HttpWebResponse流读取不均匀的块

Question

I am using the stream from HttpWebRequest.GetResponse().GetResponseStream() to read data from a streaming web API. I use Begin / EndRead on the stream with a buffer of 65K bytes. I can see that data is being returned in the following pattern:

16383 bytes read.
1 bytes read.
16383 bytes read.
1 bytes read.
16383 bytes read.
1 bytes read.
etc...

Obviously the 1 byte reads introduce a lot of inefficiency in the process, and the buffer size I provide is large enough to fit 16384 bytes or more. Is there anything I can do as a client to improve this or is simply up to the server how it's streaming data to me?

The reader code is basically:

var buffer = new byte[65536];
using (var stream = response.GetResponseStream()) {
    while (true) {
        var bytesRead = await AsyncRead(stream.BeginRead, stream.EndRead, buffer);
        Console.WriteLine($"{bytesRead} bytes read.");
        // do something with the bytes
    }
}

where AsyncRead just calls BeginRead(buffer, 0, buffer.Length, callback, null) , then EndRead in the callback and returns the return value of EndRead .

BTW this is on .NET 4.0, no HttpClient.

Answer 1

What exactly are you trying to achieve by sending the HTTPWebRequest to the target server?

Are you trying to read a live response from the server after asking the server for data or initializing a request between your client application and the target server? If you are try to send an HTTPWepRequest and an HTTPWebResponse to the target server then convert the response given from the server to a stream then use System.IO.StreamReader to read the in-coming stream!

Then just to be on the safe side convert the stream that was read by the System.IO.StreamReader method into UTF-8 if that is your overall goal!

After converting to UTF-8 you can now print the output stream into a string value which you can output to console or wherever you want to send the output string!

I hope this is what you wanted if not then I have been essentially useless! :P