How can I optimize this Python code to generate all words with word-distance 1?

Question

Profiling shows this is the slowest segment of my code for a little word game I wrote:

def distance(word1, word2):
    difference = 0
    for i in range(len(word         


        

Answer 1

Everyone else focused just on explicit distance-calculation without doing anything about constructing the distance-1 candidates. You can improve by using a well-known data-structure called a Trie to merge the implicit distance-calculation with the task of generating all distance-1 neighbor words. A Trie is a linked-list where each node stands for a letter, and the 'next' field is a dict with up to 26 entries, pointing to the next node.

Here's the pseudocode: walk the Trie iteratively for your given word; at each node add all distance-0 and distance-1 neighbors to the results; keep a counter of distance and decrement it. You don't need recursion, just a lookup function which takes an extra distance_so_far integer argument.

A minor tradeoff of extra speed for O(N) space increase can be gotten by building separate Tries for length-3, length-4, length-5 etc. words.