Unicorn::App::Inetd: reinventing Unix, poorly :)

This includes an example of tunneling the git protocol inside a
TE:chunked HTTP request.  The example is unfortunately contrived
in that it relies on the custom examples/cat-chunk-proxy.rb
script in the client.  My initial wish was to have a generic
tool like curl(1) operate like this:

  cat > ~/bin/cat-chunk-proxy.sh <<EOF
  #!/bin/sh
  exec curl -sfNT- http://$1:$2/
  EOF
  chmod +x ~/bin/cat-chunk-proxy.sh
  GIT_PROXY_COMMAND=cat-chunk-proxy.sh git clone git://0:8080/foo

Unfortunately, curl will attempt a blocking read on stdin before
reading the TCP socket; causing the git-clone consumer to
starve.  This does not appear to be a problem with the new
server code for handling chunked requests.

1 files changed, 106 insertions, 0 deletions

diff --git a/lib/unicorn/app/inetd.rb b/lib/unicorn/app/inetd.rb
new file mode 100644
index 0000000..97dc5d3
--- /dev/null
+++ b/lib/unicorn/app/inetd.rb
@@ -0,0 +1,106 @@
+# this class *must* be used with Rack::Chunked
+
+module Unicorn::App
+  class Inetd
+
+    Z = ''
+    Z.force_encoding(Encoding::BINARY) if Z.respond_to?(:force_encoding)
+
+    class CatBody
+      def initialize(env, cmd)
+        @cmd = cmd
+        @input, @errors = env['rack.input'], env['rack.errors']
+        in_rd, in_wr = IO.pipe
+        @err_rd, err_wr = IO.pipe
+        @out_rd, out_wr = IO.pipe
+
+        @cmd_pid = fork {
+          inp, out, err = (0..2).map { |i| IO.new(i) }
+          inp.reopen(in_rd)
+          out.reopen(out_wr)
+          err.reopen(err_wr)
+          [ in_rd, in_wr, @err_rd, err_wr, @out_rd, out_wr ].each { |io|
+            io.close
+          }
+          exec(*cmd)
+        }
+        [ in_rd, err_wr, out_wr ].each { |io| io.close }
+        [ in_wr, @err_rd, @out_rd ].each { |io| io.binmode }
+        in_wr.sync = true
+
+        # Unfortunately, input here must be processed inside a seperate
+        # thread/process using blocking I/O since env['rack.input'] is not
+        # IO.select-able and attempting to make it so would trip Rack::Lint
+        @inp_pid = fork {
+          [ @err_rd, @out_rd ].each { |io| io.close }
+          buf = Z.dup
+
+          # this is dependent on @input.read having readpartial semantics:
+          while @input.read(16384, buf)
+            in_wr.write(buf)
+          end
+          in_wr.close
+        }
+        in_wr.close
+      end
+
+      def each(&block)
+        buf = Z.dup
+        begin
+          rd, = IO.select([@err_rd, @out_rd])
+          rd && rd.first or next
+
+          if rd.include?(@err_rd)
+            begin
+              @errors.write(@err_rd.read_nonblock(16384, buf))
+            rescue Errno::EINTR
+            rescue Errno::EAGAIN
+              break
+            end while true
+          end
+
+          rd.include?(@out_rd) or next
+
+          begin
+            yield @out_rd.read_nonblock(16384, buf)
+          rescue Errno::EINTR
+          rescue Errno::EAGAIN
+            break
+          end while true
+        rescue EOFError,Errno::EPIPE,Errno::EBADF,Errno::EINVAL
+          break
+        end while true
+
+        self
+      end
+
+      def close
+        @input = nil
+        [ [ @cmd.inspect, @cmd_pid ], [ 'input streamer', @inp_pid ]
+        ].each { |str, pid|
+          begin
+            pid, status = Process.waitpid2(pid)
+            status.success? or
+              @errors.write("#{str}: #{status.inspect} (PID:#{pid})\n")
+          rescue Errno::ECHILD
+            @errors.write("Failed to reap #{str} (PID:#{pid})\n")
+          end
+        }
+      end
+
+    end
+
+    def initialize(*cmd)
+      # enable streaming input mode in Unicorn
+      Unicorn::HttpRequest::DEFAULTS["unicorn.stream_input"] = true
+      @cmd = cmd
+    end
+
+    def call(env)
+      [ 200, { 'Content-Type' => 'application/octet-stream' },
+       CatBody.new(env, @cmd) ]
+    end
+
+  end
+
+end