12

What is the best way to get a file (in this case, a .PDF, but any file will do) from a WebResponse and put it into a MemoryStream? Using .GetResponseStream() from WebResponse gets a Stream object, but if you want to convert that Stream to a specific type of stream, what do you do?

Andrei
  • 55,890
  • 9
  • 87
  • 108

3 Answers3

30

There is a serious issue with SoloBold's answer that I discovered while testing it. When using it to read a file via an FtpWebRequest into a MemoryStream it intermittently failed to read the entire stream into memory. I tracked this down to Peek() sometimes returning -1 after the first 1460 bytes even though a Read() would have succeeded (the file was significantly larger than this).

Instead I propose the solution below:

MemoryStream memStream;
using (Stream response = request.GetResponseStream()) {
    memStream = new MemoryStream();

    byte[] buffer = new byte[1024];
    int byteCount;
    do {
        byteCount = stream.Read(buffer, 0, buffer.Length);
        memStream.Write(buffer, 0, byteCount);
    } while (byteCount > 0);
}

// If you're going to be reading from the stream afterwords you're going to want to seek back to the beginning.
memStream.Seek(0, SeekOrigin.Begin);

// Use memStream as required
Community
  • 1
  • 1
Redwood
  • 66,744
  • 41
  • 126
  • 187
  • 1
    Thanks for the answer, Lawrence. I was getting inconsistent results trying to do this a few other ways. – Cashley Mar 21 '13 at 15:33
  • 1
    When you look at the IL generated (via dotPeek) by the CopyTo(Stream, Int32) method in .net 4.0 you will notice that it generates code with the same pattern but using a single line of code. https://msdn.microsoft.com/en-us/library/system.io.stream.copyto(v=vs.100).aspx. response.CopytTo(memStream,1024); memStream.Position = 0; – Jamie Clayton Feb 23 '15 at 23:18
0

Copied this from the web a year or so ago.

//---------- Start HttpResponse
if(objHttpWebResponse.StatusCode == HttpStatusCode.OK)
    {
        //Get response stream
        objResponseStream = objHttpWebResponse.GetResponseStream();

        //Load response stream into XMLReader
        objXMLReader = new XmlTextReader(objResponseStream);

        //Declare XMLDocument
        XmlDocument xmldoc = new XmlDocument();
        xmldoc.Load(objXMLReader);

        //Set XMLResponse object returned from XMLReader
        XMLResponse = xmldoc;

        //Close XMLReader
        objXMLReader.Close();
    }

    //Close HttpWebResponse
    objHttpWebResponse.Close();
}
Siddharth Rout
  • 147,039
  • 17
  • 206
  • 250
-3

I found the following at http://social.msdn.microsoft.com/Forums/en-US/vbgeneral/thread/eeeefd81-8800-41b2-be63-71acdaddce0e/

    Dim request As WebRequest
    Dim response As WebResponse = Nothing
    Dim s As Stream = Nothing
    Dim fs As FileStream = Nothing
    Dim file As MemoryStream = Nothing

    Dim uri As New Uri(String.Format("http://forums.microsoft.com/forums/ShowPost.aspx?PostID=2992978&SiteID=1"))
    request = WebRequest.Create(uri)
    request.Timeout = 10000
    response = request.GetResponse
    s = response.GetResponseStream

    '2 - Receive file as memorystream
    Dim read(256) As Byte
    Dim count As Int32 = s.Read(read, 0, read.Length)
    File = New MemoryStream
    Do While (count > 0)
        File.Write(read, 0, count)
        count = s.Read(read, 0, read.Length)
    Loop
    File.Position = 0
    'Close responsestream
    s.Close()
    response.Close()

    '3 - Save file
    fs = New FileStream("c:\test.html", FileMode.CreateNew)
    count = file.Read(read, 0, read.Length)
    Do While (count > 0)
        fs.Write(read, 0, count)
        count = file.Read(read, 0, read.Length)
    Loop
    fs.Close()
    File.Close()