Question

I would like to use the WebClient (or there is another better option?) but there is a problem. I understand that opening up the stream takes some time and this can not be avoided. However, reading it takes a strangely much more amount of time compared to read it entirely immediately.

Is there a best way to do this? I mean two ways, to string and to file. Progress is my own delegate and it’s working good.

---

FIFTH UPDATE:

Finally, I managed to do it. In the meantime I checked out some solutions what made me realize that the problem lies elsewhere.

I’ve tested custom WebResponse and WebRequest objects, library libCURL.NET and even Sockets.

The difference in time was gzip compression. Compressed stream lenght was simply half the normal stream lenght and thus download time was less than 3 seconds with the browser.

I put some code if someone will want to know how i solved this: (some headers are not needed)

public static string DownloadString(string URL)
    {
        WebClient client = new WebClient();
        client.Headers["User-Agent"] = "Mozilla/5.0 (Windows; U; Windows NT 6.1; en-US) AppleWebKit/532.5 (KHTML, like Gecko) Chrome/4.1.249.1045 Safari/532.5";
        client.Headers["Accept"] = "application/xml,application/xhtml+xml,text/html;q=0.9,text/plain;q=0.8,image/png,*/*;q=0.5";
        client.Headers["Accept-Encoding"] = "gzip,deflate,sdch";
        client.Headers["Accept-Charset"] = "ISO-8859-2,utf-8;q=0.7,*;q=0.3";

        Stream inputStream = client.OpenRead(new Uri(URL));
        MemoryStream memoryStream = new MemoryStream();
        const int size = 32 * 4096;
        byte[] buffer = new byte[size];

        if (client.ResponseHeaders["Content-Encoding"] == "gzip")
        {
            inputStream = new GZipStream(inputStream, CompressionMode.Decompress);
        }

        int count = 0;
        do
        {
            count = inputStream.Read(buffer, 0, size);
            if (count > 0)
            {
                memoryStream.Write(buffer, 0, count);
            }
        }
        while (count > 0); 

        string result = Encoding.Default.GetString(memoryStream.ToArray());
        memoryStream.Close();
        inputStream.Close();
        return result;
    }

I think that asyncro functions will be almost the same. But i will simply use another thread to fire this function. I dont need percise progress indication.

Answer 1

You only get the last iSize bytes from your file since you overwrite your buffer on each iteration, you’re not saving the buffer anywhere. Here’s a sample on how to store the file in memory using a MemoryStream.

var totalBytes = new MemoryStream(1024 * 1024);
while ((iByteSize = streamRemote.Read(byteBuffer, 0, iByteSize)) > 0)
{
    totalBytes.Write(byteBuffer, 0, iByteSize);
    iRunningByteTotal += iByteSize;

    //Some progress calculation
    if (Progress != null) Progress(iProgressPercentage);
}

When the whole download is complete, then you can convert it into text.

var byteArray = totalBytes.GetBuffer();
var numberOfBytes = totalBytes.Length;
var text = Encoding.Default.GetString(byteArray, 0, numberOfBytes);

Update: the DownloadStringAsync method basically does the same as above, but will not give you any progress indication. There are some other async methods though which will fire the DownloadProgressChanged event.

Update 2: Regarding response time. Have you timed download of the resource using some other tool? Major browsers have builtin support for timing such tings.

Further, is it a static file your serving up or is the content generated on the serverside?

A third thing that comes to mind is serverside buffering. E.g. if the Response.Buffer property in ASP.Net is used, nothing will be sent to the client until the whole file/page is done serverside. Thus the client will have to wait before it can start downloading.