

// Close readerdr.Close();

}catch (System.InvalidOperationException ex){

catch (System.Data.SqlClient.SqlException ex){

string str;

str = "Source: " + ex.Source;

str += "\n" + "Exception Message: " + ex.Message;MessageBox.Show (str, "Database Exception");}

catch (System.Exception ex){

string str;

str = "Source: " + ex.Source;

str += "\n" + "Exception Message: " + ex.Message;MessageBox.Show (str, "Non-Database Exception");}

if (conn.State == ConnectionState.Open){

MessageBox.Show ("Finally block closing the connection", "Finally");

conn.Close();

}}

■ Tip Testing whether a connection is open before attempting to close it isn’t actually necessary The

Closemethod doesn’t throw any exceptions, and calling it multiple times on the same connection, even

if it’s already closed, causes no errors

3. Run the program with Ctrl+F5 Click the ADO.NET Exception-2 button, and you’llsee the message box shown in Figure 13-5 Click OK When the finallyblock mes-sage appears, click OK, then close the window

4. For a quick comparison, now generate a SQL Server exception, an error thatoccurs within the database Alter the name of the stored procedure in the code

to a name that doesn’t exist within the Northwind database For example:

cmd.CommandText = "sp_Select_No_Employees";

Figure 13-5.Handling a specific ADO.NET exception

5. Run the program with Ctrl+F5 Click the ADO.NET Exception-2 button, and you’llsee the message box shown in Figure 13-6 Click OK When the finallyblock mes-sage appears, click OK, then close the window.

How It Works

First you create the data reader and try to access an invalid column:

// Create data readerSqlDataReader dr = cmd.ExecuteReader();

// Access nonexistent columnstring str = dr.GetValue(20).ToString();

An exception is thrown, because you tried to get the value of column 20, whichdoesn’t exist You add a new catchclause to handle this kind of ADO.NET error:

catch (System.InvalidOperationException ex){

string str;

str = "Source: " + ex.Source;

str += "\n" + "Message: "+ ex.Message;

str += "\n" + "\n";

str += "\n" + "Stack Trace: " + ex.StackTrace;

MessageBox.Show (str, "Specific Exception");

}When an exception of type System.InvalidOperationExceptionis thrown, this catchclause executes, displaying the source, message, and stack trace for the exception.Without this specific catchclause, the generic catchclause would have handled theexception (Try commenting out this catchclause and reexecuting the code to seewhich catchclause handles the exception.)

Figure 13-6.Handling a SQL Server exception

Next, you run the program for a nonexistent stored procedure:

// Specify that a stored procedure is to be executedcmd.CommandType = CommandType.StoredProcedure;

cmd.CommandText = "sp_Select_No_Employees";

You catch your (first) database exception withcatch (System.Data.SqlClient.SqlException ex)which leads into the next topic: handling exceptions thrown by the database manager

Handling Database Exceptions

An exception of type System.Data.SqlClient.SqlExceptionis thrown when SQL Server

returns a warning or error This class is derived from System.SystemExceptionand is sealed

so it can’t be inherited, but it has several useful members that you can interrogate to

obtain valuable information about the exception

An instance of SqlExceptionis thrown whenever the NET data provider for SQLServer encounters an error or warning from the database Table 13-1 describes the prop-

erties of this class that provide information about the exception

Table 13-1.SqlExceptionProperties

Property Name Description

Class Gets the severity level of the error returned from the SqlClient data provider.

The severity level is a numeric code that’s used to indicate the nature of the error Levels 1 to 10 are informational errors; 11 to 16 are user-level errors; and

17 to 25 are software or hardware errors At level 20 or greater, the connection

is usually closed.

Data Gets a collection of key-value pairs that contain user-defined information.

ErrorCode The HRESULT of the error.

Errors Contains one or more SqlError objects that have detailed information about

the exception This is a collection that can be iterated through.

HelpLink The help file associated with this exception.

InnerException Gets the exception instance that caused the current exception.

LineNumber Gets the line number within the Transact-SQL command batch or stored

procedure that generated the exception.

Message The text describing the exception.

Number The number that identifies the type of exception.

Continued

Table 13-1.Continued

Property Name Description

Procedure The name of the stored procedure that generated the exception.

Server The name of the computer running the instance of SQL Server that generated

the exception.

Source The name of the provider that generated the exception.

StackTrace A string representation of the call stack when the exception was thrown State Numeric error code from SQL Server that represents an exception, warning, or

“no data found” message For more information, see SQL Server Books Online TargetSite The method that throws the current exception.

When an error occurs within SQL Server, it uses a T-SQL RAISERRORstatement to raise

an error and send it back to the calling program A typical error message looks like thefollowing:

Server: Msg 2812, Level 16, State 62, Line 1

Could not find stored procedure 'sp_DoesNotExist'

In this message, 2812represents the error number, 16represents the severity level,and 62represents the state of the error

You can also use the RAISERRORstatement to display specific messages within a storedprocedure The RAISERRORstatement in its simplest form takes three parameters The firstparameter is the message itself that needs to be shown The second parameter is theseverity level of the error Any users can use severity levels 11 through 16 They representmessages that can be categorized as information, software, or hardware problems Thethird parameter is an arbitrary integer from 1 through 127 that represents informationabout the state or source of the error

Let’s see how a SQL error, raised by a stored procedure, is handled in C# You’ll ate a stored procedure and use the following T-SQL to raise an error when the number

cre-of orders in the Orderstable exceeds ten:

if @orderscount > 10

raiserror ('Orders Count is greater than 10 - Notify the Business Manager',16,

1)Note that in this RAISERRORstatement, you specify a message string, a severity level

of 16, and an arbitrary state number of 1 When a RAISERRORstatement that you writecontains a message string, the error number is given automatically as 50000 When

SQL Server raises errors using RAISERROR, it uses a predefined dictionary of messages to

give out the corresponding error numbers (See SQL Server Books Online to learn how

to add your own messages to SQL Server’s predefined messages.)

Try It Out: Handling a Database Exception (Part 1): RAISERROR

Let’s raise a database error and handle the exception:

1. Add a button to the Database tab page and change its Textproperty to DatabaseException-1 Add a label to the right of this button, and change its Textproperty

to Calls a stored procedure that uses RAISERROR

2. Add a second button to the tab page, and change its Textproperty to DatabaseException-2 Add a label to the right of this button, and change its Textproperty

to Calls a stored procedure that encounters an error

3. Add a third button to the tab page, and change its Textproperty to DatabaseException-3 Add a label to the right of this button, and change its Textproperty

to Creates multiple SqlError objects The layout should look like Figure 13-7

4. Using SSMSE, create a stored procedure in Northwind named sp_DbException_1, asfollows:

create procedure sp_DbException_1as

set nocount on

declare @ordercount int

Figure 13-7.Database tab page

@ordercount = count(*)from

orders

if @ordercount > 10raiserror ('Orders Count is greater than 10 - Notify the Business Manager',16,

1)

5. Add the code in Listing 13-4 to the button3_Clickmethod

Listing 13-4.button3_Click()

// Create connectionSqlConnection conn = new SqlConnection(@"

data source = \sqlexpress;

integrated security = true;

database = northwind

");

// Create commandSqlCommand cmd = conn.CreateCommand();

// Specify that a stored procedure to be executedcmd.CommandType = CommandType.StoredProcedure;

cmd.CommandText = "sp_DbException_1";

try{// Open connectionconn.Open();

// Execute stored procedurecmd.ExecuteNonQuery();

}catch (System.Data.SqlClient.SqlException ex){

string str;

str = "Source: " + ex.Source;

string str;

str = "Source: " + ex.Source;

str += "\n" + "Exception Message: " + ex.Message;

MessageBox.Show (str, "General Exception");

}finally{

if (conn.State == ConnectionState.Open){

6. Run the program with Ctrl+F5, then click the Database Exception-1button You’llsee the message box shown in Figure 13-8 Click OK to close the message box,then OK to close the next one, then close the window

Figure 13-8.RAISERROR Database Exception message

Observe the caption and contents of the message box The source, message, name

of the stored procedure, exact line number where the error was found, and name of theserver are all displayed You obtain this detailed information about the exception fromthe SqlExceptionobject

ordersThen, if @ordercountis greater than ten, you raise an error using the RAISERRORstatement:

if @ordercount > 10raiserror ('Orders Count is greater than 10 - Notify the Business Manager',16,

1)Then, in the button3_Clickmethod, you execute the stored procedure using theExecuteNonQuerymethod within a tryblock:

try{// Open connectionconn.Open();

// Create data readercmd.ExecuteNonQuery();

}When the stored procedure executes, the RAISERRORstatement raises an error, which

is converted to an exception by ADO.NET The following code handles the exception:

catch (System.Data.SqlClient.SqlException ex){

Try It Out: Handling a Database Exception (Part 2):

Stored Procedure Error

Now let’s see what happens when a statement in a stored procedure encounters an error

You’ll create a stored procedure that attempts an illegal INSERT, and then you’ll extract

information from the SqlExceptionobject:

1. Using SSMSE, create a stored procedure in Northwind named sp_DbException_2, asfollows:

create procedure sp_DBException_2as

set nocount oninsert into employees(

employeeid,firstname)

values (50, 'Cinderella')

2. Insert the code in Listing 13-5 into the button4_Clickmethod:

Listing 13-5.button4_Click()

// Create connectionSqlConnection conn = new SqlConnection(@"

data source = \sqlexpress;

integrated security = true;

database = northwind

");

// Create commandSqlCommand cmd = conn.CreateCommand();

// Specify stored procedure to be executedcmd.CommandType = CommandType.StoredProcedure;

cmd.CommandText = "sp_DbException_2";

try{// Open connectionconn.Open();

// Execute stored procedurecmd.ExecuteNonQuery();

}catch (System.Data.SqlClient.SqlException ex){

MessageBox.Show (str, "Database Exception");

}catch (System.Exception ex)

{string str;

str = "Source: " + ex.Source;

str += "\n" + "Exception Message: " + ex.Message;

MessageBox.Show (str, "ADO.NET Exception");

}finally{

if (conn.State == ConnectionState.Open){

3. Run the program with Ctrl+F5, and then click the Database Exception-2button

You’ll see the message box shown in Figure 13-9 Click OK to close the messagebox, then OK to close the next one, then close the window

How It Works

The stored procedure tries to insert a new employee into the Employeestable:

insert into employees(

employeeid,firstname)

values (50, 'Cinderella')

Figure 13-9.Stored procedure Database Exception message

However, since the EmployeeIDcolumn in the Employeestable is an IDENTITYcolumn,you can’t explicitly assign a value to it.

■ Tip Actually, you can—as the message indicates—if you use SET IDENTITY_INSERT employees OFF

in the stored procedure before you attempt the INSERT This would allow you to insert explicit EmployeeIDvalues, but this seldom is, or should be, done

When this SQL error occurs, the specific SqlException catchclause traps it anddisplays the information The finallyblock then closes the connection

It’s possible for stored procedures to encounter several errors You can trap anddebug these using the SqlExceptionobject, as you’ll see next

Try It Out: Handling a Database Exception (Part 3):

data source = \sqlexpress;

integrated security = true;

database = northwnd

");

// Create commandSqlCommand cmd = conn.CreateCommand();

// Specify stored procedure to be executedcmd.CommandType = CommandType.StoredProcedure;

cmd.CommandText = "sp_DbException_2";

try{// Open connectionconn.Open();

// Execute stored procedurecmd.ExecuteNonQuery();

}catch (System.Data.SqlClient.SqlException ex){

}catch (System.Exception ex){

string str;

str = "Source: " + ex.Source;

str += "\n" + "Exception Message: " + ex.Message;

MessageBox.Show (str, "ADO.NET Exception");

}finally{

if (conn.State == ConnectionState.Open){

2. Run the program with Ctrl+F5, and then click the Database Exception-2button.You’ll see the message box shown in Figure 13-10

Observe that two items are found in the Errorscollection, and their error numbersare different

How It Works

In the connection string, you specify a database that doesn’t exist on the server; here youmisspell Northwindas Northwnd:

// Create connectionSqlConnection conn = new SqlConnection(@"

data source = \sqlexpress;

integrated security = true;

database = northwnd

");

When you try to open the connection, an exception of type SqlExceptionis thrown,and you loop through the items of the Errorscollection and get each Errorobject usingits indexer:

catch (System.Data.SqlClient.SqlException ex){

string str ="";

for (int i = 0; i < ex.Errors.Count; i++)

Figure 13-10.Handling multiple database errors

{str +=

"\n" + "Index #" + i + "\n"

+ "Exception: " + ex.Errors[i].ToString() + "\n"

+ "Number: " + ex.Errors[i].Number.ToString() + "\n"

;}MessageBox.Show (str, "Database Exception");

}This example shows that the SqlExceptionobject carries detailed information aboutevery SQL error in its Errorscollection

Summary

In this chapter, you saw how to handle exceptions thrown by ADO.NET and by SQL

Server In particular, you learned how to handle both single and multiple database

errors with the System.Data.SqlClient.SqlExceptionclass

In the next chapter, you’ll look at transactions and how to maintain databaseintegrity when multiple users are working concurrently

Using Transactions

Atransaction is a set of operations performed so all operations are guaranteed to

suc-ceed or fail as one unit

A common example of a simple transaction is transferring money from a checkingaccount to a savings account This involves two operations: deducting money from the

checking account and adding it to the savings account Both must succeed together and

be committed to the accounts, or both must fail together and be rolled back so that the

accounts are maintained in a consistent state Under no circumstances should money

be deducted from the checking account but not added to the savings account (or vice

versa) By using a transaction, both operations can be guaranteed to succeed or fail

together

Transactions may comprise many individual operations and even other transactions

Transactions are essential for maintaining data integrity, both for multiple related

oper-ations and when multiple users update the database concurrently

In this chapter, we’ll cover:

• When to use transactions

• The ACID properties of a transaction

• How to code transactions

When to Use Transactions

You should use transactions when several operations must succeed or fail as a unit

The following are some frequent scenarios where you must use transactions:

• In batch processing, where multiple rows must be inserted or deleted as

• When modifying data in two or more databases, concurrently

• In distributed transactions, where data is manipulated in databases on differentservers

When you use transactions, you place locks on data pending permanent change tothe database No other operations can take place on locked data until you lift the lock.You could lock anything from a single row up to the whole database This is called con- currency; that is, concurrency is how the database handles multiple updates at one time

In the bank example, locks ensure that two separate transactions don’t access thesame accounts at the same time If they did, either deposits or withdrawals could be lost

■ Note It’s important to keep transactions pending for the shortest period of time A lock stops others fromaccessing the locked database resource Too many locks, or locks on frequently accessed resources, canseriously degrade performance

Understanding ACID Properties

A transaction is characterized by four properties, often referred to as the ACID properties:

atomicity, consistency, isolation, and durability

Atomicity: A transaction is atomic if it’s regarded as a single action rather than

a collection of separate operations So, a transaction succeeds and is committed tothe database only when all the separate operations succeed On the other hand, if

a single operation fails during the transaction, everything is considered to have failedand must be undone (rolled back) if it has already taken place In the case of theorder-entry system of the Northwind database, when you enter an order into theOrdersand Order Detailstables, data will be saved together in both tables, or

it won’t be saved at all

Consistency: The transaction should leave the database in a consistent state—

whether or not it completed successfully The data modified by the transaction mustcomply with all the constraints placed on the columns in order to maintain dataintegrity In the case of Northwind, you can’t have rows in the Order Detailstablewithout a corresponding row in the Orderstable, as this would leave the data in

an inconsistent state

Isolation: Every transaction has a well-defined boundary One transaction shouldn’t

affect other transactions running at the same time Data modifications made by onetransaction must be isolated from the data modifications made by all other trans-actions A transaction sees data in the state it was in before another concurrenttransaction modified it, or it sees the data after the second transaction has com-pleted, but it doesn’t see an intermediate state

Durability: Data modifications that occur within a successful transaction are kept

permanently within the system regardless of what else occurs Transaction logs aremaintained so that should a failure occur the database can be restored to its originalstate before the failure As each transaction is completed a row is entered in the data-base transaction log If you have a major system failure that requires the database to

be restored from a backup, you could then use this transaction log to insert (roll ward) any successful transactions that had taken place

for-Every database server that offers support for transactions enforces these four ties automatically All you need to do is create the transactions in the first place, which is

proper-what you’ll look at next

How to Code Transactions

The following three T-SQL statements control transactions in SQL Server:

BEGIN TRANSACTION: This marks the beginning of a transaction

COMMIT TRANSACTION: This marks the successful end of a transaction It signals thedatabase to save the work

ROLLBACK TRANSACTION: This denotes that a transaction hasn’t been successful and nals the database to roll back to the state it was in prior to the transaction

sig-Note that there is no END TRANSACTIONstatement Transactions end on (explicit orimplicit) commits and rollbacks

■ Note All our example programs to this point have run in SQL Server’s default autocommit mode, i.e.,

SQL Server implicitly committed or rolled back each statement depending on its success or failure

Auto-commit mode can’t provide atomicity and consistency for multiple statements It’s also a potentially

pro-hibitively expensive way to do things if you need to perform many (thousands or millions, but sometimes

even just hundreds) of inserts How to design transactions to handle such heavy loads is beyond the scope

of this book, but what you learn here forms the basis for designing solutions for all transactional situations

Coding Transactions in T-SQL

We’ll use a stored procedure to practice coding transactions in SQL It’s an intentionallyartificial example but representative of transaction processing fundamentals It keepsthings simple so you can focus on the important issue of what can happen in a trans-action That’s what you really need to understand, especially when you later code thesame transaction in C#

■ Warning Using ROLLBACKand COMMITinside stored procedures typically requires careful consideration

of what transactions may already be in progress and led to the stored procedure call Our example runs byitself, so we’re not concerned with this here, but you should always consider whether it’s a potential issue

Try It Out: Coding a Transaction in T-SQL

Let’s code a transaction to both add a customer to and delete one from the NorthwindCustomerstable Customershas eleven columns, but only two, CustomerIDand CompanyName,don’t allow nulls, so we’ll use just those columns for insertion We’ll also use arbitrarycustomer IDs to make it easy to find the rows we manipulate when viewing customerssorted by ID

1. In SSMSE, create a stored procedure named sp_Trans_Test, using the code in ing 14-1 Note that you’re using several new SQL statements for stored procedureprogramming We don’t explain the ones not involved in transactions, but theirmeaning and usage should be obvious

declare @inserr intdeclare @delerr intdeclare @maxerr int

set @maxerr = 0

begin transaction

Add a customerinsert into customers(

customerid,companyname)

values(@newcustid, @newconame)

Save error numberset @inserr = @@error

if @inserr > @maxerrset @maxerr = @inserr Delete a customerdelete from customerswhere

customerid = @oldcustid Save error numberset @delerr = @@error

if @delerr > @maxerrset @maxerr = @delerr

If an error occurred, roll back

if @maxerr <> 0beginrollbackprint 'Transaction rolled back'end

elsebegincommitprint 'Transaction committed'end

print 'INSERT error number:' + cast(@inserr as nvarchar(8))print 'DELETE error number:' + cast(@delerr as nvarchar(8)) return @maxerr

2. Run the stored procedure from Object Explorer When prompted, enter just aforboth @newcustidand @newconameand zfor @oldcustid A new edit window willappear, displaying your input values The Results window should show a ReturnValue of zero Click on the Messages tab and you should see the same messages

Figure 14-1.Rows inserted (but not deleted) in a transaction

Figure 14-2.Row inserted in a transaction