forward-proxy 0.1.5 → 0.5.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
-SHA1:
-  metadata.gz: 3e94626ce91d870564fcbed5db9133ddee5a1167
-  data.tar.gz: 44c9c4e4ca3a4036a8c1e9698dac6a203c3fe367
+SHA256:
+  metadata.gz: ccf19df114eeec55eaef8765999561904d8f5b6cd91b7ed2016053ec9e717408
+  data.tar.gz: ce712f520c0ed32b93ea211ff658d330bef7eeb26c2ddfcea1246d955ce78175
 SHA512:
-  metadata.gz: c13a9fa915fb77699702226408cbd160d61cd375789beb947299423c788594c7f5d7af3783624fcf5fb30e8521e8a5e98c4953e7f1b5c7e477a756bda153634b
-  data.tar.gz: 609126058336e8cc597c0df21b3a3cb65c4d1216d7af11d77dec09eda104c6277811d59b942d38c46f8776c0cf1b8f137d09778793343a88c138872f65b06b54
+  metadata.gz: b71608455b079161950b09521d8fd99496767463f3e2aa0f0d293f21e3955b03d7b0ba12f06f3b3df1ff6674d598ed2e7eec7492ffb8fd9047ebb19771625251
+  data.tar.gz: '092035474f162e29d2c7d0735b43cd38e41580aadca7ae50e97e02dee8e6f3825f50393497c0a8db8a5e0b3622d3e44e25e69e03ae02b56eba1cad55abdc1388'

data/CHANGELOG.md

@@ -1,5 +1,9 @@
 # CHANGELOG
 
+## 0.2.0
+
+- Extract errors into module.
+
 ## 0.1.5
 
 - stream http proxy requests.

data/README.md

@@ -2,12 +2,12 @@
 
 ![Gem Version][3] ![Gem][1] ![Build Status][2]
 
-100 LOC Ruby forward proxy using only standard libraries.
+Minimal forward proxy using 150LOC and only standard libraries. Useful for development, testing, and learning.
 
 ```
 $ forward-proxy --binding 0.0.0.0 --port 3182 --threads 2
-[2021-01-14 19:37:47 +1100] INFO Listening 0.0.0.0:3182
-[2021-01-14 19:38:24 +1100] INFO CONNECT raw.githubusercontent.com:443 HTTP/1.1
+I, [2021-07-04T10:33:32.947653 #1790]  INFO -- : Listening 0.0.0.0:3182
+I, [2021-07-04T10:33:32.998298 #1790]  INFO -- : CONNECT raw.githubusercontent.com:443 HTTP/1.1
 ```
 
 ## Installation

data/exe/forward-proxy

@@ -7,7 +7,7 @@ options = {}
 OptionParser.new do |parser|
   parser.banner = "Usage: forward-proxy [options]"
 
-  parser.on("-pPORT", "--port=PORT", "Bind to specified port. Default: 9292", String) do |bind_port|
+  parser.on("-pPORT", "--port=PORT", String, "Bind to specified port. Default: 9292") do |bind_port|
     options[:bind_port] = bind_port
   end
 
@@ -15,7 +15,8 @@ OptionParser.new do |parser|
     options[:bind_address] = bind_address
   end
 
-  parser.on("-tTHREADS", "--threads=THREADS", Integer, "Specify the number of client threads. Default: 32") do |threads|
+  parser.on("-tTHREADS", "--threads=THREADS", Integer,
+            "Specify the number of client threads. Default: 128") do |threads|
     options[:threads] = threads
   end
 
@@ -26,6 +27,5 @@ OptionParser.new do |parser|
 end.parse!
 
 require_relative '../lib/forward_proxy'
-
 server = ForwardProxy::Server.new(options)
 server.start

data/lib/forward_proxy.rb

@@ -1,4 +1,4 @@
-$:.unshift File.dirname(__FILE__)
+$LOAD_PATH.unshift File.dirname(__FILE__)
 
 require 'forward_proxy/version'
 require 'forward_proxy/server'

data/lib/forward_proxy/errors/http_method_not_implemented.rb

@@ -0,0 +1,5 @@
+module ForwardProxy
+  module Errors
+    class HTTPMethodNotImplemented < StandardError; end
+  end
+end

data/lib/forward_proxy/errors/http_parse_error.rb

@@ -0,0 +1,5 @@
+module ForwardProxy
+  module Errors
+    class HTTPParseError < StandardError; end
+  end
+end

data/lib/forward_proxy/server.rb

@@ -1,64 +1,44 @@
-require 'forward_proxy/thread_pool'
+require 'logger'
 require 'socket'
 require 'webrick'
 require 'net/http'
+require 'forward_proxy/errors/http_method_not_implemented'
+require 'forward_proxy/errors/http_parse_error'
+require 'forward_proxy/thread_pool'
 
 module ForwardProxy
-  class HTTPMethodNotImplemented < StandardError; end
-  class HTTPParseError < StandardError; end
-
   class Server
-    attr_reader :bind_address, :bind_port
+    attr_reader :bind_address, :bind_port, :logger
 
-    def initialize(bind_address: "127.0.0.1", bind_port: 9292, threads: 32)
-      @thread_pool = ThreadPool.new(threads)
+    def initialize(bind_address: "127.0.0.1", bind_port: 9292, threads: 128, logger: default_logger)
       @bind_address = bind_address
       @bind_port = bind_port
+      @logger = logger
+      @thread_pool = ThreadPool.new(threads)
     end
 
     def start
       thread_pool.start
 
-      @server = TCPServer.new(bind_address, bind_port)
+      @socket = TCPServer.new(bind_address, bind_port)
 
-      log("Listening #{bind_address}:#{bind_port}")
+      logger.info("Listening #{bind_address}:#{bind_port}")
 
       loop do
-        # The following comments are from https://stackoverflow.com/q/5124320/273101
-        #
-        # accept(): POSIX.1-2001, POSIX.1-2008, SVr4, 4.4BSD (accept() first appeared in 4.2BSD).
-        #
-        # We see that POSIX.1-2008 is a viable reference (check this for a description of relevant
-        # standards for Linux systems). As already said in other answers, POSIX.1 standard specifies
-        # accept function as (POSIX-)thread safe (as defined in Base Definitions, section 3.399 Thread Safe)
-        # by not listing it on System Interfaces, section 2.9.1 Thread Safety.
-        thread_pool.schedule(server.accept) do |client_conn|
+        thread_pool.schedule(socket.accept) do |client_conn|
           begin
             req = parse_req(client_conn)
 
-            log(req.request_line)
+            logger.info(req.request_line.strip)
 
             case req.request_method
             when METHOD_CONNECT then handle_tunnel(client_conn, req)
             when METHOD_GET, METHOD_POST then handle(client_conn, req)
             else
-              raise HTTPMethodNotImplemented
+              raise Errors::HTTPMethodNotImplemented
             end
           rescue => e
-            # The following comments are from the IETF document
-            # "Hypertext Transfer Protocol (HTTP/1.1): Semantics and Content"
-            # https://tools.ietf.org/html/rfc7231#section-6.6.3
-
-            # The 502 (Bad Gateway) status code indicates that the server, while
-            # acting as a gateway or proxy, received an invalid response from an
-            # inbound server it accessed while attempting to fulfill the request.
-            client_conn.puts <<~eos.chomp
-              HTTP/1.1 502
-              Via: #{HEADER_VIA}
-            eos
-
-            puts e.message
-            puts e.backtrace.map { |line| "  #{line}" }
+            handle_error(e, client_conn)
           ensure
             client_conn.close
           end
@@ -67,22 +47,33 @@ module ForwardProxy
     rescue Interrupt
       shutdown
     rescue IOError, Errno::EBADF => e
-      log(e.message, "ERROR")
+      logger.error(e.message)
     end
 
     def shutdown
-      log("Stoping server...")
-      server.close if server
+      if socket
+        logger.info("Shutting down")
+
+        socket.close
+      end
     end
 
     private
 
-    attr_reader :server, :thread_pool
+    attr_reader :socket, :thread_pool
+
+    # The following comments are from the IETF document
+    # "Hypertext Transfer Protocol -- HTTP/1.1: Basic Rules"
+    # https://datatracker.ietf.org/doc/html/rfc2616#section-2.2
 
     METHOD_CONNECT = "CONNECT"
     METHOD_GET = "GET"
     METHOD_POST = "POST"
 
+    # HTTP/1.1 defines the sequence CR LF as the end-of-line marker for all
+    # protocol elements except the entity-body.
+    HEADER_EOP = "\r\n"
+
     # The following comments are from the IETF document
     # "Hypertext Transfer Protocol (HTTP/1.1): Semantics and Content"
     # https://tools.ietf.org/html/rfc7231#section-4.3.6
@@ -115,7 +106,10 @@ module ForwardProxy
       # blank line that concludes the successful response's header section;
       # data received after that blank line is from the server identified by
       # the request-target.
-      client_conn.write "HTTP/1.1 200 OK\n\n"
+      client_conn.write <<~eos.chomp
+        HTTP/1.1 200 OK
+        #{HEADER_EOP}
+      eos
 
       # The CONNECT method requests that the recipient establish a tunnel to
       # the destination origin server identified by the request-target and,
@@ -132,12 +126,38 @@ module ForwardProxy
     def handle(client_conn, req)
       Net::HTTP.start(req.host, req.port) do |http|
         http.request(map_webrick_to_net_http_req(req)) do |resp|
+          # The following comments are from the IETF document
+          # "Hypertext Transfer Protocol (HTTP/1.1): Semantics and Content"
+          # https://tools.ietf.org/html/rfc7231#section-4.3.6
+
+          # An intermediary MAY combine an ordered subsequence of Via header
+          # field entries into a single such entry if the entries have identical
+          # received-protocol values.  For example,
+          #
+          #   Via: 1.0 ricky, 1.1 ethel, 1.1 fred, 1.0 lucy
+          #
+          # could be collapsed to
+          #
+          #   Via: 1.0 ricky, 1.1 mertz, 1.0 lucy
+          #
+          # A sender SHOULD NOT combine multiple entries unless they are all
+          # under the same organizational control and the hosts have already been
+          # replaced by pseudonyms.  A sender MUST NOT combine entries that have
+          # different received-protocol values.
+          headers = resp.to_hash.merge(Via: [HEADER_VIA, resp['Via']].compact.join(', '))
+
           client_conn.puts <<~eos.chomp
             HTTP/1.1 #{resp.code}
-            Via: #{[HEADER_VIA, resp['Via']].compact.join(', ')}
-            #{resp.each.map { |header, value| "#{header}: #{value}" }.join("\n")}\n\n
+            #{headers.map { |header, value| "#{header}: #{value}" }.join("\n")}
+            #{HEADER_EOP}
           eos
 
+          # The following comments are taken from:
+          # https://docs.ruby-lang.org/en/2.0.0/Net/HTTP.html#class-Net::HTTP-label-Streaming+Response+Bodies
+
+          # By default Net::HTTP reads an entire response into memory. If you are
+          # handling large files or wish to implement a progress bar you can
+          # instead stream the body directly to an IO.
           resp.read_body do |chunk|
             client_conn << chunk
           end
@@ -145,6 +165,21 @@ module ForwardProxy
       end
     end
 
+    def handle_error(err, client_conn)
+      client_conn.puts <<~eos.chomp
+        HTTP/1.1 502
+        Via: #{HEADER_VIA}
+        #{HEADER_EOP}
+      eos
+
+      logger.error(err.message)
+      logger.debug(err.backtrace.join("\n"))
+    end
+
+    def default_logger
+      Logger.new(STDOUT, level: :info)
+    end
+
     def map_webrick_to_net_http_req(req)
       req_headers = Hash[req.header.map { |k, v| [k, v.first] }]
 
@@ -152,16 +187,16 @@ module ForwardProxy
               when METHOD_GET then Net::HTTP::Get
               when METHOD_POST then Net::HTTP::Post
               else
-                raise HTTPMethodNotImplemented
+                raise Errors::HTTPMethodNotImplemented
               end
 
       klass.new(req.path, req_headers)
     end
 
-    def transfer(dest_conn, src_conn)
+    def transfer(src_conn, dest_conn)
       IO.copy_stream(src_conn, dest_conn)
     rescue => e
-      log(e.message, "WARNING")
+      logger.warn(e.message)
     end
 
     def parse_req(client_conn)
@@ -169,11 +204,7 @@ module ForwardProxy
         req.parse(client_conn)
       end
     rescue => e
-      throw HTTPParseError.new(e.message)
-    end
-
-    def log(str, level = 'INFO')
-      puts "[#{Time.now}] #{level} #{str}"
+      throw Errors::HTTPParseError.new(e.message)
     end
   end
 end

data/lib/forward_proxy/thread_pool.rb

@@ -1,21 +1,18 @@
 module ForwardProxy
   class ThreadPool
-    attr_reader :queue, :threads, :size
+    attr_reader :queue, :size
 
     def initialize(size)
-      @size    = size
-      @queue   = Queue.new
-      @threads = []
+      @size  = size
+      @queue = Queue.new
     end
 
     def start
       size.times do
-        threads << Thread.new do
-          catch(:exit) do
-            loop do
-              job, args = queue.pop
-              job.call(*args)
-            end
+        Thread.new do
+          loop do
+            job, args = queue.pop
+            job.call(*args)
           end
         end
       end
@@ -24,13 +21,5 @@ module ForwardProxy
     def schedule(*args, &block)
       queue.push([block, args])
     end
-
-    def shutdown
-      threads.each do
-        schedule { throw :exit }
-      end
-
-      threads.each(&:join)
-    end
   end
 end

data/lib/forward_proxy/version.rb

@@ -1,3 +1,3 @@
 module ForwardProxy
-  VERSION = "0.1.5"
+  VERSION = "0.5.0"
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: forward-proxy
 version: !ruby/object:Gem::Version
-  version: 0.1.5
+  version: 0.5.0
 platform: ruby
 authors:
 - James Moriarty
 autorequire: 
 bindir: exe
 cert_chain: []
-date: 2021-01-20 00:00:00.000000000 Z
+date: 2021-07-13 00:00:00.000000000 Z
 dependencies: []
 description: Forward proxy using just Ruby standard libraries.
 email:
@@ -32,6 +32,8 @@ files:
 - exe/forward-proxy
 - forward-proxy.gemspec
 - lib/forward_proxy.rb
+- lib/forward_proxy/errors/http_method_not_implemented.rb
+- lib/forward_proxy/errors/http_parse_error.rb
 - lib/forward_proxy/server.rb
 - lib/forward_proxy/thread_pool.rb
 - lib/forward_proxy/version.rb
@@ -57,8 +59,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubyforge_project: 
-rubygems_version: 2.5.2.3
+rubygems_version: 3.0.3
 signing_key: 
 specification_version: 4
 summary: Forward proxy.