ircb is an awk irc bot. It accepts raw data from IRC via stdin, and outputs IRC commands to stdout. It is completely nawk-compatible; you will find no GNU/awk-exclusive functionality.
Sample invocation using a FIFO in might look like:
netcat my.irc.network 6667 <in | awk -f ircb.awk >in
Another invocation that uses the bash /dev/tcp pseudofile extension:
exec 3<>/dev/tcp/my.irc.network/6667; awk -f ircb.awk <&3 >&3
This setup is extremely versatile; run the data streams through as many pipes and sed and grep commands as you want.
ircb supports a host of modules, most of which you don't need. In fact, maintainers are actively encouraged to dispose of any modules they do not need.
modules/boot.awk is considered the only "necessary" module. Anything else is optional.
Including modules can be done using awk's -f option:
netcat my.irc.network 6667 <in | awk -f ircb.awk -f modules/boot.awk -f modules/tell.awk >in
Or:
exec 3<>/dev/tcp/my.irc.network/6667; tee irc/log <&3 | awk -f ircb.awk >&3
For information on module usage, check the actual .awk file.
How not to configure ircb: open each individual module file and modify the variables in the BEGIN block.
How to configure ircb: write a config.awk with a BEGIN block where you declare your configuration variables. Include this after every other module using -f
Write a lang.awk file where you overwrite response templates. Include this after all other modules. For example:
# es.awk
BEGIN {
dbinterface_Template["conflict"] = "[%s] Error fatal: opciones conflictivas `%s` y `%s` especificado."
}
irc logger: /usr/bin/tee
configuration variables are denoted with ###.
ircd weirdness is documented using ##