Chapter 5 Connector/Net Programming

Once you have created a connection string it can be used to open a connection to the MySQL server.

The following code is used to create a MySqlConnection object, assign the connection string, and open the connection.

Connector/Net can also connect using the native Windows authentication plugin. See Section 5.5, “Using the Windows Native Authentication Plugin” for details.

You can further extend the authentication mechanism by writing your own authentication plugin. See Section 5.6, “Writing a Custom Authentication Plugin” for details.

Visual Basic Example

Dim conn As New MySql.Data.MySqlClient.MySqlConnection
Dim myConnectionString as String

myConnectionString = "server=127.0.0.1;" _
            & "uid=root;" _
            & "pwd=12345;" _
            & "database=test;"

Try
  conn.ConnectionString = myConnectionString
  conn.Open()

Catch ex As MySql.Data.MySqlClient.MySqlException
  MessageBox.Show(ex.Message)
End Try

C# Example

MySql.Data.MySqlClient.MySqlConnection conn;
string myConnectionString;

myConnectionString = "server=127.0.0.1;uid=root;" +
    "pwd=12345;database=test;";

try
{
    conn = new MySql.Data.MySqlClient.MySqlConnection();
    conn.ConnectionString = myConnectionString;
    conn.Open();
}
catch (MySql.Data.MySqlClient.MySqlException ex)
{
    MessageBox.Show(ex.Message);
}

You can also pass the connection string to the constructor of the MySqlConnection class:

Visual Basic Example

Dim myConnectionString as String

myConnectionString = "server=127.0.0.1;" _
              & "uid=root;" _
              & "pwd=12345;" _
              & "database=test;"

Try
    Dim conn As New MySql.Data.MySqlClient.MySqlConnection(myConnectionString)
    conn.Open()
Catch ex As MySql.Data.MySqlClient.MySqlException
   MessageBox.Show(ex.Message)
End Try

C# Example

MySql.Data.MySqlClient.MySqlConnection conn;
string myConnectionString;

myConnectionString = "server=127.0.0.1;uid=root;" +
    "pwd=12345;database=test;";

try
{
    conn = new MySql.Data.MySqlClient.MySqlConnection(myConnectionString);
    conn.Open();
}
catch (MySql.Data.MySqlClient.MySqlException ex)
{
    MessageBox.Show(ex.Message);
}

Once the connection is open it can be used by the other Connector/Net classes to communicate with the MySQL server.

Because connecting to an external server is unpredictable, it is important to add error handling to your .NET application. When there is an error connecting, the MySqlConnection class will return a MySqlException object. This object has two properties that are of interest when handling errors:

When handling errors, you can your application's response based on the error number. The two most common error numbers when connecting are as follows:

The following code shows how to adapt the application's response based on the actual error:

Visual Basic Example

Dim myConnectionString as String

myConnectionString = "server=127.0.0.1;" _
          & "uid=root;" _
          & "pwd=12345;" _
          & "database=test;"

Try
    Dim conn As New MySql.Data.MySqlClient.MySqlConnection(myConnectionString)
    conn.Open()
Catch ex As MySql.Data.MySqlClient.MySqlException
    Select Case ex.Number
        Case 0
            MessageBox.Show("Cannot connect to server. Contact administrator")
        Case 1045
            MessageBox.Show("Invalid username/password, please try again")
    End Select
End Try

C# Example

MySql.Data.MySqlClient.MySqlConnection conn;
string myConnectionString;

myConnectionString = "server=127.0.0.1;uid=root;" +
    "pwd=12345;database=test;";

try
{
    conn = new MySql.Data.MySqlClient.MySqlConnection(myConnectionString);
    conn.Open();
}
    catch (MySql.Data.MySqlClient.MySqlException ex)
{
    switch (ex.Number)
    {
        case 0:
            MessageBox.Show("Cannot connect to server.  Contact administrator");
            break;
        case 1045:
            MessageBox.Show("Invalid username/password, please try again");
            break;
    }
}

MySqlConnection myConnection = new MySqlConnection("server=127.0.0.1;uid=root;" +
    "pwd=12345;database=test;Charset=latin1;");

The GetSchema() method of the connection object can be used to retrieve schema information about the database currently connected to. The schema information is returned in the form of a DataTable. The schema information is organized into a number of collections. Different forms of the GetSchema() method can be used depending on the information required. There are three forms of the GetSchema() method:

using System;
using System.Data;
using System.Text;
using MySql.Data;
using MySql.Data.MySqlClient;

namespace ConsoleApplication2
{
    class Program
    {

        private static void DisplayData(System.Data.DataTable table)
        {
            foreach (System.Data.DataRow row in table.Rows)
            {
                foreach (System.Data.DataColumn col in table.Columns)
                {
                    Console.WriteLine("{0} = {1}", col.ColumnName, row[col]);
                }
                Console.WriteLine("============================");
            }
        }

        static void Main(string[] args)
        {

            string connStr = "server=localhost;user=root;database=world;port=3306;password=******;";
            MySqlConnection conn = new MySqlConnection(connStr);

            try
            {
                Console.WriteLine("Connecting to MySQL...");
                conn.Open();

                DataTable table = conn.GetSchema("MetaDataCollections");
                //DataTable table = conn.GetSchema("UDF");
                DisplayData(table);

                conn.Close();
            }
            catch (Exception ex)
            {
                Console.WriteLine(ex.ToString());
            }
            Console.WriteLine("Done.");
        }
    }
}

To prepare a statement, create a command object and set the .CommandText property to your query.

After entering your statement, call the .Prepare method of the MySqlCommand object. After the statement is prepared, add parameters for each of the dynamic elements in the query.

After you enter your query and enter parameters, execute the statement using the .ExecuteNonQuery(), .ExecuteScalar(), or .ExecuteReader methods.

For subsequent executions, you need only modify the values of the parameters and call the execute method again, there is no need to set the .CommandText property or redefine the parameters.

Visual Basic Example

Dim conn As New MySqlConnection
Dim cmd As New MySqlCommand

conn.ConnectionString = strConnection

Try
   conn.Open()
   cmd.Connection = conn

   cmd.CommandText = "INSERT INTO myTable VALUES(NULL, @number, @text)"
   cmd.Prepare()

   cmd.Parameters.AddWithValue("@number", 1)
   cmd.Parameters.AddWithValue("@text", "One")

   For i = 1 To 1000
       cmd.Parameters("@number").Value = i
       cmd.Parameters("@text").Value = "A string value"

       cmd.ExecuteNonQuery()
     Next
Catch ex As MySqlException
    MessageBox.Show("Error " & ex.Number & " has occurred: " & ex.Message, "Error", MessageBoxButtons.OK, MessageBoxIcon.Error)
End Try

C# Example

MySql.Data.MySqlClient.MySqlConnection conn;
MySql.Data.MySqlClient.MySqlCommand cmd;

conn = new MySql.Data.MySqlClient.MySqlConnection();
cmd = new MySql.Data.MySqlClient.MySqlCommand();

conn.ConnectionString = strConnection;

try
{
    conn.Open();
    cmd.Connection = conn;

    cmd.CommandText = "INSERT INTO myTable VALUES(NULL, @number, @text)";
    cmd.Prepare();

    cmd.Parameters.AddWithValue("@number", 1);
    cmd.Parameters.AddWithValue("@text", "One");

    for (int i=1; i <= 1000; i++)
    {
        cmd.Parameters["@number"].Value = i;
        cmd.Parameters["@text"].Value = "A string value";

        cmd.ExecuteNonQuery();
    }
}
catch (MySql.Data.MySqlClient.MySqlException ex)
{
    MessageBox.Show("Error " + ex.Number + " has occurred: " + ex.Message,
        "Error", MessageBoxButtons.OK, MessageBoxIcon.Error);
}

Stored procedures in MySQL can be created using a variety of tools. First, stored procedures can be created using the mysql command-line client. Second, stored procedures can be created using MySQL Workbench. Finally, stored procedures can be created using the .ExecuteNonQuery method of the MySqlCommand object.

Unlike the command-line and GUI clients, you are not required to specify a special delimiter when creating stored procedures in Connector/Net.

To call a stored procedure using Connector/Net, you create a MySqlCommand object and pass the stored procedure name as the .CommandText property. You then set the .CommandType property to CommandType.StoredProcedure.

After the stored procedure is named, you create one MySqlCommand parameter for every parameter in the stored procedure. IN parameters are defined with the parameter name and the object containing the value, OUT parameters are defined with the parameter name and the data type that is expected to be returned. All parameters need the parameter direction defined.

After defining the parameters, you call the stored procedure by using the MySqlCommand.ExecuteNonQuery() method.

Once the stored procedure is called, the values of the output parameters can be retrieved by using the .Value property of the MySqlConnector.Parameters collection.

The following C# example code demonstrates the use of stored procedures. It assumes the database 'employees' has already been created:

using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;

using System.Data;
using MySql.Data;
using MySql.Data.MySqlClient;

namespace UsingStoredRoutines
{
    class Program
    {
        static void Main(string[] args)
        {
            MySqlConnection conn = new MySqlConnection();
            conn.ConnectionString = "server=localhost;user=root;database=employees;port=3306;password=******;";
            MySqlCommand cmd = new MySqlCommand();

            try
            {
                Console.WriteLine("Connecting to MySQL...");
                conn.Open();
                cmd.Connection = conn;
                cmd.CommandText = "DROP PROCEDURE IF EXISTS add_emp";
                cmd.ExecuteNonQuery();
                cmd.CommandText = "DROP TABLE IF EXISTS emp";
                cmd.ExecuteNonQuery();
                cmd.CommandText = "CREATE TABLE emp (empno INT UNSIGNED NOT NULL AUTO_INCREMENT PRIMARY KEY, first_name VARCHAR(20), last_name VARCHAR(20), birthdate DATE)";
                cmd.ExecuteNonQuery();

                cmd.CommandText = "CREATE PROCEDURE add_emp(" +
                                  "IN fname VARCHAR(20), IN lname VARCHAR(20), IN bday DATETIME, OUT empno INT)" +
                                  "BEGIN INSERT INTO emp(first_name, last_name, birthdate) " +
                                  "VALUES(fname, lname, DATE(bday)); SET empno = LAST_INSERT_ID(); END";

                cmd.ExecuteNonQuery();
            }
            catch (MySqlException ex)
            {
                Console.WriteLine ("Error " + ex.Number + " has occurred: " + ex.Message);
            }
            conn.Close();
            Console.WriteLine("Connection closed.");
            try
            {
                Console.WriteLine("Connecting to MySQL...");
                conn.Open();
                cmd.Connection = conn;

                cmd.CommandText = "add_emp";
                cmd.CommandType = CommandType.StoredProcedure;

                cmd.Parameters.AddWithValue("@lname", "Jones");
                cmd.Parameters["@lname"].Direction = ParameterDirection.Input;

                cmd.Parameters.AddWithValue("@fname", "Tom");
                cmd.Parameters["@fname"].Direction = ParameterDirection.Input;

                cmd.Parameters.AddWithValue("@bday", "1940-06-07");
                cmd.Parameters["@bday"].Direction = ParameterDirection.Input;

                cmd.Parameters.AddWithValue("@empno", MySqlDbType.Int32);
                cmd.Parameters["@empno"].Direction = ParameterDirection.Output;

                cmd.ExecuteNonQuery();

                Console.WriteLine("Employee number: "+cmd.Parameters["@empno"].Value);
                Console.WriteLine("Birthday: " + cmd.Parameters["@bday"].Value);
            }
            catch (MySql.Data.MySqlClient.MySqlException ex)
            {
                Console.WriteLine("Error " + ex.Number + " has occurred: " + ex.Message);
            }
            conn.Close();
            Console.WriteLine("Done.");
        }
    }
}

The following code shows the same application in Visual Basic:

Imports System
Imports System.Collections.Generic
Imports System.Linq
Imports System.Text

Imports System.Data
Imports MySql.Data
Imports MySql.Data.MySqlClient

Module Module1

    Sub Main()
        Dim conn As New MySqlConnection()
        conn.ConnectionString = "server=localhost;user=root;database=world;port=3306;password=******;"
        Dim cmd As New MySqlCommand()

        Try
            Console.WriteLine("Connecting to MySQL...")
            conn.Open()
            cmd.Connection = conn
            cmd.CommandText = "DROP PROCEDURE IF EXISTS add_emp"
            cmd.ExecuteNonQuery()
            cmd.CommandText = "DROP TABLE IF EXISTS emp"
            cmd.ExecuteNonQuery()
            cmd.CommandText = "CREATE TABLE emp (empno INT UNSIGNED NOT NULL AUTO_INCREMENT PRIMARY KEY, first_name VARCHAR(20), last_name VARCHAR(20), birthdate DATE)"
            cmd.ExecuteNonQuery()

            cmd.CommandText = "CREATE PROCEDURE add_emp(" & "IN fname VARCHAR(20), IN lname VARCHAR(20), IN bday DATETIME, OUT empno INT)" & "BEGIN INSERT INTO emp(first_name, last_name, birthdate) " & "VALUES(fname, lname, DATE(bday)); SET empno = LAST_INSERT_ID(); END"

            cmd.ExecuteNonQuery()
        Catch ex As MySqlException
            Console.WriteLine(("Error " & ex.Number & " has occurred: ") + ex.Message)
        End Try
        conn.Close()
        Console.WriteLine("Connection closed.")
        Try
            Console.WriteLine("Connecting to MySQL...")
            conn.Open()
            cmd.Connection = conn

            cmd.CommandText = "add_emp"
            cmd.CommandType = CommandType.StoredProcedure

            cmd.Parameters.AddWithValue("@lname", "Jones")
            cmd.Parameters("@lname").Direction = ParameterDirection.Input

            cmd.Parameters.AddWithValue("@fname", "Tom")
            cmd.Parameters("@fname").Direction = ParameterDirection.Input

            cmd.Parameters.AddWithValue("@bday", "1940-06-07")
            cmd.Parameters("@bday").Direction = ParameterDirection.Input

            cmd.Parameters.AddWithValue("@empno", MySqlDbType.Int32)
            cmd.Parameters("@empno").Direction = ParameterDirection.Output

            cmd.ExecuteNonQuery()

            Console.WriteLine("Employee number: " & cmd.Parameters("@empno").Value)
            Console.WriteLine("Birthday: " & cmd.Parameters("@bday").Value)
        Catch ex As MySql.Data.MySqlClient.MySqlException
            Console.WriteLine(("Error " & ex.Number & " has occurred: ") + ex.Message)
        End Try
        conn.Close()
        Console.WriteLine("Done.")

    End Sub

End Module

The first step is using MySQL with BLOB data is to configure the server. Let's start by creating a table to be accessed. In my file tables, I usually have four columns: an AUTO_INCREMENT column of appropriate size (UNSIGNED SMALLINT) to serve as a primary key to identify the file, a VARCHAR column that stores the file name, an UNSIGNED MEDIUMINT column that stores the size of the file, and a MEDIUMBLOB column that stores the file itself. For this example, I will use the following table definition:

CREATE TABLE file(
file_id SMALLINT UNSIGNED AUTO_INCREMENT NOT NULL PRIMARY KEY,
file_name VARCHAR(64) NOT NULL,
file_size MEDIUMINT UNSIGNED NOT NULL,
file MEDIUMBLOB NOT NULL);

After creating a table, you might need to modify the max_allowed_packet system variable. This variable determines how large of a packet (that is, a single row) can be sent to the MySQL server. By default, the server only accepts a maximum size of 1MB from the client application. If you intend to exceed 1MB in your file transfers, increase this number.

The max_allowed_packet option can be modified using the MySQL Workbench Server Administration screen. Adjust the Maximum permitted option in the Data / Memory size section of the Networking tab to an appropriate setting. After adjusting the value, click the Apply button and restart the server using the Startup / Shutdown screen of MySQL Workbench. You can also adjust this value directly in the my.cnf file (add a line that reads max_allowed_packet=xxM), or use the SET max_allowed_packet=xxM; syntax from within MySQL.

Try to be conservative when setting max_allowed_packet, as transfers of BLOB data can take some time to complete. Try to set a value that will be adequate for your intended use and increase the value if necessary.

To write a file to a database, we need to convert the file to a byte array, then use the byte array as a parameter to an INSERT query.

The following code opens a file using a FileStream object, reads it into a byte array, and inserts it into the file table:

Visual Basic Example

Dim conn As New MySqlConnection
Dim cmd As New MySqlCommand

Dim SQL As String

Dim FileSize As UInt32
Dim rawData() As Byte
Dim fs As FileStream

conn.ConnectionString = "server=127.0.0.1;" _
    & "uid=root;" _
    & "pwd=12345;" _
    & "database=test"

Try
    fs = New FileStream("c:\image.png", FileMode.Open, FileAccess.Read)
    FileSize = fs.Length

    rawData = New Byte(FileSize) {}
    fs.Read(rawData, 0, FileSize)
    fs.Close()

    conn.Open()

    SQL = "INSERT INTO file VALUES(NULL, @FileName, @FileSize, @File)"

    cmd.Connection = conn
    cmd.CommandText = SQL
    cmd.Parameters.AddWithValue("@FileName", strFileName)
    cmd.Parameters.AddWithValue("@FileSize", FileSize)
    cmd.Parameters.AddWithValue("@File", rawData)

    cmd.ExecuteNonQuery()

    MessageBox.Show("File Inserted into database successfully!", _
    "Success!", MessageBoxButtons.OK, MessageBoxIcon.Asterisk)

    conn.Close()
Catch ex As Exception
    MessageBox.Show("There was an error: " & ex.Message, "Error", _
        MessageBoxButtons.OK, MessageBoxIcon.Error)
End Try

C# Example

MySql.Data.MySqlClient.MySqlConnection conn;
MySql.Data.MySqlClient.MySqlCommand cmd;

conn = new MySql.Data.MySqlClient.MySqlConnection();
cmd = new MySql.Data.MySqlClient.MySqlCommand();

string SQL;
UInt32 FileSize;
byte[] rawData;
FileStream fs;

conn.ConnectionString = "server=127.0.0.1;uid=root;" +
    "pwd=12345;database=test;";

try
{
    fs = new FileStream(@"c:\image.png", FileMode.Open, FileAccess.Read);
    FileSize = fs.Length;

    rawData = new byte[FileSize];
    fs.Read(rawData, 0, FileSize);
    fs.Close();

    conn.Open();

    SQL = "INSERT INTO file VALUES(NULL, @FileName, @FileSize, @File)";

    cmd.Connection = conn;
    cmd.CommandText = SQL;
    cmd.Parameters.AddWithValue("@FileName", strFileName);
    cmd.Parameters.AddWithValue("@FileSize", FileSize);
    cmd.Parameters.AddWithValue("@File", rawData);

    cmd.ExecuteNonQuery();

    MessageBox.Show("File Inserted into database successfully!",
        "Success!", MessageBoxButtons.OK, MessageBoxIcon.Asterisk);

    conn.Close();
}
catch (MySql.Data.MySqlClient.MySqlException ex)
{
    MessageBox.Show("Error " + ex.Number + " has occurred: " + ex.Message,
        "Error", MessageBoxButtons.OK, MessageBoxIcon.Error);
}

The Read method of the FileStream object is used to load the file into a byte array which is sized according to the Length property of the FileStream object.

After assigning the byte array as a parameter of the MySqlCommand object, the ExecuteNonQuery method is called and the BLOB is inserted into the file table.

Once a file is loaded into the file table, we can use the MySqlDataReader class to retrieve it.

The following code retrieves a row from the file table, then loads the data into a FileStream object to be written to disk:

Visual Basic Example

Dim conn As New MySqlConnection
Dim cmd As New MySqlCommand
Dim myData As MySqlDataReader
Dim SQL As String
Dim rawData() As Byte
Dim FileSize As UInt32
Dim fs As FileStream

conn.ConnectionString = "server=127.0.0.1;" _
    & "uid=root;" _
    & "pwd=12345;" _
    & "database=test"

SQL = "SELECT file_name, file_size, file FROM file"

Try
    conn.Open()

    cmd.Connection = conn
    cmd.CommandText = SQL

    myData = cmd.ExecuteReader

    If Not myData.HasRows Then Throw New Exception("There are no BLOBs to save")

    myData.Read()

    FileSize = myData.GetUInt32(myData.GetOrdinal("file_size"))
    rawData = New Byte(FileSize) {}

    myData.GetBytes(myData.GetOrdinal("file"), 0, rawData, 0, FileSize)

    fs = New FileStream("C:\newfile.png", FileMode.OpenOrCreate, FileAccess.Write)
    fs.Write(rawData, 0, FileSize)
    fs.Close()

    MessageBox.Show("File successfully written to disk!", "Success!", MessageBoxButtons.OK, MessageBoxIcon.Asterisk)

    myData.Close()
    conn.Close()
Catch ex As Exception
    MessageBox.Show("There was an error: " & ex.Message, "Error", MessageBoxButtons.OK, MessageBoxIcon.Error)
End Try

C# Example

MySql.Data.MySqlClient.MySqlConnection conn;
MySql.Data.MySqlClient.MySqlCommand cmd;
MySql.Data.MySqlClient.MySqlDataReader myData;

conn = new MySql.Data.MySqlClient.MySqlConnection();
cmd = new MySql.Data.MySqlClient.MySqlCommand();

string SQL;
UInt32 FileSize;
byte[] rawData;
FileStream fs;

conn.ConnectionString = "server=127.0.0.1;uid=root;" +
    "pwd=12345;database=test;";

SQL = "SELECT file_name, file_size, file FROM file";

try
{
    conn.Open();

    cmd.Connection = conn;
    cmd.CommandText = SQL;

    myData = cmd.ExecuteReader();

    if (! myData.HasRows)
        throw new Exception("There are no BLOBs to save");

    myData.Read();

    FileSize = myData.GetUInt32(myData.GetOrdinal("file_size"));
    rawData = new byte[FileSize];

    myData.GetBytes(myData.GetOrdinal("file"), 0, rawData, 0, (int)FileSize);

    fs = new FileStream(@"C:\newfile.png", FileMode.OpenOrCreate, FileAccess.Write);
    fs.Write(rawData, 0, (int)FileSize);
    fs.Close();

    MessageBox.Show("File successfully written to disk!",
        "Success!", MessageBoxButtons.OK, MessageBoxIcon.Asterisk);

    myData.Close();
    conn.Close();
}
catch (MySql.Data.MySqlClient.MySqlException ex)
{
    MessageBox.Show("Error " + ex.Number + " has occurred: " + ex.Message,
        "Error", MessageBoxButtons.OK, MessageBoxIcon.Error);
}

After connecting, the contents of the file table are loaded into a MySqlDataReader object. The GetBytes method of the MySqlDataReader is used to load the BLOB into a byte array, which is then written to disk using a FileStream object.

The GetOrdinal method of the MySqlDataReader can be used to determine the integer index of a named column. Use of the GetOrdinal method prevents errors if the column order of the SELECT query is changed.

Connector/Net 6.5 and higher support the fractional seconds feature introduced in MySQL 5.6.4. Fractional seconds could always be specified in a date literal or passed back and forth as parameters and return values, but the fractional part was always stripped off when stored in a table column. In MySQL 5.6.4 and higher, the fractional part is now preserved in data stored and retrieved through SQL. For fractional second handling in MySQL 5.6.4 and higher, see Fractional Seconds in Time Values. For the behavior of fractional seconds prior to MySQL 5.6.4, see Fractional Seconds in Time Values.

To use the more precise date and time types, specify a value from 1 to 6 when creating the table column, for example TIME(3) or DATETIME(6), representing the number of digits of precision after the decimal point. Specifying a precision of 0 leaves the fractional part out entirely. In your C# or Visual Basic code, refer to the Millisecond member to retrieve the fractional second value from the MySqlDateTime object returned by the GetMySqlDateTime function. The DateTime object returned by the GetDateTime function also contains the fractional value, but only the first 3 digits.

For related code examples, see the following blog post: https://blogs.oracle.com/MySqlOnWindows/entry/milliseconds_value_support_on_datetime

The differences in date handling can cause problems for developers who use invalid dates. Invalid MySQL dates cannot be loaded into native .NET DateTime objects, including NULL dates.

Because of this issue, .NET DataSet objects cannot be populated by the Fill method of the MySqlDataAdapter class as invalid dates will cause a System.ArgumentOutOfRangeException exception to occur.

The best solution to the date problem is to restrict users from entering invalid dates. This can be done on either the client or the server side.

Restricting invalid dates on the client side is as simple as always using the .NET DateTime class to handle dates. The DateTime class will only allow valid dates, ensuring that the values in your database are also valid. The disadvantage of this is that it is not useful in a mixed environment where .NET and non .NET code are used to manipulate the database, as each application must perform its own date validation.

Users of MySQL 5.0.2 and higher can use the new traditional SQL mode to restrict invalid date values. For information on using the traditional SQL mode, see Server SQL Modes.

Although it is strongly recommended that you avoid the use of invalid dates within your .NET application, it is possible to use invalid dates by means of the MySqlDateTime data type.

The MySqlDateTime data type supports the same date values that are supported by the MySQL server. The default behavior of Connector/Net is to return a .NET DateTime object for valid date values, and return an error for invalid dates. This default can be modified to cause Connector/Net to return MySqlDateTime objects for invalid dates.

To instruct Connector/Net to return a MySqlDateTime object for invalid dates, add the following line to your connection string:

Allow Zero Datetime=True

The MySqlDateTime class can still be problematic. The following are some known issues:

Because of the known issues, the best recommendation is still to use only valid dates in your application.

The .NET DateTime data type cannot handle NULL values. As such, when assigning values from a query to a DateTime variable, you must first check whether the value is in fact NULL.

When using a MySqlDataReader, use the .IsDBNull method to check whether a value is NULL before making the assignment:

Visual Basic Example

If Not myReader.IsDBNull(myReader.GetOrdinal("mytime")) Then
    myTime = myReader.GetDateTime(myReader.GetOrdinal("mytime"))
Else
    myTime = DateTime.MinValue
End If

C# Example

if (! myReader.IsDBNull(myReader.GetOrdinal("mytime")))
    myTime = myReader.GetDateTime(myReader.GetOrdinal("mytime"));
else
    myTime = DateTime.MinValue;

NULL values will work in a data set and can be bound to form controls without special handling.

This section describes how to set up your application to view MySQL trace information.

The first thing you need to do is create a suitable app.config file for your application. An example is shown in the following code:

<?xml version="1.0" encoding="utf-8" ?>
<configuration>
  <system.diagnostics>
    <sources>
      <source name="mysql" switchName="SourceSwitch"
        switchType="System.Diagnostics.SourceSwitch" >
        <listeners>
          <add name="console" />
          <remove name ="Default" />
        </listeners>
      </source>
    </sources>
    <switches>
      <!-- You can set the level at which tracing is to occur -->
      <add name="SourceSwitch" value="Verbose" />
      <!-- You can turn tracing off -->
      <!--add name="SourceSwitch" value="Off" -->
    </switches>
    <sharedListeners>
      <add name="console"
        type="System.Diagnostics.ConsoleTraceListener"
        initializeData="false"/>
    </sharedListeners>
  </system.diagnostics>
</configuration>

This ensures a suitable trace source is created, along with a switch. The switch level in this case is set to Verbose to display the maximum amount of information.

In the application the only other step required is to add logging=true to the connection string. An example application could be:

using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
using System.Diagnostics;
using MySql.Data;
using MySql.Data.MySqlClient;
using MySql.Web;


namespace ConsoleApplication1
{
    class Program
    {
        static void Main(string[] args)
        {

            string connStr = "server=localhost;user=root;database=world;port=3306;password=******;logging=true;";
            MySqlConnection conn = new MySqlConnection(connStr);
            try
            {
                Console.WriteLine("Connecting to MySQL...");
                conn.Open();

                string sql = "SELECT Name, HeadOfState FROM Country WHERE Continent='Oceania'";
                MySqlCommand cmd = new MySqlCommand(sql, conn);
                MySqlDataReader rdr = cmd.ExecuteReader();

                while (rdr.Read())
                {
                    Console.WriteLine(rdr[0] + " -- " + rdr[1]);
                }

                rdr.Close();

                conn.Close();
            }
            catch (Exception ex)
            {
                Console.WriteLine(ex.ToString());
            }
            Console.WriteLine("Done.");
        }
    }
}

This simple application will then generate the following output:

Connecting to MySQL...
mysql Information: 1 : 1: Connection Opened: connection string = 'server=localhost;User Id=root;database=world;port=3306
;password=******;logging=True'
mysql Information: 3 : 1: Query Opened: SHOW VARIABLES
mysql Information: 4 : 1: Resultset Opened: field(s) = 2, affected rows = -1, inserted id = -1
mysql Information: 5 : 1: Resultset Closed. Total rows=272, skipped rows=0, size (bytes)=7058
mysql Information: 6 : 1: Query Closed
mysql Information: 3 : 1: Query Opened: SHOW COLLATION
mysql Information: 4 : 1: Resultset Opened: field(s) = 6, affected rows = -1, inserted id = -1
mysql Information: 5 : 1: Resultset Closed. Total rows=127, skipped rows=0, size (bytes)=4102
mysql Information: 6 : 1: Query Closed
mysql Information: 3 : 1: Query Opened: SET character_set_results=NULL
mysql Information: 4 : 1: Resultset Opened: field(s) = 0, affected rows = 0, inserted id = 0
mysql Information: 5 : 1: Resultset Closed. Total rows=0, skipped rows=0, size (bytes)=0
mysql Information: 6 : 1: Query Closed
mysql Information: 10 : 1: Set Database: world
mysql Information: 3 : 1: Query Opened: SELECT Name, HeadOfState FROM Country WHERE Continent='Oceania'
mysql Information: 4 : 1: Resultset Opened: field(s) = 2, affected rows = -1, inserted id = -1
American Samoa -- George W. Bush
Australia -- Elisabeth II
...
Wallis and Futuna -- Jacques Chirac
Vanuatu -- John Bani
United States Minor Outlying Islands -- George W. Bush
mysql Information: 5 : 1: Resultset Closed. Total rows=28, skipped rows=0, size (bytes)=788
mysql Information: 6 : 1: Query Closed
Done.
mysql Information: 2 : 1: Connection Closed

The first number displayed in the trace message corresponds to the MySQL event type:

The second number displayed in the trace message is the connection count.

Although this example uses the ConsoleTraceListener, any of the other standard listeners could have been used. Another possibility is to create a custom listener that uses the information passed using the TraceEvent method. For example, a custom trace listener could be created to perform active monitoring of the MySQL event messages, rather than simply writing these to an output device.

It is also possible to add listeners to the MySQL Trace Source at runtime. This can be done with the following code:

MySqlTrace.Listeners.Add(new ConsoleTraceListener());

MySQL Connector/Net 6.3.2 introduced the ability to switch tracing on and off at runtime. This can be achieved using the calls MySqlTrace.EnableQueryAnalyzer(string host, int postInterval) and MySqlTrace.DisableQueryAnalyzer(). The parameter host is the URL of the MySQL Enterprise Monitor server to monitor. The parameter postInterval is how often to post the data to MySQL Enterprise Monitor, in seconds.

To build custom listeners that work with the MySQL Connector/Net Trace Source, it is necessary to understand the key methods used, and the event data formats used.

The main method involved in passing trace messages is the TraceSource.TraceEvent method. This has the prototype:

public void TraceEvent(
    TraceEventType eventType,
    int id,
    string format,
    params Object[] args
)

This trace source method will process the list of attached listeners and call the listener's TraceListener.TraceEvent method. The prototype for the TraceListener.TraceEvent method is as follows:

public virtual void TraceEvent(
    TraceEventCache eventCache,
    string source,
    TraceEventType eventType,
    int id,
    string format,
    params Object[] args
)

The first three parameters are used in the standard as defined by Microsoft. The last three parameters contain MySQL-specific trace information. Each of these parameters is now discussed in more detail.

int id

This is a MySQL-specific identifier. It identifies the MySQL event type that has occurred, resulting in a trace message being generated. This value is defined by the MySqlTraceEventType public enum contained in the MySQL Connector/Net code:

public enum MySqlTraceEventType : int
{
    ConnectionOpened = 1,
    ConnectionClosed,
    QueryOpened,
    ResultOpened,
    ResultClosed,
    QueryClosed,
    StatementPrepared,
    StatementExecuted,
    StatementClosed,
    NonQuery,
    UsageAdvisorWarning,
    Warning,
    Error
}

The MySQL event type also determines the contents passed using the parameter params Object[] args. The nature of the args parameters are described in further detail in the following material.

string format

This is the format string that contains zero or more format items, which correspond to objects in the args array. This would be used by a listener such as ConsoleTraceListener to write a message to the output device.

params Object[] args

This is a list of objects that depends on the MySQL event type, id. However, the first parameter passed using this list is always the driver id. The driver id is a unique number that is incremented each time the connector is opened. This enables groups of queries on the same connection to be identified. The parameters that follow driver id depend on the MySQL event id, and are as follows:

This information will allow you to create custom trace listeners that can actively monitor the MySQL-specific events.

When creating a report in Crystal Reports there are two options for accessing the MySQL data while designing your report.

The first option is to use Connector/ODBC as an ADO data source when designing your report. You will be able to browse your database and choose tables and fields using drag and drop to build your report. The disadvantage of this approach is that additional work must be performed within your application to produce a data set that matches the one expected by your report.

The second option is to create a data set in VB.NET and save it as XML. This XML file can then be used to design a report. This works quite well when displaying the report in your application, but is less versatile at design time because you must choose all relevant columns when creating the data set. If you forget a column you must re-create the data set before the column can be added to the report.

The following code can be used to create a data set from a query and write it to disk:

Visual Basic Example

Dim myData As New DataSet
Dim conn As New MySqlConnection
Dim cmd As New MySqlCommand
Dim myAdapter As New MySqlDataAdapter

conn.ConnectionString = "server=127.0.0.1;" _
    & "uid=root;" _
    & "pwd=12345;" _
    & "database=world"

Try
    conn.Open()
    cmd.CommandText = "SELECT city.name AS cityName, city.population AS CityPopulation, " _
        & "country.name, country.population, country.continent " _
        & "FROM country, city ORDER BY country.continent, country.name"
    cmd.Connection = conn

    myAdapter.SelectCommand = cmd
    myAdapter.Fill(myData)

    myData.WriteXml("C:\dataset.xml", XmlWriteMode.WriteSchema)
Catch ex As Exception
    MessageBox.Show(ex.Message, "Report could not be created", MessageBoxButtons.OK, MessageBoxIcon.Error)
End Try

C# Example

DataSet myData = new DataSet();
MySql.Data.MySqlClient.MySqlConnection conn;
MySql.Data.MySqlClient.MySqlCommand cmd;
MySql.Data.MySqlClient.MySqlDataAdapter myAdapter;

conn = new MySql.Data.MySqlClient.MySqlConnection();
cmd = new MySql.Data.MySqlClient.MySqlCommand();
myAdapter = new MySql.Data.MySqlClient.MySqlDataAdapter();

conn.ConnectionString = "server=127.0.0.1;uid=root;" +
  "pwd=12345;database=test;";

try
{
  cmd.CommandText = "SELECT city.name AS cityName, city.population AS CityPopulation, " +
  "country.name, country.population, country.continent " +
  "FROM country, city ORDER BY country.continent, country.name";
  cmd.Connection = conn;

  myAdapter.SelectCommand = cmd;
  myAdapter.Fill(myData);

  myData.WriteXml(@"C:\dataset.xml", XmlWriteMode.WriteSchema);
}
catch (MySql.Data.MySqlClient.MySqlException ex)
{
  MessageBox.Show(ex.Message, "Report could not be created",
  MessageBoxButtons.OK, MessageBoxIcon.Error);
}

The resulting XML file can be used as an ADO.NET XML datasource when designing your report.

If you choose to design your reports using Connector/ODBC, it can be downloaded from dev.mysql.com.

For most purposes, the Standard Report wizard helps with the initial creation of a report. To start the wizard, open Crystal Reports and choose the New > Standard Report option from the File menu.

The wizard first prompts you for a data source. If you use Connector/ODBC as your data source, use the OLEDB provider for ODBC option from the OLE DB (ADO) tree instead of the ODBC (RDO) tree when choosing a data source. If using a saved data set, choose the ADO.NET (XML) option and browse to your saved data set.

The remainder of the report creation process is done automatically by the wizard.

After the report is created, choose the Report Options... entry of the File menu. Un-check the Save Data With Report option. This prevents saved data from interfering with the loading of data within our application.

To display a report we first populate a data set with the data needed for the report, then load the report and bind it to the data set. Finally we pass the report to the crViewer control for display to the user.

The following references are needed in a project that displays a report:

The following code assumes that you created your report using a data set saved using the code shown in Section 5.19.1, “Creating a Data Source”, and have a crViewer control on your form named myViewer.

Visual Basic Example

Imports CrystalDecisions.CrystalReports.Engine
Imports System.Data
Imports MySql.Data.MySqlClient

Dim myReport As New ReportDocument
Dim myData As New DataSet
Dim conn As New MySqlConnection
Dim cmd As New MySqlCommand
Dim myAdapter As New MySqlDataAdapter

conn.ConnectionString = _
    "server=127.0.0.1;" _
    & "uid=root;" _
    & "pwd=12345;" _
    & "database=test"

Try
    conn.Open()

    cmd.CommandText = "SELECT city.name AS cityName, city.population AS CityPopulation, " _
        & "country.name, country.population, country.continent " _
        & "FROM country, city ORDER BY country.continent, country.name"
    cmd.Connection = conn

    myAdapter.SelectCommand = cmd
    myAdapter.Fill(myData)

    myReport.Load(".\world_report.rpt")
    myReport.SetDataSource(myData)
    myViewer.ReportSource = myReport
Catch ex As Exception
    MessageBox.Show(ex.Message, "Report could not be created", MessageBoxButtons.OK, MessageBoxIcon.Error)
End Try

C# Example

using CrystalDecisions.CrystalReports.Engine;
using System.Data;
using MySql.Data.MySqlClient;

ReportDocument myReport = new ReportDocument();
DataSet myData = new DataSet();
MySql.Data.MySqlClient.MySqlConnection conn;
MySql.Data.MySqlClient.MySqlCommand cmd;
MySql.Data.MySqlClient.MySqlDataAdapter myAdapter;

conn = new MySql.Data.MySqlClient.MySqlConnection();
cmd = new MySql.Data.MySqlClient.MySqlCommand();
myAdapter = new MySql.Data.MySqlClient.MySqlDataAdapter();

conn.ConnectionString = "server=127.0.0.1;uid=root;" +
    "pwd=12345;database=test;";

try
{
    cmd.CommandText = "SELECT city.name AS cityName, city.population AS CityPopulation, " +
        "country.name, country.population, country.continent " +
        "FROM country, city ORDER BY country.continent, country.name";
    cmd.Connection = conn;

    myAdapter.SelectCommand = cmd;
    myAdapter.Fill(myData);

    myReport.Load(@".\world_report.rpt");
    myReport.SetDataSource(myData);
    myViewer.ReportSource = myReport;
}
catch (MySql.Data.MySqlClient.MySqlException ex)
{
    MessageBox.Show(ex.Message, "Report could not be created",
        MessageBoxButtons.OK, MessageBoxIcon.Error);
}

A new data set it generated using the same query used to generate the previously saved data set. Once the data set is filled, a ReportDocument is used to load the report file and bind it to the data set. The ReportDocument is the passed as the ReportSource of the crViewer.

This same approach is taken when a report is created from a single table using Connector/ODBC. The data set replaces the table used in the report and the report is displayed properly.

When a report is created from multiple tables using Connector/ODBC, a data set with multiple tables must be created in our application. This enables each table in the report data source to be replaced with a report in the data set.

We populate a data set with multiple tables by providing multiple SELECT statements in our MySqlCommand object. These SELECT statements are based on the SQL query shown in Crystal Reports in the Database menu's Show SQL Query option. Assume the following query:

SELECT `country`.`Name`, `country`.`Continent`, `country`.`Population`, `city`.`Name`, `city`.`Population`
FROM `world`.`country` `country` LEFT OUTER JOIN `world`.`city` `city` ON `country`.`Code`=`city`.`CountryCode`
ORDER BY `country`.`Continent`, `country`.`Name`, `city`.`Name`

This query is converted to two SELECT queries and displayed with the following code:

Visual Basic Example

Imports CrystalDecisions.CrystalReports.Engine
Imports System.Data
Imports MySql.Data.MySqlClient

Dim myReport As New ReportDocument
Dim myData As New DataSet
Dim conn As New MySqlConnection
Dim cmd As New MySqlCommand
Dim myAdapter As New MySqlDataAdapter

conn.ConnectionString = "server=127.0.0.1;" _
    & "uid=root;" _
    & "pwd=12345;" _
    & "database=world"

Try
    conn.Open()
    cmd.CommandText = "SELECT name, population, countrycode FROM city ORDER BY countrycode, name; " _
        & "SELECT name, population, code, continent FROM country ORDER BY continent, name"
    cmd.Connection = conn

    myAdapter.SelectCommand = cmd
    myAdapter.Fill(myData)

    myReport.Load(".\world_report.rpt")
    myReport.Database.Tables(0).SetDataSource(myData.Tables(0))
    myReport.Database.Tables(1).SetDataSource(myData.Tables(1))
    myViewer.ReportSource = myReport
Catch ex As Exception
    MessageBox.Show(ex.Message, "Report could not be created", MessageBoxButtons.OK, MessageBoxIcon.Error)
End Try

C# Example

using CrystalDecisions.CrystalReports.Engine;
using System.Data;
using MySql.Data.MySqlClient;

ReportDocument myReport = new ReportDocument();
DataSet myData = new DataSet();
MySql.Data.MySqlClient.MySqlConnection conn;
MySql.Data.MySqlClient.MySqlCommand cmd;
MySql.Data.MySqlClient.MySqlDataAdapter myAdapter;

conn = new MySql.Data.MySqlClient.MySqlConnection();
cmd = new MySql.Data.MySqlClient.MySqlCommand();
myAdapter = new MySql.Data.MySqlClient.MySqlDataAdapter();

conn.ConnectionString = "server=127.0.0.1;uid=root;" +
    "pwd=12345;database=test;";

try
{
    cmd.CommandText = "SELECT name, population, countrycode FROM city ORDER " +
        "BY countrycode, name; SELECT name, population, code, continent FROM " +
        "country ORDER BY continent, name";
    cmd.Connection = conn;

    myAdapter.SelectCommand = cmd;
    myAdapter.Fill(myData);

    myReport.Load(@".\world_report.rpt");
    myReport.Database.Tables(0).SetDataSource(myData.Tables(0));
    myReport.Database.Tables(1).SetDataSource(myData.Tables(1));
    myViewer.ReportSource = myReport;
}
catch (MySql.Data.MySqlClient.MySqlException ex)
{
    MessageBox.Show(ex.Message, "Report could not be created",
        MessageBoxButtons.OK, MessageBoxIcon.Error);
}

It is important to order the SELECT queries in alphabetic order, as this is the order the report will expect its source tables to be in. One SetDataSource statement is needed for each table in the report.

This approach can cause performance problems because Crystal Reports must bind the tables together on the client-side, which will be slower than using a pre-saved data set.

The partial trust support for Connector/Net has improved rapidly throughout the 6.5.x and 6.6.x versions. The latest enhancements do require some configuration changes in existing deployments. Here is a summary of the changes for each version.

6.6.4 and Above: Library Can Be Inside or Outside GAC

Now you can install the MySql.Data.dll library in the Global Assembly Cache (GAC) as explained in Section 5.21.2, “Configuring Partial Trust with Connector/Net Library Installed in GAC”, or in a bin or lib folder inside the project or solution as explained in Section 5.21.3, “Configuring Partial Trust with Connector/Net Library Not Installed in GAC”. If the library is not in the GAC, the only protocol supported is TCP/IP.

6.5.1 and Above: Partial Trust Requires Library in the GAC

Connector/Net 6.5 fully enables our provider to run in a partial trust environment when the library is installed in the Global Assembly Cache (GAC). The new MySqlClientPermission class, derived from the .NET DBDataPermission class, helps to simplify the permission setup.

5.0.8 / 5.1.3 and Above: Partial Trust Requires Socket Permissions

Starting with these versions, Connector/Net can be used under partial trust hosting that has been modified to allow the use of sockets for communication. By default, partial trust does not include SocketPermission. Connector/Net uses sockets to talk with the MySQL server, so the hosting provider must create a new trust level that is an exact clone of partial trust but that has the following permissions added:

Prior to 5.0.8 / 5.1.3: Partial Trust Not Supported

Connector/Net versions prior to 5.0.8 and 5.1.3 were not compatible with partial trust hosting.

If the library is installed in the GAC, you must include the connection option includesecurityasserts=true in your connection string. This is a new requirement as of Connector/Net 6.6.4.

The following list shows steps and code fragments needed to run a Connector/Net application in a partial trust environment. For illustration purposes, we use the Pipe Connections protocol in this example.

When deploying a web application to a Shared Hosted environment, where this environment is configured to run all their .NET applications under a partial or medium trust level, you might not be able to install the Connector/Net library in the GAC. Instead, you put a reference to the library in the bin or lib folder inside the project or solution. In this case, you configure the security in a different way than when the library is in the GAC.

Connector/Net is commonly used by applications that run in Windows environments where the default communication for the protocol is used via sockets or by TCP/IP. For this protocol to operate is necessary have the required socket permissions in the web configuration file as follows: