Stream Not writing last byte?

Stream Not writing last byte?

• mike22

  In your GZipCompress function you are returning the results before you have flushed and closed your stream.  In order to guarantee that the stream contents are properly written you need to flush or close the stream before using it. 
  
  Note that you don't need to flush the stream if you close it because Close will automatically flush the stream first.  You also don't need to call Dispose if you have already called Close.  They both invoke the same method.  For classes that expose both you should normally just call Close as it makes more logical sense.
  
  Note also that your code is not exception safe.  If an exception occurs while operating on the stream you will leak memory and potentially lock a resource.  Here is my recommended update to your function.  You should modify your second function as well.
  
  

  Public Function GZipCompress(ByVal compressData As String)

  Dim sr As IO.MemoryStream
  Dim compressedzipStream As GZipStream
  Dim results() As Byte
  

  Try

  Dim buffer As Byte() = System.Text.ASCIIEncoding.ASCII.GetBytes(compressData)

  sr = New IO.MemoryStream()

  compressedzipStream = New GZipStream(sr, CompressionMode.Compress)

  compressedzipStream.Write(buffer, 0, buffer.Length)

  sr.Flush()
  Return sr.ToArray
  
  'sr.Flush()
  'compressedzipStream.Flush()
  'compressedzipStream.Close()
  'sr.Close()
  'sr.Dispose()
  'destinationStream.Flush()
  'destinationStream.Close()
  
  Catch ex As Exception

  MsgBox(ex.ToString)
  

  Finally
     'Close stream if it is opened - will automatically flush data
     If Not compressedzipStream Is Nothing
        compressedzipStream.Close()
    
     'Close extra memory stream
     If Not sr Is Nothing
        sr.Close()
  End Try

  End Function
  

  My VB isn't too great so you might need to make changes but in a nutshell it cleans up the resources (streams) in the finally block to guarantee that they are always properly cleaned up.  Notice that the return statement stays in this case because this is the last thing you are doing in the function.  However you have to flush the sr stream first otherwise you might not get all the data.
  
  Hope this helps,
  Michael Taylor - 11/11/05
  
  

• Scottie J

  Hi all, i'm also facing some problem on filestream while i'm trying to Gzip a text file using VS 2005 system.io.compression. The code that i wrote work fine that it creates a zip file but the extension of the original file inside the zip file has been cut off. For example, the original file is Text.txt, after zip, the name of the file inside only left Text. Then I have to rename the file to Text.txt then can only i open the file.
  
  I included the code that i try out (download through internet) as below. Hope that someone can help me.
  
  

  Public Function compresserFichier(ByVal cheminSource As String, ByVal cheminDestination As String) As Boolean

  Try

  Dim monFileStream As FileStream

  monFileStream = New FileStream(cheminSource, FileMode.Open)

  Dim monBuffer As Byte()

  monBuffer = New Byte(monFileStream.Length) {}

  monFileStream.Read(monBuffer, 0, System.Convert.ToInt32(monFileStream.Length))

  monFileStream.Close()

  monFileStream = New FileStream(cheminDestination, FileMode.Create)

  Dim monGZipStream As GZipStream

  monGZipStream = New GZipStream(monFileStream, CompressionMode.Compress, False)

  monGZipStream.Write(monBuffer, 0, monBuffer.Length)

  monGZipStream.Close()

  monFileStream.Close()

  Return True

  Catch ex As Exception

  MessageBox.Show(ex.Message)

  Return False

  End Try

  End Function

  
  
  

• John1984

  I actually have narrowed it down now, not to the actual compress/decompress, but when I am saving it to the database. For some reason when I am updating the db it cuts off part of the data.
  
  this is where I set the data to the dataset
  

  Me.InventorydbDataSet.Tables("webXml").Rows.Find(saveName).Item("xml9") = sr.toarray 'sr.toarray is the byte array

  
  
  
  here is the update code
  

  Me.WebXmlBindingSource.EndEdit() Me.WebXmlTableAdapter.Update(Me.InventorydbDataSet.webXml) Me.InventorydbDataSet.webXml.AcceptChanges() Me.WebXmlTableAdapter.Fill(Me.InventorydbDataSet.webXml)

  
  

• acplus

  Thank you for your reply. I changed the my code according to what you said, and also changed my decompression code to what I believe matches, but the problem still exists, which sucks lol. so here is my updated decompress code. Thanks again!
  
  

  Public Function decomp(ByVal decompExp As Object) As String Dim objZS As System.IO.Compression.GZipStream Dim mem As New MemoryStream Dim abyt() As Byte abyt = CType(decompExp, Byte()) Dim ms As New IO.MemoryStream(abyt) Try objZS = New System.IO.Compression.GZipStream(ms, System.IO.Compression.CompressionMode.Decompress)   Dim buffer As Byte() = New Byte(32767) {} Dim bytesRead As Integer = 0 While True bytesRead = objZS.Read(buffer, 0, buffer.Length) If bytesRead > 0 Then mem.Write(buffer, 0, bytesRead) Else Exit While End If End While ms.Flush() bind.Filter = Nothing Return System.Text.ASCIIEncoding.ASCII.GetString(mem.ToArray) Catch ex As Exception MsgBox(ex.ToString) Finally If Not objZS Is Nothing Then objZS.Close() End If If Not ms Is Nothing Then ms.Close() End If If Not mem Is Nothing Then mem.Close() End If End Try End Function

  
  

• JerseyGRL1970

  Without some more info I'm not sure.  You seem to be trying to put a byte array into a row of a DataTable.  Do you mean to do that or do you want it as a string   If you want it as a string then you'll need to convert it first.
  
  I'm not sure what the WebXml thing you are using is other than perhaps a DataSet.  Did you verify that the column in the row contained all the data before you flushed it to the DB   It seems that maybe it isn't going to like the array.  What type is the column
  Why do you call Fill again   Is the row valid before you call Fill and invalid after   If so then maybe the underlying DataSet you are using is not pushing all the data out.
  
  Michael Taylor - 11/11/05
  
  

• Nino Crudele

  Hi all,
  
  I think I have solve the problem that I post. For the last post I call the compresserFichier function through:

  m_objFile.compresserFichier("C:\\Test.txt", "D:\\Test.gz")
  
  Now I change the call to:
  
  

  m_objFile.compresserFichier("C:\\Test.txt", "D:\\Test.txt.gz")
  
  So for the zip name I want to use, I change it to original file name + .gz extension. So the Text.txt keep it's original file name and extension. But I don't really understand why it is like that, can someone help explain to me.
  
  Thanks all.

  
  
  
• Stream Not writing last byte?