Welcome, guest | Sign In | My Account | Store | Cart
2

Many comparable or hashable classes calculate all comparison and hash results from the same simple calculation. The mixins in this recipe allow such classes to define a single __key__() method from which all the __eq__(), __lt__(), __hash__(), etc. methods are automatically defined.

Python, 59 lines
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
class KeyedEqualityMixin(object):
   def __eq__(self, other):
       return self.__key__() == other.__key__()
   def __ne__(self, other):
       return self.__key__() != other.__key__()

class KeyedComparisonMixin(KeyedEqualityMixin):
   def __lt__(self, other):
       return self.__key__() < other.__key__()
   def __le__(self, other):
       return self.__key__() <= other.__key__()
   def __gt__(self, other):
       return self.__key__() > other.__key__()
   def __ge__(self, other):
       return self.__key__() >= other.__key__()

class KeyedHashingMixin(KeyedEqualityMixin):
    def __hash__(self):
        return hash(self.__key__())


# =============================
# And at the interactive prompt
# =============================

>>> class C(object):
...     def __init__(self, x):
...         self.x = x
... 
>>> x, y, z = C(1), C(1), C(2)
>>> x == y, y < z, hash(x) == hash(y)
(False, False, False)
>>> class C(KeyedEqualityMixin):
...     def __init__(self, x):
...         self.x = x
...     def __key__(self):
...         return self.x
...        
>>> x, y, z = C(1), C(1), C(2)
>>> x == y, y < z, hash(x) == hash(y)
(True, False, False)
>>> class C(KeyedComparisonMixin):
...     def __init__(self, x):
...         self.x = x
...     def __key__(self):
...         return self.x
... 
>>> x, y, z = C(1), C(1), C(2)
>>> x == y, y < z, hash(x) == hash(y)
(True, True, False)
>>> class C(KeyedHashingMixin):
...     def __init__(self, x):
...         self.x = x
...     def __key__(self):
...         return self.x
... 
>>> x, y, z = C(1), C(1), C(2)
>>> x == y, y < z, hash(x) == hash(y)
(True, False, True)

This approach is generally cleaner than writing a __cmp__() method where you'd often have to specify the attribute(s) twice, e.g.:: <pre> def __cmp__(self, other): return cmp(self.x, other.x)</pre> With the __key__() approach, you only need to name the attribute once.

If you've been relying on __cmp__ methods to handle all your rich comparisons, you should probably start using an approach like the one in this recipe because the __cmp__ method will no longer exist in Python 3000.

Add a comment

Sign in to comment

Created by Steven Bethard on Wed, 21 Mar 2007 (PSF)
Python recipes (4258)
Steven Bethard's recipes (7)

Required Modules

  • (none specified)

Other Information and Tasks