Word Ladder Solver - https://en.wikipedia.org/wiki/Word_ladder

words.py

#!/usr/bin/env python3

from os.path import exists
from sys import argv, exit

import igraph as ig


def adjacent4(w1, w2):
    return (
        (w1[0] != w2[0] and w1[1] == w2[1] and w1[2] == w2[2] and w1[3] == w2[3])
        or (w1[1] != w2[1] and w1[0] == w2[0] and w1[2] == w2[2] and w1[3] == w2[3])
        or (w1[2] != w2[2] and w1[0] == w2[0] and w1[1] == w2[1] and w1[3] == w2[3])
        or (w1[3] != w2[3] and w1[0] == w2[0] and w1[1] == w2[1] and w1[2] == w2[2])
    )


def adjacent5(w1, w2):
    return (
        (
            w1[0] != w2[0]
            and w1[1] == w2[1]
            and w1[2] == w2[2]
            and w1[3] == w2[3]
            and w1[4] == w2[4]
        )
        or (
            w1[1] != w2[1]
            and w1[0] == w2[0]
            and w1[2] == w2[2]
            and w1[3] == w2[3]
            and w1[4] == w2[4]
        )
        or (
            w1[2] != w2[2]
            and w1[0] == w2[0]
            and w1[1] == w2[1]
            and w1[3] == w2[3]
            and w1[4] == w2[4]
        )
        or (
            w1[3] != w2[3]
            and w1[0] == w2[0]
            and w1[1] == w2[1]
            and w1[2] == w2[2]
            and w1[4] == w2[4]
        )
        or (
            w1[4] != w2[4]
            and w1[0] == w2[0]
            and w1[1] == w2[1]
            and w1[2] == w2[2]
            and w1[3] == w2[3]
        )
    )


words_filename = argv[1]
start = argv[2]
try:
    end = argv[3]
except IndexError:
    end = None

filename = words_filename.replace(".txt", ".gml")
if exists(filename):
    graph = ig.load(filename)
else:
    with open(words_filename) as fh:
        words = fh.read().split()

    n = len(start)
    adjacent = adjacent4 if n == 4 else adjacent5

    n = len(words)
    print(f"{n} words")
    edges = []
    for i, w1 in enumerate(words):
        for j, w2 in enumerate(words):
            if i == j:
                break
            if adjacent(w1, w2):
                edges.append((i, j))
    print(f"{len(edges)} adjacenies")
    graph = ig.Graph(n, set(edges))
    graph.vs["word"] = words
    graph.save(filename)
    print(f"graph built and saved")

start_i = graph.vs["word"].index(start)

if end is None:
    neighbors = [graph.vs["word"][i] for i in graph.neighbors(start_i)]
    print("\n".join(neighbors))
else:
    end_i = graph.vs["word"].index(end)

    paths = graph.get_all_shortest_paths(start_i, to=end_i)
    paths = list(
        tuple(graph.vs["word"][i] for i in p)
        for p in paths
    )
    paths.sort()
    for path in paths:
        print(" -> ".join(path))