KnightNiwrem / rs-evalbot

IRC code evaluator

Rust IRC evaluator

Like rust-playpen, but in Rust.

This repository contains:

• evalbot/: the bot itself
• evaluators/: some glue code for C# (Mono) and Python REPLs
• run/: example configuration files and a script to set up a sandbox in Arch

Simple usage

1. Compile the bot.

   $ pushd evalbot; cargo build --release; popd
   

2. Set up the sandbox.

   $ pushd run; sudo ./init.sh; popd
   

   N.B. init.sh is written for a system with pacman.

3. Edit the configuration files

   $ pushd run
   $ cp evalbot.toml.in evalbot.toml; cp evalbot.irc.toml.in evalbot.irc.toml
   $ $EDITOR evalbot.toml
   $ $EDITOR evalbot.irc.toml
   $ popd
   

4. Run the bot

   $ pushd run
   $ cp ../evalbot/target/release/evalbot .
   $ ./evalbot
   

5. Try it out!

   <$you> rs>println!("Hello!");
   <$bot> >Hello!
   <$bot> >()
   

Advanced usage

This bot depends on playpen.

If you wish to use some other means of sandboxing evaluated code (or not sandbox it at all), you can edit the code in evalbot/src/playpen.rs. However, note that the bot depends on playpen to enforce the timeout for non-daemon evaluators (i.e. persistent = false). Check the code if you're still unsure.

The chroot sandbox just needs to be able to run whatever you configure it to run (in evalbot.toml).

"Persistent" evaluator protocol

The bot will send, for each request, via standard input:

Field	Type	Description
Timeout	Int32	Timeout in milliseconds, or 0 for none
Context key length	Int32	Context key in bytes
Code length	Int32	Code length in bytes
Context key	UTF-8 string	Key of the context to use
Code	UTF-8 string	The code to evaluate

The bot expects, for each request, from standard output:

Field	Type	Description
Response length	Int32	Response length in bytes
Response	UTF-8 string	The response

Note that an evaluator will be killed by the bot if it doesn't respond within 1.5 * timeout seconds.