A brainfuck interpreter.
It takes one or multiple files this way: bf-interpreter file-1.bf ... file-n.bf.
Each file (program) has a separate "runtime", which is composed of an array of 30 000 bytes (u8 in Rust), a data pointer (usize), a program pointer (usize), and a vector of tokens (the program instructions). As explained below, I suggest compiling it with release mode as some programs rely on integer overflow. Considering the program pointer is usize, this may cause problems with programs with a very large number of instructions.
There's 3 programs in the tests: helloworld.bf, mandelbrot.bf and test.bf. mandelbrot.bf and test.bf both depends on integer overflow to run correctly. It caused me problems during testing as I ran the interpreter with cargo run file.bf, which was using debug mode, which panics on overflow.
The helloworld.bf and mandelbrot.bf programs came from https://github.com/Overv/bf.
The young.bf program comes from https://esolangs.org/wiki/FBP.
The cellwidth.bf program comes from https://esolangs.org/wiki/Brainfuck.