3
\$\begingroup\$

I am trying to improve the performance of this password cracking method:

def search_method_2(num_pass_wheels):
    global totalguesses
    result = False
    starttime = time.time()
    tests = 0
    still_searching = True
    print("Using method 2 and searching with "+str(num_pass_wheels)+" password wheels.")
    wheel = " ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789"
    # we only allow up to 8 wheels for each password for now
    if (num_pass_wheels > 8):
        print("Unable to handle the request. No more than 8 characters for a password")
        still_searching = False
    # set all of the wheels to the first position
    pass_wheel_array=array('i',[1,0,0,0,0,0,0,0,0])

    while still_searching:
        ourguess_pass = ""
        for i in range(0,num_pass_wheels):  # once for each wheel
            if pass_wheel_array[i] > 0:
                ourguess_pass = wheel[pass_wheel_array[i]] + ourguess_pass
        #print ("trying ["+ourguess_pass+"]")
        if (check_userpass(which_password, ourguess_pass)):
            print ("Success! Password  "+str(which_password)+" is " + ourguess_pass)
            still_searching = False   # we can stop now - we found it!
            result = True
        #else:
            #print ("Darn. " + ourguess + " is NOT the password.")
        tests = tests + 1
        totalguesses = totalguesses + 1

        carry = 1
        for i in range(0,num_pass_wheels): # once for each wheel
            pass_wheel_array[i] = pass_wheel_array[i] + carry
            carry = 0
            if pass_wheel_array[i] > 62:
                pass_wheel_array[i] = 1
                carry = 1
                if i == (num_pass_wheels-1):
                    still_searching = False

    seconds = time.time()-starttime
    report_search_time(tests, seconds)
    return result

This is a modified version of the Science Buddies password cracker (found here: http://www.sciencebuddies.org/science-fair-projects/project_ideas/CompSci_p046.shtml#materials

You can see the full file here if you would like to: https://files.fm/u/bg9ppx77

Basically, I need a way to run this method faster, I can currently do about 200,000 tests (or runs of this method) per second. I would like to double or triple that number so that I can "crack" longer passwords in a reasonable period of time.

If anyone could point me in the right direction that would be great, if anyone has any code ideas that would be amazing.

\$\endgroup\$
  • 2
    \$\begingroup\$ you might want to use any profiler to figure out which part/call takes most of the time and focus on it. \$\endgroup\$ – Alex Nov 1 '16 at 10:05
  • \$\begingroup\$ Your indentation was most likely not properly shown due to a copy-paste error. I fixed it for you. Let me know if this isn't exactly as your own code. Indentation is important in Python. \$\endgroup\$ – Mast Nov 1 '16 at 10:29
  • 1
    \$\begingroup\$ How is check_userpass implemented? Since it is called every loop iteration, this is quite important for the performance... \$\endgroup\$ – Graipher Nov 1 '16 at 16:27
  • \$\begingroup\$ The more you call functions, the more you spend time. \$\endgroup\$ – Laurent LAPORTE Nov 2 '16 at 23:06
  • \$\begingroup\$ In the future please declare your cross-posts. stackoverflow.com/questions/40352390/… \$\endgroup\$ – janos Nov 3 '16 at 21:22

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.