Reading from a URL Connection Java

Question

I'm trying to read html code from a URL Connection. In one case the html file I'm trying to read includes 5 line breaks before the actual doc type declaration. In this case the input reader throws an exception for EOF.

URL pageUrl = 
    new URL(
        "http://www.nytimes.com/2011/03/15/sports/basketball/15nbaround.html"
    );

URLConnection getConn = pageUrl.openConnection();
getConn.connect();
DataInputStream dis = new DataInputStream(getConn.getInputStream());
//some read method here

Has anyone ran into a problem like this?

URL pageUrl = new URL("http://www.nytimes.com/2011/03/15/sports/basketball/15nbaround.html");
URLConnection getConn = pageUrl.openConnection();
getConn.connect();
DataInputStream dis = new DataInputStream(getConn.getInputStream());
String urlData = "";
while ((urlData = dis.readUTF()) != null)
    System.out.println(urlData);

//exception thrown

java.io.EOFException at java.io.DataInputStream.readUnsignedShort(DataInputStream.java:323) at java.io.DataInputStream.readUTF(DataInputStream.java:572) at java.io.DataInputStream.readUTF(DataInputStream.java:547)

in the case of bufferedreader, it just responds null and doesn't continue

pageUrl = new URL("http://www.nytimes.com/2011/03/15/sports/basketball/15nbaround.html");
URLConnection getConn = pageUrl.openConnection();
getConn.connect();
BufferedReader br = new BufferedReader(new InputStreamReader(getConn.getInputStream()));
String urlData = "";
while(true)
     urlData = br.readLine();
     System.out.println(urlData);

outputs null

Answer 1

You're using DataInputStream to read data that wasn't encoded using DataOutputStream. Examine the documented behavior for your call to DataInputStream#readUtf(); it first reads two bytes to form a 16-bit integer, indicating the number of bytes that follow comprising the UTF-encoded string. The data you're reading from the HTTP server is not encoded in this format.

Instead, the HTTP server is sending headers encoded in ASCII, per RFC 2616 sections 6.1 and 2.2. You need to read the headers as text, and then determine how the message body (the "entity") is encoded.

Answer 2

This works fine:

package url;

import java.io.BufferedReader;
import java.io.IOException;
import java.io.InputStreamReader;
import java.io.Reader;
import java.net.URL;

/**
 * UrlReader
 * @author Michael
 * @since 3/20/11
 */
public class UrlReader
{

    public static void main(String[] args)
    {
        UrlReader urlReader = new UrlReader();

        for (String url : args)
        {
            try
            {
                String contents = urlReader.readContents(url);
                System.out.printf("url: %s contents: %s\n", url, contents);
            }
            catch (Exception e)
            {
                e.printStackTrace();
            }
        }
    }


    public String readContents(String address) throws IOException
    {
        StringBuilder contents = new StringBuilder(2048);
        BufferedReader br = null;

        try
        {
            URL url = new URL(address);
            br = new BufferedReader(new InputStreamReader(url.openStream()));
            String line = "";
            while (line != null)
            {
                line = br.readLine();
                contents.append(line);
            }
        }
        finally
        {
            close(br);
        }

        return contents.toString();
    }

    private static void close(Reader br)
    {
        try
        {
            if (br != null)
            {
                br.close();
            }
        }
        catch (Exception e)
        {
            e.printStackTrace();
        }
    }
}