C++ Client-Server Architechture with Sockets on Linux

uses a client-server architecture to implement the tokenizer system. The client and server will communicate using sockets. Use the same struct’s defined for the previous assignments to handle strings and tokens. As in Project 2, the parent program (on the client machine) will handle the string input and the tokenizer, both in threads as before and communicating with queues. Use a third thread handle sending the tokenStruct’s to the server; this will require a new queue. The parent must handle two command line parameters that indicate the server’s name (or address) and the port number, in the form –s server and –p portnumber. Make sure that the socket connection is successful before spawning either thread; if not, report an error and quit. Otherwise, the server runs “forever” – that is, when the quit command is received, only the worker thread quits – the server continues to listen for connections. The server program will take one command line parameter to the server to indicate what port number to use, in the form –p portnumber. As with the client, make sure that the socket initialization is successful before spawning the threads. If it is, display the port number to the screen; if not, report an error and quit. Note that the port number may be in use; only port numbers above 1024 are allowed; you can, if you wish, add a port stepper – a loop that increases the port number until a successful one is found. Also note that the server must be started in the background (add an & to the end of the command), before any client is started, and the server must be killed manually with the kill -9 %pid command. Otherwise, the server runs “forever” – that is, when the quit command is received, only the worker thread quits – the server continues to listen for connections. The server will use three threads – one to handle incoming messages, started after the accept( ) call; the other two threads are the token classifier/printer and the token counter. These threads will communicate using shared memory queues.