Custom Indexing Python Data Structure

Question

I have a class that wraps around python deque from collections. When I go and create a deque x=deque(), and I want to reference the first variable....

In[78]: x[0]
Out[78]: 0

My question is how can use the [] for referencing in the following example wrapper

class deque_wrapper:
    def __init__(self):
        self.data_structure = deque()

    def newCustomAddon(x):
        return len(self.data_structure)

    def __repr__(self):
        return repr(self.data_structure)

Ie, continuing from above example:

In[75]: x[0]
Out[76]: TypeError: 'deque_wrapper' object does not support indexing

I want to customize my own referencing, is that possible?

Instead of asking a separate question for each method needed to emulate a deque, you should skim the Data model section of the docs (that you've now been given links to in both questions). Almost everything you're likely to ask for is going to be there. — abarnert
– abarnert, Commented Sep 8, 2014 at 3:19
Also see collections.abc.MutableSequence. If you inherit from that, and implement the give methods listed in the table, it'll automatically give you all the other methods that deque, list, and similar types have. (If you're in Python 2.x, it's called collections.MutableSequence, no abc… but if you're in 2.x, you shouldn't be declaring a class with no bases like this, as someone else explained on your previous question.) — abarnert
– abarnert, Commented Sep 8, 2014 at 3:21

Michael0x2a · Accepted Answer · 2014-09-08 02:50:29Z

4

You want to implement the __getitem__ method:

class DequeWrapper:
    def __init__(self):
        self.data_structure = deque()

    def newCustomAddon(x):
        return len(self.data_structure)

    def __repr__(self):
        return repr(self.data_structure)

    def __getitem__(self, index):
        # etc

Whenever you do my_obj[x], Python will actually call my_obj.__getitem__(x).

You may also want to consider implementing the __setitem__ method, if applicable. (When you write my_obj[x] = y, Python will actually run my_obj.__setitem__(x, y).

The documentation on Python data models will contain more information on which methods you need to implement in order to make custom data structures in Python.

answered Sep 8, 2014 at 2:50

Michael0x2a

65.6k32 gold badges190 silver badges241 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

abarnert Over a year ago

If you're going to implement __getitem__ and __setitem__, unless you've got an explicitly-fixed-size collection, you probably also want __delitem__ (and probably insert and append and so on, too).

Collectives™ on Stack Overflow

Custom Indexing Python Data Structure

1 Answer 1

1 Comment

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

1 Comment

Your Answer

Sign up or log in

Post as a guest

Linked

Related