I feel like beginner programmers should experiment with the following concepts in order.
1. File I/O to plaintext files. fread/fwrite, or whatever equivalent is in your programming language (Python fd.read() or whatever) is your first step into I/O. Especially because you can open the file and read it yourself as a human.
2. Piped I/O using Unix Pipes and/or Fifos. Pipes (./program1 | ./program2) are by far the easiest interprocess communication mechanism. They're not very flexible, but they're obvious to understand once you have mastered simple File I/O.
3. TCP Sockets are the natural evolution from pipes. Your first TCP Server will not be socket/bind/accept, since this is a bit complicated. Instead, you should write nc -l (port number) | ./myprogram, where netcat does the heavy lifting for you and creates a listening server for ./myprogram. You then have the ability to write ./myclient (which communicates to netcat, and therefore to "myprogram").
4. You finalize this by writing the server side without the netcat training wheels, to learn the socket/bind/accept dance.
I disagree that TCP are an evolution of pipes. Pipes are unidirectional and have 2 file descriptors, which is way less confusing than TCPs behaviour with respect to EOF, shutdown and write()/read() error behaviour.
A socket API that gave you two unidirectional fd's for a TCP socket would have been a massive improvement.
With the POSIX API, you can duplicate the socket file descriptor, but calling shutdown() to close one direction of a file descriptor will actually close that direction on all file descriptors... unfortunate.
TCP connections also have 2 file descriptors, they are just on different computers. It seems quite natural to me to think of a socket as two pipes joined together.
As with so many things in computing the abstraction leaks a bit when you get deeper into it but as a first intuition for beginners I think you could do way worse than "a socket is a bidirectional pipe to another computer".
I'm intrigued by your idea of making 2 file descriptors for each socket btw. Why would it be such an improvement in your opinion?
> I disagree that TCP are an evolution of pipes. Pipes are unidirectional and have 2 file descriptors, which is way less confusing than TCPs behaviour with respect to EOF, shutdown and write()/read() error behaviour.
TCP is complicated. But "stream based sockets" aren't "just" TCP (ex: Unix domain sockets fixes all of the "glitches" of TCP).
The important bit of "sockets", in general, is the concept of:
1. "Accepting" file descriptor (creates new file descriptors whenever someone is "connecting" to you).
2. Multiplexing and "juggling" those descriptors.
TCP fits this model, though with some low level quirks. Unix Domain Sockets are arguably the place to start, but most beginner programmers wouldn't find that useful (they really want internet connections).
-------------
Its difficult to make a "server" off of one pipe. Maybe two pipes (first pipe accepts requests and divvys out new pipes to work off of). But no matter how you do it, its wonky.
A proper server needs that "accept" abstraction, which is completely missing from pipes.
> A proper server needs that "accept" abstraction, which is completely missing from pipes.
You can receive file descriptors over a UNIX socket, but yeah. That said I don't see why a file descriptor is a good abstraction for a listening socket in the first place when you don't read() or write() to it.
In two-fds tcp, FIN/ACK, RST, timeouts and desire to stay as non-blocking as possible would still be a thing, so we’d have all the same functions and behaviors, only SHUT_RD|WR parameters would go away.
Pipes guarantee delivery if both sides agree, tcp doesn’t (but it tries hard asynchronously), that’s where the difference starts.
If I want to implement some rudimentary communications protocol between two machines in Rust, just for learning, is this where I should begin?
Edit: perhaps not quite. Seems like this lives “on top of” TCP or UDP. I should probably read through this then look for something like a “how to implement ICMP” guide.
Interesting. If you don't mind me asking, what kinds of things have you made talk to each other that weren't originally designed to talk to each other?
Usually make retrocomupters talk to modern networks or modern computers talk to retro networks.
Admittedly I haven't programmed much in a while outside work, but have a number of backlogged projects including an H.320 (ISDN videoconferencing spec) to discord bridge.
Most of my ideas, are useless and kinda dumb, but fun.
Sure. But keep in mind that some people readers here on HN are still beginners when it comes to TCP/IP. Maybe high school students, maybe they work in another field than tech, maybe they work in tech but with design, legal, HR etc etc. For those readers it could potentially be a useful and interesting resource.
1. File I/O to plaintext files. fread/fwrite, or whatever equivalent is in your programming language (Python fd.read() or whatever) is your first step into I/O. Especially because you can open the file and read it yourself as a human.
2. Piped I/O using Unix Pipes and/or Fifos. Pipes (./program1 | ./program2) are by far the easiest interprocess communication mechanism. They're not very flexible, but they're obvious to understand once you have mastered simple File I/O.
3. TCP Sockets are the natural evolution from pipes. Your first TCP Server will not be socket/bind/accept, since this is a bit complicated. Instead, you should write nc -l (port number) | ./myprogram, where netcat does the heavy lifting for you and creates a listening server for ./myprogram. You then have the ability to write ./myclient (which communicates to netcat, and therefore to "myprogram").
4. You finalize this by writing the server side without the netcat training wheels, to learn the socket/bind/accept dance.