Generate cool names for things that are hard to name from a wordlist. Entries in the wordlist should start with star and a space, one per line. Other notes will be ignored

namegenerator.py

#!/usr/bin/env python
import random
import re
import argparse
parser = argparse.ArgumentParser(description="Generate names from wordlist")
parser.add_argument('--wordlist', default='wordlist', help='wordlist file to read')
parser.add_argument('--iterations', default=100, type=int, help='how many words to generate?')
parser.add_argument('--words', default=3, type=int, help='how many words to fuse together')
parser.add_argument('--min', default=2, type=int, help='min letters to pick from word')
parser.add_argument('--max', default=4, type=int, help='max letters to pick from word')
parser.add_argument('--no-offset', dest='offset', action='store_false', help='offset pick from start of word?')
parser.add_argument('--cutshut', dest='cutshut', action='store_true', help='cut n shut mode')
parser.set_defaults(offset=True)
parser.set_defaults(cutshut=False)
args = parser.parse_args()

# for reversed min/max just use max
if args.min > args.max:
  args.min = args.max

filename = args.wordlist
c = 0
wordlist=[]
# read the file line by line
with open(filename, 'r') as f:
  for line in f:
    if line.startswith("*"):
      word = line.strip()
      # eliminate leading spaces and stars
      wordlist.append(re.sub('\*\s+', '', word))
      c = c+1
print "Loaded {c} words!".format(c=c)

# pick a random element in wordlist
names = {}
for iteration in range(args.iterations):
  name = ""
  source = []
  # always cram 3 words together
  for i in range(args.words): 

    # pick a random word from the wordlist, make sure its not a dup
    # use a loop-and-a-half:  http://stackoverflow.com/questions/1662161/is-there-a-do-until-in-python
    while True:
      r = random.randint(0, len(wordlist)-1)
      random_word = wordlist[r] 
      if random_word not in source:
        break

    # store the source word so we know where we got our final name from
    source.append(random_word)

    # kill all spaces
    unspaced = re.sub('\s+', '', random_word)

    # pick random group of letter of varying length.  Can only be as long
    # as the word we are looking at
    l = min(random.randint(args.min, args.max), len(unspaced))
    
    # pick a random offset for the above group
    if args.cutshut:
      # pick from start, then middle, then end as we build the word
      if i==0:
        # first word
        o=0
      elif i==args.words-1:
        # last word
        o=len(unspaced)-l
      else:
        # middle word
        o=random.randint(1, max(len(unspaced)-l,2)) 
    else:
    # 'normal' offset mode
      if args.offset:
        o = random.randint(0, len(unspaced)-l)
      else:
        o = 0

    sect  = unspaced[o:o+l]
    name = name+sect

  # filter away words with no vowels ;-)
  if re.match('.*[aeiou].*', name):
    # use name as a key to remove duplicates and then save the stringified 
    # source words
    names[name] = str(source)

for name in names:
  print name + " <---- " + names[name]