Muokkaa

Jaa

Understanding the Script Component Object Model

  • Artikkeli

Applies to: SQL Server SSIS Integration Runtime in Azure Data Factory

As discussed in Coding and Debugging the Script Component, the Script component project contains three project items:

  1. The ScriptMain item, which contains the ScriptMain class in which you write your code. The ScriptMain class inherits from the UserComponent class.

  2. The ComponentWrapper item, which contains the UserComponent class, an instance of ScriptComponent that contains the methods and properties that you will use to process data and to interact with the package. The ComponentWrapper item also contains Connections and Variables collection classes.

  3. The BufferWrapper item, which contains classes that inherits from ScriptBuffer for each input and output, and typed properties for each column.

As you write your code in the ScriptMain item, you will use the objects, methods, and properties discussed in this topic. Each component will not use all the methods listed here; however, when used, they are used in the sequence shown.

The ScriptComponent base class does not contain any implementation code for the methods discussed in this topic. Therefore it is unnecessary, but harmless, to add a call to the base class implementation to your own implementation of the method.

For information about how to use the methods and properties of these classes in a particular type of Script component, see the section Additional Script Component Examples. The example topics also contain complete code samples.

AcquireConnections Method

Sources and destinations generally must connect to an external data source. Override the AcquireConnections method of the ScriptComponent base class to retrieve the connection or the connection information from the appropriate connection manager.

The following example returns a System.Data.SqlClient.SqlConnection from an ADO.NET connection manager.

Dim connMgr As IDTSConnectionManager100  
Dim sqlConn As SqlConnection  
  
Public Overrides Sub AcquireConnections(ByVal Transaction As Object)  
  
    connMgr = Me.Connections.MyADONETConnection  
    sqlConn = CType(connMgr.AcquireConnection(Nothing), SqlConnection)  
  
End Sub

The following example returns a complete path and file name from a Flat File Connection Manager, and then opens the file by using a System.IO.StreamReader.

Private textReader As StreamReader  
Public Overrides Sub AcquireConnections(ByVal Transaction As Object)  
  
    Dim connMgr As IDTSConnectionManager100 = _  
        Me.Connections.MyFlatFileSrcConnectionManager  
    Dim exportedAddressFile As String = _  
        CType(connMgr.AcquireConnection(Nothing), String)  
    textReader = New StreamReader(exportedAddressFile)  
  
End Sub

PreExecute Method

Override the PreExecute method of the ScriptComponent base class whenever you have processing that you must perform one time only before you start processing rows of data. For example, in a destination, you may want to configure the parameterized command that the destination will use to insert each row of data into the data source.

    Dim sqlConn As SqlConnection  
    Dim sqlCmd As SqlCommand  
    Dim sqlParam As SqlParameter  
...  
    Public Overrides Sub PreExecute()  
  
        sqlCmd = New SqlCommand("INSERT INTO Person.Address2(AddressID, City) " & _  
            "VALUES(@addressid, @city)", sqlConn)  
        sqlParam = New SqlParameter("@addressid", SqlDbType.Int)  
        sqlCmd.Parameters.Add(sqlParam)  
        sqlParam = New SqlParameter("@city", SqlDbType.NVarChar, 30)  
        sqlCmd.Parameters.Add(sqlParam)  
  
    End Sub  

SqlConnection sqlConn;   
SqlCommand sqlCmd;   
SqlParameter sqlParam;   
  
public override void PreExecute()   
{   
  
    sqlCmd = new SqlCommand("INSERT INTO Person.Address2(AddressID, City) " + "VALUES(@addressid, @city)", sqlConn);   
    sqlParam = new SqlParameter("@addressid", SqlDbType.Int);   
    sqlCmd.Parameters.Add(sqlParam);   
    sqlParam = new SqlParameter("@city", SqlDbType.NVarChar, 30);   
    sqlCmd.Parameters.Add(sqlParam);   
  
}

Processing Inputs and Outputs

Processing Inputs

Script components that are configured as transformations or destinations have one input.

What the BufferWrapper Project Item Provides

For each input that you have configured, the BufferWrapper project item contains a class that derives from ScriptBuffer and has the same name as the input. Each input buffer class contains the following properties, functions, and methods:

  • Named, typed accessor properties for each selected input column. These properties are read-only or read/write depending on the Usage Type specified for the column on the Input Columns page of the Script Transformation Editor.

  • A <column>_IsNull property for each selected input column. This property is also read-only or read/write depending on the Usage Type specified for the column.

  • A DirectRowTo<outputbuffer> method for each configured output. You will use these methods when filtering rows to one of several outputs in the same ExclusionGroup.

  • A NextRow function to get the next input row, and an EndOfRowset function to determine whether the last buffer of data has been processed. You typically do not need these functions when you use the input processing methods implemented in the UserComponent base class. The next section provides more information about the UserComponent base class.

What the ComponentWrapper Project Item Provides

The ComponentWrapper project item contains a class named UserComponent that derives from ScriptComponent. The ScriptMain class in which you write your custom code derives in turn from UserComponent. The UserComponent class contains the following methods:

  • An overridden implementation of the ProcessInput method. This is the method that the data flow engine calls next at run time after the PreExecute method, and it may be called multiple times. ProcessInput hands off processing to the <inputbuffer>_ProcessInput method. Then the ProcessInput method checks for the end of the input buffer and, if the end of the buffer has been reached, calls the overridable FinishOutputs method and the private MarkOutputsAsFinished method. The MarkOutputsAsFinished method then calls SetEndOfRowset on the last output buffer.

  • An overridable implementation of the <inputbuffer>_ProcessInput method. This default implementation simply loops through each input row and calls <inputbuffer>_ProcessInputRow.

  • An overridable implementation of the <inputbuffer>_ProcessInputRow method. The default implementation is empty. This is the method that you will normally override to write your custom data processing code.

What Your Custom Code Should Do

You can use the following methods to process input in the ScriptMain class:

  • Override <inputbuffer>_ProcessInputRow to process the data in each input row as it passes through.

  • Override <inputbuffer>_ProcessInput only if you have to do something additional while looping through input rows. (For example, you have to test for EndOfRowset to take some other action after all rows have been processed.) Call <inputbuffer>_ProcessInputRow to perform the row processing.

  • Override FinishOutputs if you have to do something to the outputs before they are closed.

The ProcessInput method ensures that these methods are called at the appropriate times.

Processing Outputs

Script components configured as sources or transformations have one or more outputs.

What the BufferWrapper Project Item Provides

For each output that you have configured, the BufferWrapper project item contains a class that derives from ScriptBuffer and has the same name as the output. Each input buffer class contains the following properties and methods:

  • Named, typed, write-only accessor properties for each output column.

  • A write-only <column>_IsNull property for each selected output column that you can use to set the column value to null.

  • An AddRow method to add an empty new row to the output buffer.

  • A SetEndOfRowset method to let the data flow engine know that no more buffers of data are expected. There is also an EndOfRowset function to determine whether the current buffer is the last buffer of data. You generally do not need these functions when you use the input processing methods implemented in the UserComponent base class.

What the ComponentWrapper Project Item Provides

The ComponentWrapper project item contains a class named UserComponent that derives from ScriptComponent. The ScriptMain class in which you write your custom code derives in turn from UserComponent. The UserComponent class contains the following methods:

  • An overridden implementation of the PrimeOutput method. The data flow engine calls this method before ProcessInput at run time, and it is only called one time. PrimeOutput hands off processing to the CreateNewOutputRows method. Then, if the component is a source (that is, the component has no inputs), PrimeOutput calls the overridable FinishOutputs method and the private MarkOutputsAsFinished method. The MarkOutputsAsFinished method calls SetEndOfRowset on the last output buffer.

  • An overridable implementation of the CreateNewOutputRows method. The default implementation is empty. This is the method that you will normally override to write your custom data processing code.

What Your Custom Code Should Do

You can use the following methods to process outputs in the ScriptMain class:

  • Override CreateNewOutputRows only when you can add and populate output rows before processing input rows. For example, you can use CreateNewOutputRows in a source, but in a transformation with asynchronous outputs, you should call AddRow during or after the processing of input data.

  • Override FinishOutputs if you have to do something to the outputs before they are closed.

The PrimeOutput method ensures that these methods are called at the appropriate times.

PostExecute Method

Override the PostExecute method of the ScriptComponent base class whenever you have processing that you must perform one time only after you have processed the rows of data. For example, in a source, you may want to close the System.Data.SqlClient.SqlDataReader that you have used to load data into the data flow.

Important

The collection of ReadWriteVariables is available only in the PostExecute method. Therefore you cannot directly increment the value of a package variable as you process each row of data. Instead, increment the value of a local variable, and set the value of the package variable to the value of the local variable in the PostExecute method after all data has been processed.

ReleaseConnections Method

Sources and destinations typically must connect to an external data source. Override the ReleaseConnections method of the ScriptComponent base class to close and release the connection that you have opened previously in the AcquireConnections method.

    Dim connMgr As IDTSConnectionManager100  
...  
    Public Overrides Sub ReleaseConnections()  
  
        connMgr.ReleaseConnection(sqlConn)  
  
    End Sub  

IDTSConnectionManager100 connMgr;  
  
public override void ReleaseConnections()  
{  
  
    connMgr.ReleaseConnection(sqlConn);  
  
}

See Also

Configuring the Script Component in the Script Component Editor
Coding and Debugging the Script Component