Avoid dependency on SOCAT

[CompuRoot]

gawk already can talk to a network without assistance from external utilities by utilizing special type of files.

From man:

The following special filenames may be used with the |& coprocess operator for creating TCP/IP network connections:
       /inet/tcp/lport/rhost/rport
       /inet4/tcp/lport/rhost/rport
       /inet6/tcp/lport/rhost/rport

So listening loop can be much simpler, as a prove of concept, below is the simplest web server in plain gawk

#!/bin/sh

gawk -v ListeningPort=12345 'BEGIN {
  content="HTTP/1.1 200 OK\r\n"
  content=content "Content-Length: 12\r\n"
  content=content "Content-Type: text/plain; charset=utf-8\r\n\r\n"
  content=content "Hello World!"

  socket = "/inet/tcp/" ListeningPort "/0/0";
  while (1) {
    printf content |& socket;
  }
}'

BTW, the same can be done with bash and that the reason why on production servers bash, gawk and so on has 700 permission and root:root ownership, to make hackers live harder because the first thing they do if they get access to user's accounts, - utilizing network capability of bash/gawk to create reverse shell.

[crossbowerbt]

Thanks for the comment. I can make a different version of the script for Linux, but at the moment my targets are also BSD systems.

SOCAT is a nice compromise, since it allows the use of standard awk (not gawk) and generic shells (like KSH)