Wordle
View on GitHub and PyPI.
Updates
Version 2.2.0: Individual Guessing!
This package can now be properly used to power the backend of a Wordle game. Initially, the .run()
method had to be used to play the game. It would open a console, request guesses as inputs, and print out responses. That's not a viable solution to power the backend of a package user's application or game.
See a demo!
The Wordle
class has two new submethods: .send_guess()
and .reset_guesses()
.
Instantiate an object with the answer and game options (dictionary, etc.). Then, send individual guesses and the method will return a response.
import wordle
game = wordle.Wordle(word = 'grape', real_words = True)
response = game.send_guess('adieu')
response = game.send_guess('yabadabadoo')
response = game.send_guess('seven')
More information on how to use this new feature is below.
Version 2.0.0
A new function to get a random answer based on the day. You can deploy this Wordle and have it give you a random answer automatically every day! Instead of manually changing the answer, the program (can) automatically change its answer every day.
Use wordle.random_answer(daily = True)
(Update: daily
argument is now optional with default value True
). You can instantiate a game using this parameter:
import wordle
wordle.Wordle(word = wordle.random_answer(), real_words = True).run()
Or, you can access the random word directly in instantiation (new and preferred):
import wordle
wordle.Wordle(random_daily = True, real_words = True).run()
All arguments are now optional. By default, word = 'hello'
, real_words = True
, and random_daily = False
. If you set random_daily = True
, word
is overwritten with a daily word. If you provide a word
and set random_daily = True
, the word
you chose will be overwritten. So, try to use either word
or random_daily
.
If this game is run in the same day, the answer word
will be the same. Tomorrow, however, it will change. I did this using UNIX. The dictionary is shuffled and an iterator is calculated using multiples of 86400 (seconds in a day). The program then gets a word from the randomized dictionary.jumbled_words
list based on the iterator.
The behavior is that the program will have the same answer intraday but will automatically run with a new unique answer the next day. By standardizing this using UNIX time I avoided the need for a central database of "used" words. To reset it all, dictionary.jumbled_words
has to be recalculated using random.sample(dictionary.words, len(dictionary.words))
.
This is a much better way to deploy to the cloud as the game is completely self-sufficient!
Version 1.9.0
Completely redesigned the Wordle
class as it was totally inefficient. It used to take the parameter self.real_words
, given on object instantiation, and then run in two separate ways. The logic worked in this way:
class Wordle:
def run(self):
if self.real_words == True:
run_with_logic_that_words_must_be_checked()
elif self.real_words == False:
run_with_logic_that_words_must_not_be_checked()
This was very inefficient because fixes and upgrades had to be made to both aspects of the Wordle.run()
function. Initially it was necessary because the logic for checking against the dictionary was unique, but now, I'm hosting a dictionary of thousands of five-letter words here. So, I rewrote the class to have this structure:
class Wordle:
def run(self):
run_with_interspersed_realWords_conditions()
Which works much better. I introduced a failed_dictionary_test
boolean which resets every time a new attempt is entered but allows the program to alert non-real words only if self.real_words == True AND guess.lower() not in dictionary.words
. It's clean, faster, and more efficient!
Version 1.8.0
An issue was brought up on GitHub. If the answer was "games"
for example, and you guessed "trees"
, the program would capitalize both of the letters "e"
in your guess. It would behave in this way:
# Answer = 'games'
Attempt 1 >>> trees
t r E E *S*
I corrected the error with a functional list(self.word)
that resets itself every attempt. Now the program behaves properly, in the way we'd expect:
# Answer = 'games'
Attempt 1 >>> trees
t r E e *S*
Documentation
Wordle is super fun and popular game. Unfortunately, it's new and nonstandard, meaning the backend technology is not prevalent and recreatable online. There are a couple web-based customizable Wordle tools, which work very well, but they're front-end only (you can't clone, copy, modify, or edit the back-end to deploy it to a website or run your own game). That said, the logic behind Wordle is quite simple which is why it's now a Python library, enabling all the functions of an open-sourced game.
Currently, the module is logic based and runs in a shell (as of v2.2.0, the new individual guess methods enable Wordle to be deployed practically anywhere). I'd love it if this project was forked into a GUI-based application, possibly using pygame
, allowing users to take this a step further and deploy a version with a user-interface. As it stands, the best way I can think of to deploy Wordle to the web is by using an embedded Python console like Trinket and putting it in 'run-only' mode so users can't see the source code (where the answer is). See an example of this in real deployment here.
In any case, usage of this module is wickedly simple.
import wordle
game = wordle.Wordle(word = 'HELLO', real_words = False)
game.run()
wordle.Wordle(word = 'hello', real_words = True).run()
Instantiate a game
object using game = wordle.Wordle()
. The two positional requirements of the Wordle
class are word
and real_words
. word
is the answer to the game. If the object is instantiated with real_words = True
, word
must be a real, five-letter word or an exception will be raised, and you'll have to change your instantiation to reflect a real word. real_words
is a boolean. If it's True
, the package will check user's (in-game) guesses against a comprehensive inbuilt database of thousands of five-letter words. If it's False
, any guess (real or not) will be accepted. In summary, real_words
applies both to the answer in the code's instantiation and in the user's in-game guesses.
In the game of Wordle, part of the challenge is that guesses must be real words. This prevents users from guessing "aeiou"
as their first attempt, for example. That's it's relevant and important for you to tell the game
object whether it should check for real words or not.
Note: In version 1.5.10, the issue disallowing a lower-case word
has been patched. word = 'hello'
and word = 'HELLO'
are both acceptable in the game initialization.
To view the docstring (explainer) of a particular function, use wordle.Wordle.run.__doc__
where run
can be repalced with any other function. For example, running print(wordle.Wordle.run.__doc__)
will return: Run the game. Depends on bool real_words from instantiation.
Depending on your IDE, hovering over a function in the editor will show the function's docstring (works in VS Code).
import wordle
if __name__ == "__main__":
print(wordle.Wordle.__doc__)
print(wordle.Wordle.run.__doc__)
Individual Guessing
This new feature was introduced in the Version 2.2.0 update at the top of the read-me. Here's some more usage information.
Here's some example usage.
Instantiate an object with the answer and game options (dictionary, etc.). Then, send individual guesses and the method will return a response.
import wordle
game = wordle.Wordle(word = 'grape', real_words = True)
response = game.send_guess('adieu')
response = game.send_guess('yabadabadoo')
response = game.send_guess('seven')
The .send_guess()
method has two parameters, one of which is optional. guess
is a string and is necessary. log_guess
is a boolean and is True
by default. Log guessing is simply the program's way of tracking how many times a valid guess has been sent to the object. It will only allow 6 valid guesses to be sent to it before it starts returning "Out of guesses."
automatically. To send a guess and return a response without adding to the guesses log, use log_guess = False
.