When to call WebResponse.Close()

Question

WebResponse response;
try
{                
 HttpWebRequest request = (HttpWebRequest)WebRequest.Create(url);
 request.Timeout = 20000;
 response = request.GetResponse();

 request = (HttpWebRequest)WebRequest.Create(url2);
 response = request.GetResponse();
}
catch(Exception ex)
{
 //do something
}              
finally
{
}

where should response.Close() be called?

• after every GetResponse() in try?

• after last GetResponse() in try - once?

• in finally block?

Answer 1

None of the above. You should be using a using block:

HttpWebRequest request = (HttpWebRequest)WebRequest.Create(url);
request.Timeout = 20000;
using (WebResponse response = request.GetResponse())
{
    using (var stream = response.GetResponseStream())
    {
        using (var reader = new StreamReader(stream))
        {
            var result = reader.ReadToEnd();
            // Do something with result
        }
    }
}

A using block will ensure that the Dispose method is called, whether or not there is an exception. Dispose will do the same thing as Close.

using (var d = new DisposableClass()){code;}

is equivalent to:

DisposableClass d = null;
try
{
    d = new DisposableClass();
    code;
}
finally
{
    if (d != null)
        ((IDisposable)d).Dispose();
}

Answer 2

Put it in the finally block. As per MSDN:

The finally block is useful for cleaning up any resources allocated in the try block as well as running any code that must execute even if there is an exception. Control is always passed to the finally block regardless of how the try block exits.

Answer 3

Note that nested using blocks don't need curly braces, improving readability. So John Saunder's code could be written:

HttpWebRequest request = (HttpWebRequest)WebRequest.Create(url);
request.Timeout = 20000;
using (WebResponse response = request.GetResponse())
using (var stream = response.GetResponseStream())
using (var reader = new StreamReader(stream))
{
    var result = reader.ReadToEnd();
    // Do something with result
}

VS.NET understands that such nested blocks don't need indenting. Note btw that if you know the encoding of the response or are going to ignore it anyhow, WebClient provides a simpler API - missing header information, so Header-based (transfer/text) encoding detection becomes impossible, but otherwise it works fine.

Answer 4

I would suggest the below

        try
        {
            HttpWebRequest request = (HttpWebRequest)WebRequest.Create("http://www.google.com");
            request.Timeout = 20000;
            using (var response = request.GetResponse())
            {
                //Do something with response.
            }


            request = (HttpWebRequest)WebRequest.Create("http://www.bing.com");
            using (var response = request.GetResponse())
            {
                //Do somehing with response
            }
        }
        catch (Exception ex)
        {
            //do something
        }
        finally
        {
        }