# Words by the Numbers 4, Exercise 3: how often does shuffling produce a pair as often as the real text?
import random, re
from collections import Counter
from pydoc_data.topics import topics

def tokens(text):
    return re.findall(r"[a-z]+(?:'[a-z]+)?", text.lower())

docs = [tokens(doc) for doc in topics.values()]
real = Counter()
for w in docs:
    real.update(zip(w, w[1:]))

PAIRS = [("floating", "point"), ("except", "clause"), ("can", "be"), ("is", "not")]
TRIALS = 20
beaten = Counter()
best = Counter()
for seed in range(TRIALS):
    random.seed(seed)
    shuffled = Counter()
    for w in docs:
        w = w[:]
        random.shuffle(w)
        shuffled.update(zip(w, w[1:]))
    for pair in PAIRS:
        best[pair] = max(best[pair], shuffled[pair])
        if shuffled[pair] >= real[pair]:
            beaten[pair] += 1
for pair in PAIRS:
    print(f"{' '.join(pair):15} real {real[pair]:4d} | best of {TRIALS} shuffles {best[pair]:4d} | shuffles that matched it: {beaten[pair]}")

# Output when run (Python 3.14; the documentation-text figures depend on the Python version):
# floating point  real   36 | best of 20 shuffles    2 | shuffles that matched it: 0
# except clause   real   48 | best of 20 shuffles    5 | shuffles that matched it: 0
# can be          real  193 | best of 20 shuffles    5 | shuffles that matched it: 0
# is not          real  167 | best of 20 shuffles   22 | shuffles that matched it: 0