I implemented a variation long ago to parse regular expressions (regex) [0]. Albeit I did some pre parsing for things like character classes and lookarounds.

[0]: https://github.com/nitely/nim-regex/blob/5e447c329ce826deb61...

edit: added permalink