BlueHensCTF2022 - Wordles with Dads Solve Scripts

wordle_dad_solve.py

from pwn import *
import requests
import pickle
from bs4 import BeautifulSoup

url = "https://icanhazdadjoke.com/"
suburl = "search?term=&page="

jokes = []

try:
    with open('jokes.pickle', 'rb') as handle:
        jokes = pickle.load(handle)

except FileNotFoundError:
    for i in range(1, 34):

        r = requests.get(url + suburl + str(i))
        bs = BeautifulSoup(r.content, "html.parser")
        results = bs.find_all('pre')

        for result in results:
            jokes.append(str(result).replace('<pre style="background-color:inherit; white-space: pre-wrap">', "").replace("</pre>", "").strip())

    with open('jokes.pickle', 'wb') as handle:
        pickle.dump(jokes, handle, protocol=pickle.HIGHEST_PROTOCOL)

# print(jokes)

joke_wordles = []

try:
    with open('joke_wordles.pickle', 'rb') as handle:
        joke_wordles = pickle.load(handle)

except FileNotFoundError:
    for joke in jokes:
        tmp = ""
        for char in joke:
            if char.isalpha():
                tmp += char.upper()
            elif char == "ñ":
                tmp += "N"
        joke_wordles.append(tmp)

    with open('joke_wordles.pickle', 'wb') as handle:
        pickle.dump(joke_wordles, handle, protocol=pickle.HIGHEST_PROTOCOL)

# print(len(joke_wordles))

p = remote("0.cloud.chals.io", 33282)

for i in range(10):
    p.recvuntil(b"Your joke is ")
    ans_len = int(p.recvuntil(b" ").decode("utf-8"))
    print(ans_len)

    len_filtered = filter(lambda x: len(x)==ans_len, joke_wordles)
    possibilities = list(len_filtered)

    while(1):
        total_possibilities = len(possibilities)
        print(f"Possibilities left: {total_possibilities}")

        p.recvuntil(b"Guess? >")
        maybe = possibilities[0]
        p.sendline(maybe.encode())

        p.recvline()
        data = p.recvline().decode('utf-8')

        corrects = []

        if("one down" in data.lower()):
            break
        else:
            try:
                tmp = data[13:].split("],")[0]
                corrects = list(map(int, tmp.strip().split(",")))
            except:
                pass

        # print(corrects)

        pop_count = 0

        for i in range(total_possibilities):
            for j in range(ans_len):
                if (j in corrects and maybe[j] != possibilities[i-pop_count][j]) or (j not in corrects and maybe[j] == possibilities[i-pop_count][j]):
                    possibilities.pop(i-pop_count)
                    pop_count+=1
                    break

p.interactive()

wordle_dad_solve_manual.py

import requests
import pickle
from bs4 import BeautifulSoup

url = "https://icanhazdadjoke.com/"
suburl = "search?term=&page="

jokes = []

try:
    with open('jokes.pickle', 'rb') as handle:
        jokes = pickle.load(handle)

except FileNotFoundError:
    for i in range(1, 34):

        r = requests.get(url + suburl + str(i))
        bs = BeautifulSoup(r.content, "html.parser")
        results = bs.find_all('pre')

        for result in results:
            jokes.append(str(result).replace('<pre style="background-color:inherit; white-space: pre-wrap">', "").replace("</pre>", "").strip())

    with open('jokes.pickle', 'wb') as handle:
        pickle.dump(jokes, handle, protocol=pickle.HIGHEST_PROTOCOL)

# print(jokes)

joke_wordles = []

try:
    with open('joke_wordles.pickle', 'rb') as handle:
        joke_wordles = pickle.load(handle)

except FileNotFoundError:
    for joke in jokes:
        tmp = ""
        for char in joke:
            if char.isalpha():
                tmp += char.upper()
            elif char == "ñ":
                tmp += "N"
        joke_wordles.append(tmp)

    with open('joke_wordles.pickle', 'wb') as handle:
        pickle.dump(joke_wordles, handle, protocol=pickle.HIGHEST_PROTOCOL)

# print(len(joke_wordles))


easy = False
mode = str(input("Easy mode? "))

if mode and mode[0].lower() == "y":
    easy = True

while(1):
    ans_len = int(input("Length: ")) 

    if easy:
        first_two_char = str(input("First two characters: ")).upper().strip()

        f2c_filtered = filter(lambda x: x[:2]==first_two_char, joke_wordles)
        len_filtered = filter(lambda x: len(x)==ans_len, list(f2c_filtered))

    else:
        len_filtered = filter(lambda x: len(x)==ans_len, joke_wordles)

    possibilities = list(len_filtered)

    while(1):
        # print(possibilities)

        total_possibilities = len(possibilities)
        print(f"Possibilities left: {total_possibilities}")

        maybe = possibilities[0]
        print(maybe)

        if total_possibilities==1:
            break

        found = str(input("found? "))
        if found and found[0].lower() == "y":
            break
        
        try:
            corrects = list(map(int, str(input("Corrects: ")).strip().split(",")))
        except:
            corrects = []

        print(corrects)

        pop_count = 0

        for i in range(total_possibilities):
            for j in range(ans_len):
                if (j in corrects and maybe[j] != possibilities[i-pop_count][j]) or (j not in corrects and maybe[j] == possibilities[i-pop_count][j]):
                    possibilities.pop(i-pop_count)
                    pop_count+=1
                    break

manual is my earlier script which I try to input by hand, but I wasn't fast enough hence why I made the solve with pwntools (could've used sockets as well)

process:
webscraping all the jokes -> pickle the scraped jokes as cache as to not redo the webscraping -> solve it based on length with filters and also go through the correct/wrong character indexes and filter them out