#design-pattern #game #buffer #game-state #double #immutability #programming

bufferbuffer

This is an implementation of the Double Buffer design pattern from 'Game Programming Patterns' by Robert Nystrom

2 unstable releases

0.2.0 Aug 10, 2024
0.1.0 Aug 9, 2024

#628 in Game dev

MIT license

7KB
65 lines

bufferbuffer

This is an implementation of the Double Buffer design pattern from 'Game Programming Patterns' by Robert Nystrom.

In a simulation, you often have to do a lot of processing to prepare the next "frame", but if you're iterating through the current-state data while mutating it, things can slip. The Double Buffer design pattern solves this by keeping two copies of the simulation state (or any variable): the "current" (or previous) state which is immutable, and the "next" (or future) state which is being prepared. When a turn of the simulation is completed, you simply switch the buffers.

Unlike other implementations on crates.io, this one wraps both buffers in std::cell::RefCell so that it is possible to borrow one buffer as mutable at the same time the other is borrowed as immutable -- a typical use case is to iterate over objects in the world (current state) and write updated versions of them to the next state.

Usage

For a simple variable:

let mut my_double_buf: DoubleBuffer<i32> = DoubleBuffer::new(0,0);
*my_double_buf.next() += 10;
my_double_buf.switch();
assert_eq!(*my_double_buf.current(), 10);

For a vector:

let mut my_double_buf: DoubleBuffer<Vec<i32>> = DoubleBuffer::new( vec!(2,4,6), Vec::new());
for number in my_double_buf.current().iter() {
    my_double_buf.next().push(*number + 1);
}
my_double_buf.switch();
assert_eq!(*my_double_buf.current(), vec!(3,5,7));

No runtime deps