I am having a problem with my program: it does as required but largest_blanagram() is too slow. I'm not sure whether the best way to speed it up is (1) try to speed up blanagram(), which is being called a lot (2) figure out a way to call blanagram() less. I'm not sure how to go about either of these things particularly.
Things I've been thinking: is there a way to use another dictionary more efficiently, such as using map()?
# CW Q2 blanagram.py
import string
from Q1_anagram import readfile, make_anagram_dict # change
def blanagram(word, anagramdict): # split up into smaller functions?
"""
"""
word = sorted(word) # sort alphabetically
alphabet = list(string.ascii_lowercase)
index = 0
blanagrams = []
for char in word:
original_char = char # save original char, so can revert back to it later
for letter in alphabet:
word[index] = letter
word_copy = word # as order messes up when order alphabetically, so save current state and revert back to it after the iteration
word = ''.join(sorted(word)) # list -> string to be able to check dict and sort alphabetically
if word in anagramdict.keys() and anagramdict[word] not in blanagrams:
for anagramdict[word] in anagramdict[word]:
blanagrams.append(anagramdict[word])# change name?
word = word_copy
# next character in word
word = list(word) # string -> list in order to index
word[index] = original_char # set back to original char, as only 1 char can be different from original word
index += 1 # to change following char
return blanagrams # get rid of anagrams?
def largest_blanagram(words, word_length): # greatest number of blanagram variants
"""
"""
blanagrams = {}
length = 0
for word in words:
if len(word) == word_length:
variants = blanagram(word, make_anagram_dict(words))
#print(variants)
if len(variants) > length:
blanagrams[word] = variants
length = len(variants)
return blanagrams
if __name__ == "__main__":
# print(blanagram('ghost', make_anagram_dict(readfile('words.txt'))))
import sys
print(largest_blanagram(readfile('words.txt'), 13)) #sys.argv[1]
Imports -
def readfile(filename):
"""
"""
file = open(filename,'r').readlines() # look into WITH open file
words = []
for line in file:
line = line.lower().split() # orang and Orang? unique? or is asteer another answer
for word in line:
words.append(word)
return words
def make_anagram_dict(words):
"""
"""
assert len(words) != 0
anagram_dict = {}
for word in words:
alpha = ''.join(sorted(word)) # Order word alphabetically
# Key is alphabetically ordered word, value is list of anagrams of the key
anagram_dict.setdefault(alpha, []).append(word) # So not appending NoneType (as list.append returns None)
return anagram_dict
And here is a pastebin equivalent (pastebin has file size limit) link to the words.txt file that I use.
