collections

New In Python 2.4

Counter

New In Python 2.7

A Counter is a dict subclass for counting hashable objects. It is an unordered collection where elements are stored as dictionary keys and their counts are stored as dictionary values. Counts are allowed to be any integer value including zero or negative counts. The Counter class is similar to bags or multisets in other languages.

In [1]: import collections

In [2]: c = collections.Counter('Hello Python!')

In [3]: c
Out[3]:
Counter({' ': 1,
         '!': 1,
         'H': 1,
         'P': 1,
         'e': 1,
         'h': 1,
         'l': 2,
         'n': 1,
         'o': 2,
         't': 1,
         'y': 1})

• TopN

  In [4]: c.most_common()
  Out[4]:
  [('l', 2),
   ('o', 2),
   ('!', 1),
   (' ', 1),
   ('e', 1),
   ('H', 1),
   ('n', 1),
   ('P', 1),
   ('t', 1),
   ('h', 1),
   ('y', 1)]
  
  In [5]: c.most_common(3)
  Out[5]: [('l', 2), ('o', 2), ('!', 1)]
  

• -TopN

  In [6]: c.most_common()[:-2-1:-1]
  Out[6]: [('y', 1), ('h', 1)]
  

• Remove Zero and Negative Counts

  In [7]: c = collections.Counter({'a': 1, 'b': 0, 'c': -1})
  
  In [8]: c
  Out[8]: Counter({'a': 1, 'b': 0, 'c': -1})
  
  In [9]: c += collections.Counter()
  
  In [10]: c
  Out[10]: Counter({'a': 1})
  

OrderedDict

New In Python 2.7

Ordered dictionaries are just like regular dictionaries but they remember the order that items were inserted. When iterating over an ordered dictionary, the items are returned in the order their keys were first added.

In [11]: items = (
   ....:     ('A', 1),
   ....:     ('B', 2),
   ....:     ('C', 3)
   ....: )

In [12]: dict(items)
Out[12]: {'A': 1, 'B': 2, 'C': 3}

In [13]: collections.OrderedDict(items)
Out[13]: OrderedDict([('A', 1), ('B', 2), ('C', 3)])

In [14]: print _12
{'A': 1, 'C': 3, 'B': 2}

In [15]: print _13
OrderedDict([('A', 1), ('B', 2), ('C', 3)])

In [16]: for k, v in _12.iteritems():
   ....:     print k, v
   ....:
A 1
C 3
B 2

In [17]: for k, v in _13.iteritems():
   ....:     print k, v
   ....:
A 1
B 2
C 3

References

[1] Docs@Python, 8.3. collections — High-performance container datatypes