0

I have a question. How can I get "reference-pointer effect" in Python 2.x?

I have a class, containing 2 dictionaries - 1 with character representation and 1 with integer representation (retrieved with ord(character)). Main problem is I will print them a lot of times, so converting them in-the-fly is a bad idea (I think). However, switching between them would be useful. 

class Character_Dict(object):
    def __init__(self, key_length):
        self.char_dict = {}
        self.ASCII_dict = {}
        self.key_length = key_length
        self.dictionary = <<here should be a dictionary in usage>>

I could just assign wanted one to self.dictionary, but when assigning for example self.char_dict any change in self.dictionary would not apply to self.char_dict which in my case is a point of whole this construction.

Is there any mechanism in Python 2.x which would allow to do things like that?

[EDIT 1]: Dictionaries contains lists of symbols encrypted with usage of n-th byte of some key: 

n [list of character encrypted with n-th byte of key]

0 ['\xc5', '\x9a', '\xa5', '\x8d', '\xc8', '\xc8', '\x92', '\x9b', '\x82', '\x92', '\x86']
1 ['a', 'm', '.', 'a', '%', ',', ' ', '*', '$', ' ', '(']
2 ['\x18', '~', '4', '4', '?', ',', ',', '0', '9', ',', '\xe7']
3 ['\xe8', '\xe2', '\xe8', '\xec', ':', '\xe6', '\xe6', '0', '\xe6', '\xf3', '\xa9']
...
255 ['\x12', 'S', '\xcc', '_', '\xc0', 'S', '\x01', 'S', 'S', 'S']

[EDIT 2]: My encryption key has 256 bytes. Message which was encrypted by usage of that key is 2688 bytes long. That means that encryption key was repeated 10.5 times.

Think about changing the 3rd letter which was encrypted with usage of 10th letter of key. That's (3-1)*256+10 byte. Instead of reading that letter, I can simply read whole stream and use my class

fileXOR = open('2011061.xor', 'r')
key_length = 256

number_of_bytes = os.path.getsize('2011061.xor')
print number_of_bytes
amount_of_key_repetition = number_of_bytes/key_length.__float__()
print "Key has been repeated", amount_of_key_repetition, "times"

character_dict = Character_Dict(key_length)
for counter_x in range(0, key_length):
    character_dict.dictionary[counter_x] = []
print character_dict

for current_byte in range(0, number_of_bytes):
    read_character = fileXOR.read(1)
    character_dict.dictionary[current_byte % key_length].append(read_character)

fileXOR.close()

and I can access my character simply by:

character_dict.dictionary[10][2]

Now, imagine that I need change character_dict.dictionary[10][2]. In constructor I had assigned self.char_dict to self.dictionary. Changing object_name.dictionary will not modify object_name.char_dict (AFAIK).

I want object_name.dictionary to be sometimes a ASCII representation, and sometimes an integer representation. That would reduce a lot of code and simplify any changes made into ciphertext.

Improve this question
12
  • 3
    Could you provide an example of what the dicts would contain? Commented May 26, 2014 at 19:44
  • 2
    Can you post some code that reproduces the problem you are trying to solve? Commented May 26, 2014 at 19:46
  • um... in python, if A = ObjA and B = A then B = ObjA. Any change in the state of either A or B will be reflected in the other variable, because they're both references to the same object. In python, everything that isn't an immutable type like str or the numeric classes are references. Commented May 26, 2014 at 19:49
  • @aruisdante - that rule apply to standard Python dictionaries? Commented May 26, 2014 at 19:51
  • 1
    @aruisdante Every type is a reference type. Including str, "numeric types", tuples, and everything else you can lay your hand on using Python. Commented May 26, 2014 at 20:01

2 Answers 2

Reset to default
1

In Python, the weakref module lets you store references to objects. If the source object gets deleted or garbage collected, this can be detected by calling the weakref.

Example from documentation:

>>> import weakref
>>> class Object:
...     pass
...
>>> o = Object()
>>> r = weakref.ref(o)
>>> o2 = r()
>>> o is o2
True

If the referent no longer exists, calling the reference object returns None:

>>> del o, o2
>>> print r()
None
Improve this answer
Sign up to request clarification or add additional context in comments.

7 Comments

I'm not actually sure what problem OP is facing, but from what I understand I don't see even the slightest relation to weak references. Care to explain how this solves what problem?
@delnan >>> o = Object() >>> r = weakref.ref(o) >>> o2 = r() Any change made in o would be applied to o2. The same thing I wanted in my code. Any change into objectA.dictionary will apply to currently assigned dictionary (ASCII or int). When I need to change current dictionary, I will just use r = weakref.ref(wantedDict) objectA.dictionary = r()
@wojciechowskip But the exact same effect is achieved with o = Object(); o2 = o.
@delnan Does Python create a copy of that object instead of passing reference, doesn't it? And changes in o2 will not apply in o?
@wojciechowskip No such thing happens. As several people already told you. Try it out. You appear to be speculating without any basis. It works. Just do it. (Also, any sane rule set which would cause a copy in o = o2 would also cause at least one copy in o = r().) Also, if after trying it you find that something is actually not working as you expect, post a minimal example that displays the problem so that we can tell you what exactly went wrong (either in the code or in your thought process).
|
0

So, I totally was in misconception of assigning. I was assured that every time you are assigning to objX there is created a copy of this objX. The source of this assumption was simply:

a = 8
b = a
b = 20
print a
>>> 8

I feel ashamed now. At least I understand how memory in Python is managed :)

Improve this answer

Comments

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.