There is, I think, one thing missing: a comprehensive test suite that checks all the possible edge cases.
Edit: as you can see below, the benchmark also checks for correctness so you can disregard this.
Edit: as you can see below, the benchmark also checks for correctness so you can disregard this.
[1] https://github.com/phoboslab/qoi/blob/master/qoifuzz.c
[2] https://qoiformat.org/benchmark/
[3] https://github.com/phoboslab/qoi/blob/master/qoibench.c#L440...