Getting the response of a Httpurl connection back to the web browser correctly

Question

Problem:Writing the response off a http message back to the client(web browser) does not return the full page if the page has images when using strings so i decided to use bytes but i still have the same issue.I have been able to get the header from the request as string and flushed it to the client but i am not sure what to do with the message to ensure that it appears properly on the web browser.

        //This portion takes the message from the Httpurl connection inputstream
        //the header has already been exttracted
        //uc here represents a httpurlconnection
       byte[] data = new byte[uc.getContentLength()];
        int bytesRead = 0;
        int offset = 0;
        InputStream in = new BufferedInputStream(uc.getInputStream());
        while (offset < uc.getContentLength()) {
        bytesRead =in.read(data, offset, data.length-offset);
        if (bytesRead == -1) break;
        offset += bytesRead;

score 0 · Answer 1 · answered Feb 14 '13 at 23:32

I would advice you to write the bytes to your response as you read them, and use a small buffer, this way your server won't be affected by large memory usage. It is not a good practice to load up all the bytes into an array in the server's memory.

Here is a quick sample:

response.setContentType("text/html;charset=UTF-8");
    OutputStream out = response.getOutputStream();
    HttpURLConnection uc;

// Setup the HTTP connection...

InputStream in = uc.getInputStream();
byte[] b = new byte[1024];
int bytesRead = 0;
while ( bytesRead != -1 ) {
    bytesRead = in.read(b);
    out.write(b);;
}

// Close the streams...

Setting the response type to charset=utf-8 is what i am trying to avoid.It does not work well on some pages which contain images.That is the problem that i have — user1850254, Feb 14 '13 at 23:36

score 0 · Answer 2 · edited May 23 '17 at 12:27

You seem to be proxying a HTML page with images and you seem to be expecting that images in a HTML page are somehow automagically inlined in the HTML source code. This is thus absolutely not true. Images in HTML are represented by the <img> element with a src attribute pointing to an URL which the webbrowser has to invoke and download individually. Exactly the same story applies to other resources like CSS and JS files.

You basically need to parse the obtained HTML, scan for all <img src> (and if necessary also <link href> and <script src>) elements and change their URL to the URL of your proxy so that it can serve the desired image (and CSS/JS) resources individually.

You can find a kickoff example in this answer on a related question: Make HttpURLConnection load web pages with images

Getting the response of a Httpurl connection back to the web browser correctly

2 Answers2