Creating a single page proxy using Ruby Sinatra

Question

I am trying to use Ruby Sinatra to create a simple proxy for a specific web page. I can do it in C#, I just can't seem to work it out for Sinatra, the C# code is below:

<%@ WebHandler Language="C#" Class="Map" %>

using System;
using System.Web;
using System.Net;
using System.IO;

public class Map : IHttpHandler {

static void CopyStream(Stream input, Stream output)
{
    byte[] buffer = new byte[0x1000];
    int read;
    while ((read = input.Read(buffer, 0, buffer.Length)) > 0)
        output.Write(buffer, 0, read);
}

public void ProcessRequest(HttpContext context)
{
    string gmapUri = string.Format("http://maps.google.com/maps/api/staticmap{0}", context.Request.Url.Query);
    WebRequest request = WebRequest.Create(gmapUri);

    using (WebResponse response = request.GetResponse())
    {
        context.Response.ContentType = response.ContentType;
        Stream responseStream = response.GetResponseStream();

        CopyStream(responseStream, context.Response.OutputStream);
    }
}

public bool IsReusable {
    get {
        return false;
    }
}

}

The Ruby Sinatra code I have tried is as follows:

require 'rubygems'
require 'sinatra'

get '/mapsproxy/staticmap' do
  request.path_info = 'http://maps.google.com/maps/api/staticmap'
  pass
end

I am assuming that the Sinatra one does not work (get a 404) as is is only passing the request to pages in the same domain. Any hep would be greatly appreciated.

EDIT:

With the Tin Man's help I've come up with a nice succinct solution, which works well for me:

get '/proxy/path' do
   URI.parse(<URI> + request.query_string.gsub("|", "%7C")).read
end

Thanks for all the help.

Answer 1

If you want your Sinatra app to retrieve the URL, you'll need to fire up a HTTP client of some sort:

get '/mapsproxy/staticmap' do
  require 'open-uri'
  open('http://maps.google.com/maps/api/staticmap').read
end

I think this will work and is about as minimal as you can get.

You could use HTTPClient if you need more tweakability.

Also, I think that Rack can do it. Sinatra is built on top of Rack, but it's been a while since I played at that level.

I still need to find a way to extract the contentType from the response

From the Open-URI docs:

The opened file has several methods for meta information as follows since
it is extended by OpenURI::Meta.

open("http://www.ruby-lang.org/en") {|f|
    f.each_line {|line| p line}
    p f.base_uri         # <URI::HTTP:0x40e6ef2 URL:http://www.ruby-lang.org/en/>
    p f.content_type     # "text/html"
    p f.charset          # "iso-8859-1"
    p f.content_encoding # []
    p f.last_modified    # Thu Dec 05 02:45:02 UTC 2002
}

For your purposes something like this should work:

content_type = ''
body = open("http://www.ruby-lang.org/en") {|f|
  content_type = f.content_type     # "text/html"
  f.read
}

I haven't tested that, but I think the return value of the block will be assigned to body. If that doesn't work then try:

content_type = ''
body = ''
open("http://www.ruby-lang.org/en") {|f|
  content_type = f.content_type     # "text/html"
  body = f.read
}

but I think the first will work.

Answer 2

With the help of the Tin Man and TK-421 I've worked out a solution, see the Sinatra route below:

get '/proxy/path' do
  require 'open-uri'
  uri = URI.parse(<URI>)
  getresult = uri.read
  halt 200, {'Content-Type' => getresult.content_type}, getresult
end

Just replace the <URI> with the page you require, and you're good to go.

After some more playing this is what I've come up with:

get '/proxy/path' do
   URI.parse(<URI> + request.query_string.gsub("|", "%7C")).read
end

As mentioned else where you need to require 'open-uri' at the top of the code. The reason for the gsub is that for some reason the parse fails if they are left in, and my browser doesn't encode them automatically.