Question

We have a text file that gets generated automatically and put in a web server. The task is to read the file line by line and insert the records in a database. The following code is in C#:

            WebRequest request = WebRequest.Create(url);
            WebResponse response = request.GetResponse();
            StreamReader r = new StreamReader(response.GetResponseStream());

            while (r.Peek() > -1)
            {
                string s = r.ReadLine().Trim();
                //insert string into a db.
            }

When I do this I constantly get the entire file which ranges from 9000 – 10000 lines. On the other hand when I use the following sometimes I get a truncated file (less lines)

            WebClient client = new WebClient();
            StreamReader r = new StreamReader(client.OpenRead(url));

            while (r.Peek() > -1)
            {
                string s = r.ReadLine().Trim();
                //insert string into a db.
            }

Can anyone explain the difference? Why would the results be different? I was under the impression that WebClient was just a wrapper of HttpWebRequest.

Answer 1

using Peek method, you are guaranteed to have a complete line ready to be red.

I bet you are getting partial lines (actually breaking at the current stream position).

In fact, I don’t think it’s related to either of the two class, but more to how you read the result.

Did you tried WebClient.DownloadString() method ?