Orsdxy

As an exercise, I wrote a program to decrypt a PDF file. The only thing that is known about the encryption password is that it is a single English word (all capital or lowercase).

The program uses a dictionary file containing over 45000 words to unlock the file. It looks like this:

dictionary.txt

AARHUS

AARON

ABABA

ABACK

ABAFT

ABANDON

ABANDONED

ABANDONING

ABANDONMENT

ABANDONS

ABASE

ABASED

ABASEMENT

ABASEMENTS

ABASES

ABASH

ABASHED

ABASHES

⋮

ZEROTH

ZEST

ZEUS

ZIEGFELD

ZIEGFELDS

ZIEGLER

ZIGGY

ZIGZAG

ZILLIONS

ZIMMERMAN

ZINC

ZION

ZIONISM

ZIONIST

ZIONISTS

ZIONS

ZODIAC

ZOE

ZOMBA

ZONAL

ZONALLY

ZONE

ZONED

ZONES

ZONING

ZOO

ZOOLOGICAL

ZOOLOGICALLY

ZOOM

ZOOMS

ZOOS

ZORN

ZOROASTER

ZOROASTRIAN

ZULU

ZULUS

ZURICH

pdf_password_breaker

"""

Brute force password breaker using a dictionary containing

English words.

"""



import sys

import PyPDF2

from pathlib import Path



def get_filename_from_user() -> Path:

    """Asks for a path from the User"""

    while True:

        filename: str = input("Enter filename in folder of script:")

        path: Path = Path(sys.path[0], filename)



        if path.is_file():

            return path.as_posix()

        print("File doesn't existn")





def decrypt(pdf_filename: Path, password: str) -> bool:

    """

    Try to decrypt a file. If not successful a false is returned.

    If the file passed is not encrypted also a false is passed

    """

    with open(pdf_filename, 'rb') as pdf_file:

        pdf_reader = PyPDF2.PdfFileReader(pdf_file)

        pdf_reader.decrypt(password)

        pdf_writer = PyPDF2.PdfFileWriter()



        try:

            for page_number in range(pdf_reader.numPages):

                pdf_writer.addPage(pdf_reader.getPage(page_number))

        except PyPDF2.utils.PdfReadError:

            return False



        new_name: str = pdf_filename.stem + "_decrypted.pdf"

        filename_decrypted = pdf_filename.parent / new_name



        with open(filename_decrypted, 'wb') as pdf_file_decrypted:

            pdf_writer.write(pdf_file_decrypted)

    return True





def break_encryption(pdf_filename: Path, dictionary_filename: str) -> bool:

    """Try's out words from a dictionary to break encryption"""

    with open(dictionary_filename, 'r') as dictionary_file:

        keyword: str = dictionary_file.readline().strip()



        if decrypt(pdf_filename, keyword):

            return keyword

        if decrypt(pdf_filename, keyword.lower()):

            return keyword.lower()



        while keyword:

            keyword = dictionary_file.readline().strip()



            if decrypt(pdf_filename, keyword):

                return keyword

            if decrypt(pdf_filename, keyword.lower()):

                return keyword.lower()

    return None





def pdf_password_breaker():

    """main loop"""

    filename: Path = get_filename_from_user()

    password: str = break_encryption(filename, "dictionary.txt")



    if password:

        print("File unlocked. Password was:" + password)

        return

    print("File could not be unlocked")



if __name__ == "__main__":

    pdf_password_breaker()

I tried a file with a simple password like "hello". It works but it takes a lot of time until it reaches the "hello" in the file.

I wonder if theres a way to improve the speed, or any other aspect of the code.

Looking at the documentation of PdfFileReader.decrypt It states that it return 0 if the file fails or 1 if it succeeds.

You can also use isEncrypted to check if the file was encrypted in the first place.

So you can turn the loop around. First open the pdf file see if it was encrypted at all and if so try to call decrypt with every password in your dictionary.