wordle_env/words.py

import os
import urllib.request

_COMPLETE_VOCABULARY_URL = "https://gist.githubusercontent.com/scholtes/94f3c0303ba6a7768b47583aff36654d/raw/d9cddf5e16140df9e14f19c2de76a0ef36fd2748/wordle-Ta.txt"
_TARGET_VOCABULARY_URL = "https://gist.githubusercontent.com/scholtes/94f3c0303ba6a7768b47583aff36654d/raw/d9cddf5e16140df9e14f19c2de76a0ef36fd2748/wordle-La.txt"
_DOWNLOADS_DIR = '.'
_COMPLETE_VOCABULARY_FILENAME = "complete_vocabulary.txt"
_TARGET_VOCABULARY_FILENAME = "target_vocabulary.txt"

def _retrieve_vocabulary(url, filename, dir):
    vocabulary_file = os.path.join(dir, filename)
    
    # Download the file if it does not exist
    if not os.path.isfile(vocabulary_file):
        urllib.request.urlretrieve(url, vocabulary_file)

    with open(vocabulary_file) as file:
        return  [line.rstrip().upper() for line in file]

target_vocabulary = _retrieve_vocabulary(_TARGET_VOCABULARY_URL, _TARGET_VOCABULARY_FILENAME, _DOWNLOADS_DIR )
complete_vocabulary = _retrieve_vocabulary(_COMPLETE_VOCABULARY_URL, _COMPLETE_VOCABULARY_FILENAME, _DOWNLOADS_DIR ) + target_vocabulary