GServer implements a generic server, featuring thread pool management, simple logging, and multi-server management. See HttpServer in xmlrpc/httpserver.rb in the Ruby standard library for an example of GServer in action.
Any kind of application-level server can be implemented using this class. It accepts multiple simultaneous connections from clients, up to an optional maximum number. Several services (i.e. one service per TCP port) can be run simultaneously, and stopped at any time through the class method GServer.stop(port). All the threading issues are handled, saving you the effort. All events are optionally logged, but you can provide your own event handlers if you wish.
Using GServer is simple. Below we implement a simple time server, run it, query it, and shut it down. Try this code in irb:
require 'gserver' # # A server that returns the time in seconds since 1970. # class TimeServer < GServer def initialize(port=10001, *args) super(port, *args) end def serve(io) io.puts(Time.now.to_s) end end # Run the server with logging enabled (it's a separate thread). server = TimeServer.new server.audit = true # Turn logging on. server.start # *** Now point your browser to http://localhost:10001 to see it working *** # See if it's still running. GServer.in_service?(10001) # -> true server.stopped? # -> false # Shut the server down gracefully. server.shutdown # Alternatively, stop it immediately. GServer.stop(10001) # or, of course, "server.stop".
All the business of accepting connections and exception handling is taken care of. All we have to do is implement the method that actually serves the client.
As the example above shows, the way to use GServer is to subclass it to create a specific server, overriding the serve method. You can override other methods as well if you wish, perhaps to collect statistics, or emit more detailed logging.
The above methods are only called if auditing is enabled.
DEFAULT_HOST = "127.0.0.1"