# Words by the Numbers 6, Exercise 1
import re

FUNCTION_WORDS = ["the", "of", "and", "to", "a", "in", "it", "but", "so", "we"]

sample_a = "the road to the harbour was long and the wind was cold but the men walked on and the night fell over the sea"      # invented
sample_b = "it was a long walk to a harbour in a cold wind so we rested in a barn and ate a loaf of bread in the dark"       # invented
mystery = "so it was a long road to a town in the north but we came in a wagon and slept in a shed"                           # invented

def profile(text):
    words = re.findall(r"[a-z]+", text.lower())
    return [100 * words.count(f) / len(words) for f in FUNCTION_WORDS], len(words)

pa, na = profile(sample_a)
pb, nb = profile(sample_b)
pm, nm = profile(mystery)
print("words:", na, nb, nm)
print(f"{'word':6} {'A':>6} {'B':>6} {'mystery':>8}   (uses per 100 words)")
for f, x, y, z in zip(FUNCTION_WORDS, pa, pb, pm):
    print(f"{f:6} {x:6.1f} {y:6.1f} {z:8.1f}")
dist_a = sum(abs(x - z) for x, z in zip(pa, pm))
dist_b = sum(abs(y - z) for y, z in zip(pb, pm))
print(f"distance to A: {dist_a:.1f}   distance to B: {dist_b:.1f}")
print("closer to", "A" if dist_a < dist_b else "B")

# Output when run (Python 3.14; Federalist text from the Project Gutenberg plain-text file):
# words: 24 27 23
# word        A      B  mystery   (uses per 100 words)
# the      25.0    3.7      4.3
# of        0.0    3.7      0.0
# and       8.3    3.7      4.3
# to        4.2    3.7      4.3
# a         0.0   18.5     17.4
# in        0.0   11.1     13.0
# it        0.0    3.7      4.3
# but       4.2    0.0      4.3
# so        0.0    3.7      4.3
# we        0.0    3.7      4.3
# distance to A: 68.5   distance to B: 15.0
# closer to B