IMPLEMENTATION ARCHITECTURE
CLIENT:

1. Run a client with <the port number (PORT-S), IP address of the server 
(IP-S)> and <Client Name(ID-C), Port number (PORT-C) of the client > as command 
line argument. The client also gets its IP address (IP-C) from the local 
system.

2. The client program will need three independent control execution paths: one 
can be the main execution path and you will need to create two threads Thread1 
and Thread2 for the other two.

3. Now, the three control execution paths operate as follows.

In the main control execution path, the client program sends the information of 
the client including <ID-C, IP-C, PORT-C> to the server via UDP. The server 
will then sends the list of online-clients to this client. Once the client 
receives the list, <ID-C, IP-C, PORT-C>, the client program stores it in the
local storage (memory or disk). Then, the main execution path of the client 
program waits for an updated online-client list from the server via UDP over 
the UDP socket that it used to connect to the server above.

Thread1 is used to handle incoming chat requests from other online clients. 
Thread1 starts waiting for a TCP connection request over PORT-C to handle a 
connection request for chatting from another client. Any at time, at most one 
connection can be established. Note that, if this client is the one initiating 
a chat session to another online client, then Thread1 running on this client 
should reject all other chat requests coming from other online clients. This is 
so that a given client is involved in one chat session at a given time.

Thread2 is used to handle the user input from the keyboard. The user can see 
the list of the online-clients coming from the server. It can tehn terminate 
the client itself, or make a connection with another client or wait for 
incoming chat requests from other clients. A user can type "LIST", "CONNECT", 
or "TERMINATE", or can wait for other clients to connect.

If the command "LIST" is given, the Thread2 should show the list of other 
clients from the server on the screen. If "TERMINATE" is typed, the client 
process will be terminated. Before termination, the client should send a 
message to the server over the UDP socket to inform the server about his 
leaving the system.

"CONNECT" command should be given with the ID of the other clients. When done, 
this client will connect to the client with this ID assuming that the selected 
client is still online. While connected, one of the clients can terminate their 
connection using "BYE" command. For simplicity, one process handles at most one 
connection request.

SERVER:

1. Run the server with the port number as a command line argument.

2. A server creates a listening socket and waits for a message from the 
clients, which will look like <ID-C, IP-C, PORT-C1, PORT-C2>. The server adds 
this to its list and also sends this updated list back to the client.

3. Whenever the client list is updated, the updated list should be sent to the 
all clients over PORT-C1.