#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""Simple regex creator.

Creates regex expressions from examples.
"""

from typing import Set
import sys


def lcs(str1: str, str2: str) -> Set[str]:
    """Find longest common substring.

    https://bogotobogo.com/python/python_longest_common_substring_lcs_algorithm_generalized_suffix_tree.php
    """
    str1_len = len(str1)
    str2_len = len(str2)
    counter = [[0]*(str2_len+1) for x in range(str1_len+1)]
    longest = 0
    lcs_set: Set[str] = set()

    for i in range(str1_len):
        for j in range(str2_len):
            if str1[i] == str2[j]:
                count = counter[i][j] + 1
                counter[i+1][j+1] = count
                if count > longest:
                    lcs_set = set()
                    longest = count
                    lcs_set.add(str1[i-count+1:i+1])
                elif count == longest:
                    lcs_set.add(str1[i-count+1:i+1])

    return lcs_set


def split(str1: str, str2: str) -> str:
    """Find the least restrictive regex of two strings."""
    # we only want to return * if at least one string has characters
    if str1 is str2 is '':
        return ''

    ret = lcs(str1, str2)

    try:
        sub_str = next(iter(ret))

        # if we split around 1 character matches
        # then we get a load of potentially spurious rules
        # e.g.
        # split('total', 'example') -> *a*l*
        # which is probably not right
        if len(sub_str) is 1:
            return '*'

        front_str1, back_str1 = str1.split(sub_str)
        front_str2, back_str2 = str2.split(sub_str)

        return split(front_str1, front_str2) + sub_str + split(back_str1, back_str2)

    except StopIteration as sub_str:
        # No common substrings
        return '*'


def regex(*args: str) -> str:
    """Find the least restrictive regex of n strings."""
    return args[0] if len(args) is 1 else split(args[0], regex(*args[1:]))


if __name__ == "__main__":
    print(regex(*sys.argv[1:]))