singleton_domains.py

#!/bin/python
# author: Ryan Brigden

# The "brute" in brute force
# Question: Given a (large) list of words and a list of top-level domains (TLDs)
# from the Internet Assigned Numbers Authority (IANA), such as ".com" and ".net",
# find all of the possible "singleton" domains that can be registered with words
# from the word list. A singleton domain is defined as a sensical word (ie from
# the word list) whose suffix is a legitimate TLD (ie from the TLD list). You are
# given a function (is_available) that checks whether a given domain name is
# available for registration.
#
# Example:
# open(wordlist, "r").readlines() = ["surprisingly", "lifeboat", "google"]
# open(output, "r").readlines()  = ["surprising.ly", "lifebo.at"]

# Model Solution:

WORDS = "path/to/words.txt"
TLDS = "path/to/tlds.txt"
OUT = "path/to/out.txt"

# make request to determine if the domain is available for registration
# (function provided)
def is_available(domain):
    # ...
    return availability

def normalize(s):
    return s.strip().lower()

def main():

    # open and init file descriptors
    wf = open(WORDS, "r")
    tf = open(TLDS, "r")
    of = open(OUT, "w")

    # ensure the lazy-loading of file data by line (word list can be very large)
    for word in wf.xreadlines():
        for tld in tf.xreadlines():
            suffix = word[-(len(tld)):]
            prefix = word[:-(len(tld))]
            domain = "%s.%s" % (prefix, suffix)
            if suffix == tld and is_available(domain):
                of.write("%s\n" % domain)

    # cleanup
    fw.close()
    ft.close()
    fo.close()

if __name__ == "__main__":
    main()