findregex

.findregex(
   winners: set, losers: set, calls: int = 1000, restarts: int = 3, max_len: int = 5
)

---

The entrypoint into generating regex expressions from lists You give the function a set of things you want an expression to match (winners), and a set of things you want it not to match (losers), and it generates an expression that does so.

calls sets the max number of iterations of the algorithm we do before we give up and return what we have restarts sets how many times we restart to algorithm in order to shake things up in case we've gone down a sub-optimal path max_len sets the maximum length of induvidual chunks that we OR together

The algorithm runs for calls * restarts times, so tread lightly when setting those values. Increasing them causes the algorithm to run longer and find a shorter regex that meets the specifications, but it will also be significantly slower.

Wrap this call in raw() to use it as an EZRegex expression

Credit for this algorithm goes to Peter Norvig and Stefan Pochmann. See: https://nbviewer.jupyter.org/url/norvig.com/ipython/xkcd1313-part2.ipynb