I’ve enjoyed the series of articles and appreciate that when the author received feedback, they incorporated it and used that as fodder for subsequent blog posts. As a reader, it’s felt as if I were along for the ride.
These are all interesting optimizations (had no idea char was 2x slower to iterate over than the u8s in a &str, though that makes some sense in hindsight).
But I feel like this post is missing a full comparison of all the various algorithms and strategies tested. What was the final result? How much faster did it get?
This post is part of a series, and benchmark results were in a previous post (this is seemingly out of order, because author first implemented them, and now it’s distilling this knowledge).