mindblown

A brainf**k to x86 ELF compiler written in Rust meant for Linux and Windows under WSL. Codegen is made from brainf**k directly to x86 Intel assembly without any IR. Most of the optimization work is done when parsing, so an integrated interpreter can benefit from it too.

This is in a very early state. Stay tuned for more features and optimizations like the ones described in TODO.

Note: samples/patterns.py was solely used for detecting common loop patterns in classical brainf**k programs for present and future refactoring.

Table of Contents

• mindblown
  • Table of Contents
  • Prerequisites
  • Getting started
  • Debug using the integrated interpreter
  • Inspiration
  • TODO

Prerequisites

• Rust
• GCC
• WSL (for Windows users)

Getting started

1. Get the latest binary release installed on your machine ```sh

   Get the latest cargo release

$ cargo install mindblown

Build from source

$ git clone https://github.com/Noxtal/mindblown.git $ cd mindblown $ cargo build --release ```

1. Use the mindblown command to compile your brainf**k program ```sh $ mindblown

   The resulting binary will be named as the input program BUT without extension

$ chmod +x

$ ./

```

YOU ARE NOW READY TO USE MINDBLOWN! 🎉

Debug using the integrated interpreter

bash $ mindblown

Note: The integrated interpreter uses the snailquote syntax for input. Please refer to their documentation for more information.

Inspiration

• https://github.com/pretzelhammer/rust-blog/blob/master/posts/too-many-brainfuck-compilers.md#intro-to-x86 for the x86 assembly codegen
• https://github.com/pretzelhammer/brainfuckcompilers/blob/master/src/x8664/compiler.rs for examples of the x86_64 assembly codegen in Rust
• http://calmerthanyouare.org/2015/01/07/optimizing-brainfuck.html for optimizing brainf**k
• http://brainfuck.org for most programs in the samples folder

TODO

• [x] Make the interpreter take in newlines
• [ ] Add various CLI features such as choosing the output file, etc.
• [ ] Implement clearer and smarter error handling
• [ ] Optimize loop handling to Assembly
• [ ] Allow for more optimization