Raymond Hettinger | 53dbe39 | 2008-02-12 20:03:09 +0000 | [diff] [blame] | 1 | :mod:`collections` --- Container datatypes |
| 2 | ========================================== |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 3 | |
| 4 | .. module:: collections |
Raymond Hettinger | 53dbe39 | 2008-02-12 20:03:09 +0000 | [diff] [blame] | 5 | :synopsis: Container datatypes |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 6 | .. moduleauthor:: Raymond Hettinger <python@rcn.com> |
| 7 | .. sectionauthor:: Raymond Hettinger <python@rcn.com> |
| 8 | |
Christian Heimes | fe337bf | 2008-03-23 21:54:12 +0000 | [diff] [blame] | 9 | .. testsetup:: * |
| 10 | |
| 11 | from collections import * |
| 12 | import itertools |
| 13 | __name__ = '<doctest>' |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 14 | |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 15 | This module implements high-performance container datatypes. Currently, |
Raymond Hettinger | 2d32f63 | 2009-03-02 21:24:57 +0000 | [diff] [blame] | 16 | there are four datatypes, :class:`Counter`, :class:`deque`, :class:`OrderedDict` and |
Raymond Hettinger | acd82b9 | 2009-02-17 20:06:51 +0000 | [diff] [blame] | 17 | :class:`defaultdict`, and one datatype factory function, :func:`namedtuple`. |
Christian Heimes | 0bd4e11 | 2008-02-12 22:59:25 +0000 | [diff] [blame] | 18 | |
Raymond Hettinger | ebcee3f | 2008-02-06 19:54:00 +0000 | [diff] [blame] | 19 | The specialized containers provided in this module provide alternatives |
Christian Heimes | fe337bf | 2008-03-23 21:54:12 +0000 | [diff] [blame] | 20 | to Python's general purpose built-in containers, :class:`dict`, |
Raymond Hettinger | ebcee3f | 2008-02-06 19:54:00 +0000 | [diff] [blame] | 21 | :class:`list`, :class:`set`, and :class:`tuple`. |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 22 | |
Mark Summerfield | 08898b4 | 2007-09-05 08:43:04 +0000 | [diff] [blame] | 23 | In addition to containers, the collections module provides some ABCs |
Christian Heimes | fe337bf | 2008-03-23 21:54:12 +0000 | [diff] [blame] | 24 | (abstract base classes) that can be used to test whether a class |
Raymond Hettinger | acd82b9 | 2009-02-17 20:06:51 +0000 | [diff] [blame] | 25 | provides a particular interface, for example, whether it is hashable or |
| 26 | a mapping. |
Raymond Hettinger | ebcee3f | 2008-02-06 19:54:00 +0000 | [diff] [blame] | 27 | |
| 28 | ABCs - abstract base classes |
| 29 | ---------------------------- |
| 30 | |
| 31 | The collections module offers the following ABCs: |
Mark Summerfield | 08898b4 | 2007-09-05 08:43:04 +0000 | [diff] [blame] | 32 | |
Georg Brandl | 86b2fb9 | 2008-07-16 03:43:04 +0000 | [diff] [blame] | 33 | ========================= ===================== ====================== ==================================================== |
| 34 | ABC Inherits Abstract Methods Mixin Methods |
| 35 | ========================= ===================== ====================== ==================================================== |
| 36 | :class:`Container` ``__contains__`` |
| 37 | :class:`Hashable` ``__hash__`` |
| 38 | :class:`Iterable` ``__iter__`` |
| 39 | :class:`Iterator` :class:`Iterable` ``__next__`` ``__iter__`` |
Georg Brandl | a1c6a1c | 2009-01-03 21:26:05 +0000 | [diff] [blame] | 40 | :class:`Sized` ``__len__`` |
Georg Brandl | 86b2fb9 | 2008-07-16 03:43:04 +0000 | [diff] [blame] | 41 | :class:`Callable` ``__call__`` |
Georg Brandl | 48310cd | 2009-01-03 21:18:54 +0000 | [diff] [blame] | 42 | |
Georg Brandl | 86b2fb9 | 2008-07-16 03:43:04 +0000 | [diff] [blame] | 43 | :class:`Sequence` :class:`Sized`, ``__getitem__`` ``__contains__``. ``__iter__``, ``__reversed__``. |
Raymond Hettinger | d23e013 | 2009-01-29 00:01:27 +0000 | [diff] [blame] | 44 | :class:`Iterable`, ``index``, and ``count`` |
Georg Brandl | 48310cd | 2009-01-03 21:18:54 +0000 | [diff] [blame] | 45 | :class:`Container` |
| 46 | |
Raymond Hettinger | d23e013 | 2009-01-29 00:01:27 +0000 | [diff] [blame] | 47 | :class:`MutableSequence` :class:`Sequence` ``__setitem__`` Inherited Sequence methods and |
Georg Brandl | 86b2fb9 | 2008-07-16 03:43:04 +0000 | [diff] [blame] | 48 | ``__delitem__``, ``append``, ``reverse``, ``extend``, ``pop``, |
Raymond Hettinger | d23e013 | 2009-01-29 00:01:27 +0000 | [diff] [blame] | 49 | and ``insert`` ``remove``, and ``__iadd__`` |
Georg Brandl | 48310cd | 2009-01-03 21:18:54 +0000 | [diff] [blame] | 50 | |
Raymond Hettinger | d23e013 | 2009-01-29 00:01:27 +0000 | [diff] [blame] | 51 | :class:`Set` :class:`Sized`, ``__le__``, ``__lt__``, ``__eq__``, ``__ne__``, |
| 52 | :class:`Iterable`, ``__gt__``, ``__ge__``, ``__and__``, ``__or__`` |
| 53 | :class:`Container` ``__sub__``, ``__xor__``, and ``isdisjoint`` |
Georg Brandl | 48310cd | 2009-01-03 21:18:54 +0000 | [diff] [blame] | 54 | |
Georg Brandl | 86b2fb9 | 2008-07-16 03:43:04 +0000 | [diff] [blame] | 55 | :class:`MutableSet` :class:`Set` ``add`` and Inherited Set methods and |
| 56 | ``discard`` ``clear``, ``pop``, ``remove``, ``__ior__``, |
| 57 | ``__iand__``, ``__ixor__``, and ``__isub__`` |
Georg Brandl | 48310cd | 2009-01-03 21:18:54 +0000 | [diff] [blame] | 58 | |
Raymond Hettinger | d23e013 | 2009-01-29 00:01:27 +0000 | [diff] [blame] | 59 | :class:`Mapping` :class:`Sized`, ``__getitem__`` ``__contains__``, ``keys``, ``items``, ``values``, |
| 60 | :class:`Iterable`, ``get``, ``__eq__``, and ``__ne__`` |
| 61 | :class:`Container` |
Georg Brandl | 48310cd | 2009-01-03 21:18:54 +0000 | [diff] [blame] | 62 | |
Raymond Hettinger | d23e013 | 2009-01-29 00:01:27 +0000 | [diff] [blame] | 63 | :class:`MutableMapping` :class:`Mapping` ``__setitem__`` and Inherited Mapping methods and |
| 64 | ``__delitem__`` ``pop``, ``popitem``, ``clear``, ``update``, |
| 65 | and ``setdefault`` |
| 66 | |
Georg Brandl | 48310cd | 2009-01-03 21:18:54 +0000 | [diff] [blame] | 67 | |
Georg Brandl | 86b2fb9 | 2008-07-16 03:43:04 +0000 | [diff] [blame] | 68 | :class:`MappingView` :class:`Sized` ``__len__`` |
| 69 | :class:`KeysView` :class:`MappingView`, ``__contains__``, |
| 70 | :class:`Set` ``__iter__`` |
| 71 | :class:`ItemsView` :class:`MappingView`, ``__contains__``, |
| 72 | :class:`Set` ``__iter__`` |
| 73 | :class:`ValuesView` :class:`MappingView` ``__contains__``, ``__iter__`` |
| 74 | ========================= ===================== ====================== ==================================================== |
Mark Summerfield | 08898b4 | 2007-09-05 08:43:04 +0000 | [diff] [blame] | 75 | |
Mark Summerfield | 08898b4 | 2007-09-05 08:43:04 +0000 | [diff] [blame] | 76 | These ABCs allow us to ask classes or instances if they provide |
| 77 | particular functionality, for example:: |
| 78 | |
Mark Summerfield | 08898b4 | 2007-09-05 08:43:04 +0000 | [diff] [blame] | 79 | size = None |
Raymond Hettinger | ebcee3f | 2008-02-06 19:54:00 +0000 | [diff] [blame] | 80 | if isinstance(myvar, collections.Sized): |
Georg Brandl | a1c6a1c | 2009-01-03 21:26:05 +0000 | [diff] [blame] | 81 | size = len(myvar) |
Mark Summerfield | 08898b4 | 2007-09-05 08:43:04 +0000 | [diff] [blame] | 82 | |
Raymond Hettinger | ebcee3f | 2008-02-06 19:54:00 +0000 | [diff] [blame] | 83 | Several of the ABCs are also useful as mixins that make it easier to develop |
| 84 | classes supporting container APIs. For example, to write a class supporting |
| 85 | the full :class:`Set` API, it only necessary to supply the three underlying |
| 86 | abstract methods: :meth:`__contains__`, :meth:`__iter__`, and :meth:`__len__`. |
| 87 | The ABC supplies the remaining methods such as :meth:`__and__` and |
| 88 | :meth:`isdisjoint` :: |
| 89 | |
| 90 | class ListBasedSet(collections.Set): |
Raymond Hettinger | c1b6a4a | 2008-02-08 23:46:23 +0000 | [diff] [blame] | 91 | ''' Alternate set implementation favoring space over speed |
| 92 | and not requiring the set elements to be hashable. ''' |
Raymond Hettinger | ebcee3f | 2008-02-06 19:54:00 +0000 | [diff] [blame] | 93 | def __init__(self, iterable): |
Raymond Hettinger | c1b6a4a | 2008-02-08 23:46:23 +0000 | [diff] [blame] | 94 | self.elements = lst = [] |
| 95 | for value in iterable: |
| 96 | if value not in lst: |
| 97 | lst.append(value) |
Raymond Hettinger | ebcee3f | 2008-02-06 19:54:00 +0000 | [diff] [blame] | 98 | def __iter__(self): |
| 99 | return iter(self.elements) |
| 100 | def __contains__(self, value): |
| 101 | return value in self.elements |
| 102 | def __len__(self): |
| 103 | return len(self.elements) |
| 104 | |
| 105 | s1 = ListBasedSet('abcdef') |
| 106 | s2 = ListBasedSet('defghi') |
| 107 | overlap = s1 & s2 # The __and__() method is supported automatically |
| 108 | |
Raymond Hettinger | 7aebb64 | 2008-02-09 03:25:08 +0000 | [diff] [blame] | 109 | Notes on using :class:`Set` and :class:`MutableSet` as a mixin: |
| 110 | |
Christian Heimes | fe337bf | 2008-03-23 21:54:12 +0000 | [diff] [blame] | 111 | (1) |
Raymond Hettinger | 7aebb64 | 2008-02-09 03:25:08 +0000 | [diff] [blame] | 112 | Since some set operations create new sets, the default mixin methods need |
Christian Heimes | fe337bf | 2008-03-23 21:54:12 +0000 | [diff] [blame] | 113 | a way to create new instances from an iterable. The class constructor is |
| 114 | assumed to have a signature in the form ``ClassName(iterable)``. |
Benjamin Peterson | 2b7411d | 2008-05-26 17:36:47 +0000 | [diff] [blame] | 115 | That assumption is factored-out to an internal classmethod called |
Raymond Hettinger | 7aebb64 | 2008-02-09 03:25:08 +0000 | [diff] [blame] | 116 | :meth:`_from_iterable` which calls ``cls(iterable)`` to produce a new set. |
| 117 | If the :class:`Set` mixin is being used in a class with a different |
Christian Heimes | fe337bf | 2008-03-23 21:54:12 +0000 | [diff] [blame] | 118 | constructor signature, you will need to override :meth:`from_iterable` |
| 119 | with a classmethod that can construct new instances from |
Raymond Hettinger | 7aebb64 | 2008-02-09 03:25:08 +0000 | [diff] [blame] | 120 | an iterable argument. |
| 121 | |
| 122 | (2) |
| 123 | To override the comparisons (presumably for speed, as the |
| 124 | semantics are fixed), redefine :meth:`__le__` and |
| 125 | then the other operations will automatically follow suit. |
Raymond Hettinger | ebcee3f | 2008-02-06 19:54:00 +0000 | [diff] [blame] | 126 | |
Raymond Hettinger | 0dbdab2 | 2008-02-09 03:48:16 +0000 | [diff] [blame] | 127 | (3) |
| 128 | The :class:`Set` mixin provides a :meth:`_hash` method to compute a hash value |
| 129 | for the set; however, :meth:`__hash__` is not defined because not all sets |
| 130 | are hashable or immutable. To add set hashabilty using mixins, |
| 131 | inherit from both :meth:`Set` and :meth:`Hashable`, then define |
| 132 | ``__hash__ = Set._hash``. |
| 133 | |
Raymond Hettinger | be075b1 | 2009-03-20 18:33:06 +0000 | [diff] [blame] | 134 | .. seealso:: |
| 135 | |
| 136 | * `OrderedSet recipe <http://code.activestate.com/recipes/576694/>`_ for an |
| 137 | example built on :class:`MutableSet`. |
| 138 | |
| 139 | * For more about ABCs, see the :mod:`abc` module and :pep:`3119`. |
Mark Summerfield | 08898b4 | 2007-09-05 08:43:04 +0000 | [diff] [blame] | 140 | |
| 141 | |
Raymond Hettinger | b8baf63 | 2009-01-14 02:20:07 +0000 | [diff] [blame] | 142 | :class:`Counter` objects |
| 143 | ------------------------ |
| 144 | |
| 145 | A counter tool is provided to support convenient and rapid tallies. |
| 146 | For example:: |
| 147 | |
Raymond Hettinger | 1c62dc9 | 2009-02-04 11:41:45 +0000 | [diff] [blame] | 148 | >>> # Tally occurrences of words in a list |
Raymond Hettinger | b8baf63 | 2009-01-14 02:20:07 +0000 | [diff] [blame] | 149 | >>> cnt = Counter() |
Raymond Hettinger | 670eaec | 2009-01-21 23:14:07 +0000 | [diff] [blame] | 150 | >>> for word in ['red', 'blue', 'red', 'green', 'blue', 'blue']: |
Raymond Hettinger | b8baf63 | 2009-01-14 02:20:07 +0000 | [diff] [blame] | 151 | ... cnt[word] += 1 |
| 152 | >>> cnt |
| 153 | Counter({'blue': 3, 'red': 2, 'green': 1}) |
| 154 | |
Raymond Hettinger | 1c62dc9 | 2009-02-04 11:41:45 +0000 | [diff] [blame] | 155 | >>> # Find the ten most common words in Hamlet |
Raymond Hettinger | b8baf63 | 2009-01-14 02:20:07 +0000 | [diff] [blame] | 156 | >>> import re |
| 157 | >>> words = re.findall('\w+', open('hamlet.txt').read().lower()) |
Raymond Hettinger | 0bae662 | 2009-01-20 13:00:59 +0000 | [diff] [blame] | 158 | >>> Counter(words).most_common(10) |
Raymond Hettinger | b8baf63 | 2009-01-14 02:20:07 +0000 | [diff] [blame] | 159 | [('the', 1143), ('and', 966), ('to', 762), ('of', 669), ('i', 631), |
| 160 | ('you', 554), ('a', 546), ('my', 514), ('hamlet', 471), ('in', 451)] |
| 161 | |
| 162 | .. class:: Counter([iterable-or-mapping]) |
| 163 | |
Raymond Hettinger | 670eaec | 2009-01-21 23:14:07 +0000 | [diff] [blame] | 164 | A :class:`Counter` is a :class:`dict` subclass for counting hashable objects. |
Raymond Hettinger | b8baf63 | 2009-01-14 02:20:07 +0000 | [diff] [blame] | 165 | It is an unordered collection where elements are stored as dictionary keys |
| 166 | and their counts are stored as dictionary values. Counts are allowed to be |
| 167 | any integer value including zero or negative counts. The :class:`Counter` |
| 168 | class is similar to bags or multisets in other languages. |
| 169 | |
| 170 | Elements are counted from an *iterable* or initialized from another |
Benjamin Peterson | 25c95f1 | 2009-05-08 20:42:26 +0000 | [diff] [blame] | 171 | *mapping* (or counter): |
Raymond Hettinger | b8baf63 | 2009-01-14 02:20:07 +0000 | [diff] [blame] | 172 | |
Raymond Hettinger | 73662a5 | 2009-01-27 02:38:22 +0000 | [diff] [blame] | 173 | >>> c = Counter() # a new, empty counter |
| 174 | >>> c = Counter('gallahad') # a new counter from an iterable |
| 175 | >>> c = Counter({'red': 4, 'blue': 2}) # a new counter from a mapping |
| 176 | >>> c = Counter(cats=4, dogs=8) # a new counter from keyword args |
Raymond Hettinger | b8baf63 | 2009-01-14 02:20:07 +0000 | [diff] [blame] | 177 | |
Raymond Hettinger | 670eaec | 2009-01-21 23:14:07 +0000 | [diff] [blame] | 178 | Counter objects have a dictionary interface except that they return a zero |
Benjamin Peterson | 25c95f1 | 2009-05-08 20:42:26 +0000 | [diff] [blame] | 179 | count for missing items instead of raising a :exc:`KeyError`: |
Raymond Hettinger | b8baf63 | 2009-01-14 02:20:07 +0000 | [diff] [blame] | 180 | |
Raymond Hettinger | 94adc8e | 2009-01-22 05:27:37 +0000 | [diff] [blame] | 181 | >>> c = Counter(['eggs', 'ham']) |
Raymond Hettinger | b8baf63 | 2009-01-14 02:20:07 +0000 | [diff] [blame] | 182 | >>> c['bacon'] # count of a missing element is zero |
| 183 | 0 |
| 184 | |
Raymond Hettinger | 94adc8e | 2009-01-22 05:27:37 +0000 | [diff] [blame] | 185 | Setting a count to zero does not remove an element from a counter. |
| 186 | Use ``del`` to remove it entirely: |
Raymond Hettinger | b8baf63 | 2009-01-14 02:20:07 +0000 | [diff] [blame] | 187 | |
Raymond Hettinger | 94adc8e | 2009-01-22 05:27:37 +0000 | [diff] [blame] | 188 | >>> c['sausage'] = 0 # counter entry with a zero count |
| 189 | >>> del c['sausage'] # del actually removes the entry |
Raymond Hettinger | b8baf63 | 2009-01-14 02:20:07 +0000 | [diff] [blame] | 190 | |
Benjamin Peterson | d45bf58 | 2009-03-02 21:44:54 +0000 | [diff] [blame] | 191 | .. versionadded:: 3.1 |
Raymond Hettinger | b8baf63 | 2009-01-14 02:20:07 +0000 | [diff] [blame] | 192 | |
| 193 | |
| 194 | Counter objects support two methods beyond those available for all |
| 195 | dictionaries: |
| 196 | |
| 197 | .. method:: elements() |
| 198 | |
Raymond Hettinger | 670eaec | 2009-01-21 23:14:07 +0000 | [diff] [blame] | 199 | Return an iterator over elements repeating each as many times as its |
| 200 | count. Elements are returned in arbitrary order. If an element's count |
| 201 | is less than one, :meth:`elements` will ignore it. |
Raymond Hettinger | b8baf63 | 2009-01-14 02:20:07 +0000 | [diff] [blame] | 202 | |
Raymond Hettinger | 0bae662 | 2009-01-20 13:00:59 +0000 | [diff] [blame] | 203 | >>> c = Counter(a=4, b=2, c=0, d=-2) |
Raymond Hettinger | b8baf63 | 2009-01-14 02:20:07 +0000 | [diff] [blame] | 204 | >>> list(c.elements()) |
| 205 | ['a', 'a', 'a', 'a', 'b', 'b'] |
| 206 | |
| 207 | .. method:: most_common([n]) |
| 208 | |
Raymond Hettinger | 73662a5 | 2009-01-27 02:38:22 +0000 | [diff] [blame] | 209 | Return a list of the *n* most common elements and their counts from the |
Raymond Hettinger | d04fa31 | 2009-02-04 19:45:13 +0000 | [diff] [blame] | 210 | most common to the least. If *n* is not specified, :func:`most_common` |
Raymond Hettinger | 73662a5 | 2009-01-27 02:38:22 +0000 | [diff] [blame] | 211 | returns *all* elements in the counter. Elements with equal counts are |
Benjamin Peterson | 25c95f1 | 2009-05-08 20:42:26 +0000 | [diff] [blame] | 212 | ordered arbitrarily: |
Raymond Hettinger | b8baf63 | 2009-01-14 02:20:07 +0000 | [diff] [blame] | 213 | |
| 214 | >>> Counter('abracadabra').most_common(3) |
| 215 | [('a', 5), ('r', 2), ('b', 2)] |
| 216 | |
Raymond Hettinger | 670eaec | 2009-01-21 23:14:07 +0000 | [diff] [blame] | 217 | The usual dictionary methods are available for :class:`Counter` objects |
| 218 | except for two which work differently for counters. |
Raymond Hettinger | b8baf63 | 2009-01-14 02:20:07 +0000 | [diff] [blame] | 219 | |
| 220 | .. method:: fromkeys(iterable) |
| 221 | |
Raymond Hettinger | 73662a5 | 2009-01-27 02:38:22 +0000 | [diff] [blame] | 222 | This class method is not implemented for :class:`Counter` objects. |
Raymond Hettinger | b8baf63 | 2009-01-14 02:20:07 +0000 | [diff] [blame] | 223 | |
| 224 | .. method:: update([iterable-or-mapping]) |
| 225 | |
Raymond Hettinger | 73662a5 | 2009-01-27 02:38:22 +0000 | [diff] [blame] | 226 | Elements are counted from an *iterable* or added-in from another |
| 227 | *mapping* (or counter). Like :meth:`dict.update` but adds counts |
| 228 | instead of replacing them. Also, the *iterable* is expected to be a |
| 229 | sequence of elements, not a sequence of ``(key, value)`` pairs. |
Raymond Hettinger | b8baf63 | 2009-01-14 02:20:07 +0000 | [diff] [blame] | 230 | |
| 231 | Common patterns for working with :class:`Counter` objects:: |
| 232 | |
Raymond Hettinger | 73662a5 | 2009-01-27 02:38:22 +0000 | [diff] [blame] | 233 | sum(c.values()) # total of all counts |
| 234 | c.clear() # reset all counts |
| 235 | list(c) # list unique elements |
| 236 | set(c) # convert to a set |
| 237 | dict(c) # convert to a regular dictionary |
| 238 | c.items() # convert to a list of (elem, cnt) pairs |
| 239 | Counter(dict(list_of_pairs)) # convert from a list of (elem, cnt) pairs |
| 240 | c.most_common()[:-n:-1] # n least common elements |
| 241 | c += Counter() # remove zero and negative counts |
Raymond Hettinger | b8baf63 | 2009-01-14 02:20:07 +0000 | [diff] [blame] | 242 | |
Raymond Hettinger | 72a95cc | 2009-02-25 22:51:40 +0000 | [diff] [blame] | 243 | Several mathematical operations are provided for combining :class:`Counter` |
| 244 | objects to produce multisets (counters that have counts greater than zero). |
| 245 | Addition and subtraction combine counters by adding or subtracting the counts |
| 246 | of corresponding elements. Intersection and union return the minimum and |
| 247 | maximum of corresponding counts. Each operation can accept inputs with signed |
| 248 | counts, but the output will exclude results with counts of zero or less. |
Raymond Hettinger | 4d2073a | 2009-01-20 03:41:22 +0000 | [diff] [blame] | 249 | |
Raymond Hettinger | e0d1b9f | 2009-01-21 20:36:27 +0000 | [diff] [blame] | 250 | >>> c = Counter(a=3, b=1) |
| 251 | >>> d = Counter(a=1, b=2) |
Raymond Hettinger | 73662a5 | 2009-01-27 02:38:22 +0000 | [diff] [blame] | 252 | >>> c + d # add two counters together: c[x] + d[x] |
Raymond Hettinger | 4d2073a | 2009-01-20 03:41:22 +0000 | [diff] [blame] | 253 | Counter({'a': 4, 'b': 3}) |
Raymond Hettinger | 73662a5 | 2009-01-27 02:38:22 +0000 | [diff] [blame] | 254 | >>> c - d # subtract (keeping only positive counts) |
Raymond Hettinger | 4d2073a | 2009-01-20 03:41:22 +0000 | [diff] [blame] | 255 | Counter({'a': 2}) |
Raymond Hettinger | 73662a5 | 2009-01-27 02:38:22 +0000 | [diff] [blame] | 256 | >>> c & d # intersection: min(c[x], d[x]) |
Raymond Hettinger | 4d2073a | 2009-01-20 03:41:22 +0000 | [diff] [blame] | 257 | Counter({'a': 1, 'b': 1}) |
Raymond Hettinger | 73662a5 | 2009-01-27 02:38:22 +0000 | [diff] [blame] | 258 | >>> c | d # union: max(c[x], d[x]) |
Raymond Hettinger | 4d2073a | 2009-01-20 03:41:22 +0000 | [diff] [blame] | 259 | Counter({'a': 3, 'b': 2}) |
| 260 | |
Raymond Hettinger | b14043c | 2009-01-20 23:44:31 +0000 | [diff] [blame] | 261 | .. seealso:: |
Raymond Hettinger | b8baf63 | 2009-01-14 02:20:07 +0000 | [diff] [blame] | 262 | |
Raymond Hettinger | 94adc8e | 2009-01-22 05:27:37 +0000 | [diff] [blame] | 263 | * `Counter class <http://code.activestate.com/recipes/576611/>`_ |
| 264 | adapted for Python 2.5 and an early `Bag recipe |
| 265 | <http://code.activestate.com/recipes/259174/>`_ for Python 2.4. |
| 266 | |
Raymond Hettinger | b14043c | 2009-01-20 23:44:31 +0000 | [diff] [blame] | 267 | * `Bag class <http://www.gnu.org/software/smalltalk/manual-base/html_node/Bag.html>`_ |
| 268 | in Smalltalk. |
Raymond Hettinger | b8baf63 | 2009-01-14 02:20:07 +0000 | [diff] [blame] | 269 | |
Raymond Hettinger | b14043c | 2009-01-20 23:44:31 +0000 | [diff] [blame] | 270 | * Wikipedia entry for `Multisets <http://en.wikipedia.org/wiki/Multiset>`_\. |
Raymond Hettinger | b8baf63 | 2009-01-14 02:20:07 +0000 | [diff] [blame] | 271 | |
Raymond Hettinger | b14043c | 2009-01-20 23:44:31 +0000 | [diff] [blame] | 272 | * `C++ multisets <http://www.demo2s.com/Tutorial/Cpp/0380__set-multiset/Catalog0380__set-multiset.htm>`_ |
Raymond Hettinger | 94adc8e | 2009-01-22 05:27:37 +0000 | [diff] [blame] | 273 | tutorial with examples. |
Raymond Hettinger | b14043c | 2009-01-20 23:44:31 +0000 | [diff] [blame] | 274 | |
Raymond Hettinger | 94adc8e | 2009-01-22 05:27:37 +0000 | [diff] [blame] | 275 | * For mathematical operations on multisets and their use cases, see |
Raymond Hettinger | b14043c | 2009-01-20 23:44:31 +0000 | [diff] [blame] | 276 | *Knuth, Donald. The Art of Computer Programming Volume II, |
| 277 | Section 4.6.3, Exercise 19*\. |
| 278 | |
Raymond Hettinger | 670eaec | 2009-01-21 23:14:07 +0000 | [diff] [blame] | 279 | * To enumerate all distinct multisets of a given size over a given set of |
Raymond Hettinger | d07d939 | 2009-01-27 04:20:44 +0000 | [diff] [blame] | 280 | elements, see :func:`itertools.combinations_with_replacement`. |
Raymond Hettinger | b14043c | 2009-01-20 23:44:31 +0000 | [diff] [blame] | 281 | |
Raymond Hettinger | 94adc8e | 2009-01-22 05:27:37 +0000 | [diff] [blame] | 282 | map(Counter, combinations_with_replacement('ABC', 2)) --> AA AB AC BB BC CC |
Raymond Hettinger | b8baf63 | 2009-01-14 02:20:07 +0000 | [diff] [blame] | 283 | |
| 284 | |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 285 | :class:`deque` objects |
| 286 | ---------------------- |
| 287 | |
Georg Brandl | c2a4f4f | 2009-04-10 09:03:43 +0000 | [diff] [blame] | 288 | .. class:: deque([iterable, [maxlen]]) |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 289 | |
| 290 | Returns a new deque object initialized left-to-right (using :meth:`append`) with |
| 291 | data from *iterable*. If *iterable* is not specified, the new deque is empty. |
| 292 | |
| 293 | Deques are a generalization of stacks and queues (the name is pronounced "deck" |
| 294 | and is short for "double-ended queue"). Deques support thread-safe, memory |
| 295 | efficient appends and pops from either side of the deque with approximately the |
| 296 | same O(1) performance in either direction. |
| 297 | |
| 298 | Though :class:`list` objects support similar operations, they are optimized for |
| 299 | fast fixed-length operations and incur O(n) memory movement costs for |
| 300 | ``pop(0)`` and ``insert(0, v)`` operations which change both the size and |
| 301 | position of the underlying data representation. |
| 302 | |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 303 | |
Georg Brandl | 9afde1c | 2007-11-01 20:32:30 +0000 | [diff] [blame] | 304 | If *maxlen* is not specified or is *None*, deques may grow to an |
| 305 | arbitrary length. Otherwise, the deque is bounded to the specified maximum |
| 306 | length. Once a bounded length deque is full, when new items are added, a |
| 307 | corresponding number of items are discarded from the opposite end. Bounded |
| 308 | length deques provide functionality similar to the ``tail`` filter in |
| 309 | Unix. They are also useful for tracking transactions and other pools of data |
| 310 | where only the most recent activity is of interest. |
| 311 | |
Georg Brandl | 9afde1c | 2007-11-01 20:32:30 +0000 | [diff] [blame] | 312 | |
Benjamin Peterson | e41251e | 2008-04-25 01:59:09 +0000 | [diff] [blame] | 313 | Deque objects support the following methods: |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 314 | |
Benjamin Peterson | e41251e | 2008-04-25 01:59:09 +0000 | [diff] [blame] | 315 | .. method:: append(x) |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 316 | |
Benjamin Peterson | e41251e | 2008-04-25 01:59:09 +0000 | [diff] [blame] | 317 | Add *x* to the right side of the deque. |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 318 | |
| 319 | |
Benjamin Peterson | e41251e | 2008-04-25 01:59:09 +0000 | [diff] [blame] | 320 | .. method:: appendleft(x) |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 321 | |
Benjamin Peterson | e41251e | 2008-04-25 01:59:09 +0000 | [diff] [blame] | 322 | Add *x* to the left side of the deque. |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 323 | |
| 324 | |
Benjamin Peterson | e41251e | 2008-04-25 01:59:09 +0000 | [diff] [blame] | 325 | .. method:: clear() |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 326 | |
Benjamin Peterson | e41251e | 2008-04-25 01:59:09 +0000 | [diff] [blame] | 327 | Remove all elements from the deque leaving it with length 0. |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 328 | |
| 329 | |
Benjamin Peterson | e41251e | 2008-04-25 01:59:09 +0000 | [diff] [blame] | 330 | .. method:: extend(iterable) |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 331 | |
Benjamin Peterson | e41251e | 2008-04-25 01:59:09 +0000 | [diff] [blame] | 332 | Extend the right side of the deque by appending elements from the iterable |
| 333 | argument. |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 334 | |
| 335 | |
Benjamin Peterson | e41251e | 2008-04-25 01:59:09 +0000 | [diff] [blame] | 336 | .. method:: extendleft(iterable) |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 337 | |
Benjamin Peterson | e41251e | 2008-04-25 01:59:09 +0000 | [diff] [blame] | 338 | Extend the left side of the deque by appending elements from *iterable*. |
| 339 | Note, the series of left appends results in reversing the order of |
| 340 | elements in the iterable argument. |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 341 | |
| 342 | |
Benjamin Peterson | e41251e | 2008-04-25 01:59:09 +0000 | [diff] [blame] | 343 | .. method:: pop() |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 344 | |
Benjamin Peterson | e41251e | 2008-04-25 01:59:09 +0000 | [diff] [blame] | 345 | Remove and return an element from the right side of the deque. If no |
| 346 | elements are present, raises an :exc:`IndexError`. |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 347 | |
| 348 | |
Benjamin Peterson | e41251e | 2008-04-25 01:59:09 +0000 | [diff] [blame] | 349 | .. method:: popleft() |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 350 | |
Benjamin Peterson | e41251e | 2008-04-25 01:59:09 +0000 | [diff] [blame] | 351 | Remove and return an element from the left side of the deque. If no |
| 352 | elements are present, raises an :exc:`IndexError`. |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 353 | |
| 354 | |
Benjamin Peterson | e41251e | 2008-04-25 01:59:09 +0000 | [diff] [blame] | 355 | .. method:: remove(value) |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 356 | |
Benjamin Peterson | e41251e | 2008-04-25 01:59:09 +0000 | [diff] [blame] | 357 | Removed the first occurrence of *value*. If not found, raises a |
| 358 | :exc:`ValueError`. |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 359 | |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 360 | |
Benjamin Peterson | e41251e | 2008-04-25 01:59:09 +0000 | [diff] [blame] | 361 | .. method:: rotate(n) |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 362 | |
Benjamin Peterson | e41251e | 2008-04-25 01:59:09 +0000 | [diff] [blame] | 363 | Rotate the deque *n* steps to the right. If *n* is negative, rotate to |
| 364 | the left. Rotating one step to the right is equivalent to: |
| 365 | ``d.appendleft(d.pop())``. |
| 366 | |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 367 | |
Raymond Hettinger | 5bb0f0e | 2009-03-10 12:56:32 +0000 | [diff] [blame] | 368 | Deque objects also provide one read-only attribute: |
| 369 | |
| 370 | .. attribute:: maxlen |
| 371 | |
| 372 | Maximum size of a deque or *None* if unbounded. |
| 373 | |
Raymond Hettinger | 150fb9c | 2009-03-10 22:48:06 +0000 | [diff] [blame] | 374 | .. versionadded:: 3.1 |
Raymond Hettinger | 5bb0f0e | 2009-03-10 12:56:32 +0000 | [diff] [blame] | 375 | |
| 376 | |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 377 | In addition to the above, deques support iteration, pickling, ``len(d)``, |
| 378 | ``reversed(d)``, ``copy.copy(d)``, ``copy.deepcopy(d)``, membership testing with |
Benjamin Peterson | 206e307 | 2008-10-19 14:07:49 +0000 | [diff] [blame] | 379 | the :keyword:`in` operator, and subscript references such as ``d[-1]``. Indexed |
| 380 | access is O(1) at both ends but slows to O(n) in the middle. For fast random |
| 381 | access, use lists instead. |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 382 | |
Christian Heimes | fe337bf | 2008-03-23 21:54:12 +0000 | [diff] [blame] | 383 | Example: |
| 384 | |
| 385 | .. doctest:: |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 386 | |
| 387 | >>> from collections import deque |
| 388 | >>> d = deque('ghi') # make a new deque with three items |
| 389 | >>> for elem in d: # iterate over the deque's elements |
Neal Norwitz | 752abd0 | 2008-05-13 04:55:24 +0000 | [diff] [blame] | 390 | ... print(elem.upper()) |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 391 | G |
| 392 | H |
| 393 | I |
| 394 | |
| 395 | >>> d.append('j') # add a new entry to the right side |
| 396 | >>> d.appendleft('f') # add a new entry to the left side |
| 397 | >>> d # show the representation of the deque |
| 398 | deque(['f', 'g', 'h', 'i', 'j']) |
| 399 | |
| 400 | >>> d.pop() # return and remove the rightmost item |
| 401 | 'j' |
| 402 | >>> d.popleft() # return and remove the leftmost item |
| 403 | 'f' |
| 404 | >>> list(d) # list the contents of the deque |
| 405 | ['g', 'h', 'i'] |
| 406 | >>> d[0] # peek at leftmost item |
| 407 | 'g' |
| 408 | >>> d[-1] # peek at rightmost item |
| 409 | 'i' |
| 410 | |
| 411 | >>> list(reversed(d)) # list the contents of a deque in reverse |
| 412 | ['i', 'h', 'g'] |
| 413 | >>> 'h' in d # search the deque |
| 414 | True |
| 415 | >>> d.extend('jkl') # add multiple elements at once |
| 416 | >>> d |
| 417 | deque(['g', 'h', 'i', 'j', 'k', 'l']) |
| 418 | >>> d.rotate(1) # right rotation |
| 419 | >>> d |
| 420 | deque(['l', 'g', 'h', 'i', 'j', 'k']) |
| 421 | >>> d.rotate(-1) # left rotation |
| 422 | >>> d |
| 423 | deque(['g', 'h', 'i', 'j', 'k', 'l']) |
| 424 | |
| 425 | >>> deque(reversed(d)) # make a new deque in reverse order |
| 426 | deque(['l', 'k', 'j', 'i', 'h', 'g']) |
| 427 | >>> d.clear() # empty the deque |
| 428 | >>> d.pop() # cannot pop from an empty deque |
| 429 | Traceback (most recent call last): |
| 430 | File "<pyshell#6>", line 1, in -toplevel- |
| 431 | d.pop() |
| 432 | IndexError: pop from an empty deque |
| 433 | |
| 434 | >>> d.extendleft('abc') # extendleft() reverses the input order |
| 435 | >>> d |
| 436 | deque(['c', 'b', 'a']) |
| 437 | |
| 438 | |
Georg Brandl | 9afde1c | 2007-11-01 20:32:30 +0000 | [diff] [blame] | 439 | :class:`deque` Recipes |
| 440 | ^^^^^^^^^^^^^^^^^^^^^^ |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 441 | |
| 442 | This section shows various approaches to working with deques. |
| 443 | |
Raymond Hettinger | d2ee64d | 2009-03-31 22:52:48 +0000 | [diff] [blame] | 444 | Bounded length deques provide functionality similar to the ``tail`` filter |
| 445 | in Unix:: |
| 446 | |
| 447 | def tail(filename, n=10): |
| 448 | 'Return the last n lines of a file' |
| 449 | return deque(open(filename), n) |
| 450 | |
| 451 | Another approach to using deques is to maintain a sequence of recently |
| 452 | added elements by appending to the right and popping to the left:: |
| 453 | |
| 454 | def moving_average(iterable, n=3): |
| 455 | # moving_average([40, 30, 50, 46, 39, 44]) --> 40.0 42.0 45.0 43.0 |
| 456 | # http://en.wikipedia.org/wiki/Moving_average |
| 457 | it = iter(iterable) |
Raymond Hettinger | d40285a | 2009-05-22 01:11:26 +0000 | [diff] [blame] | 458 | d = deque(itertools.islice(it, n-1)) |
| 459 | d.appendleft(0) |
Raymond Hettinger | d2ee64d | 2009-03-31 22:52:48 +0000 | [diff] [blame] | 460 | s = sum(d) |
Raymond Hettinger | d2ee64d | 2009-03-31 22:52:48 +0000 | [diff] [blame] | 461 | for elem in it: |
| 462 | s += elem - d.popleft() |
| 463 | d.append(elem) |
| 464 | yield s / n |
| 465 | |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 466 | The :meth:`rotate` method provides a way to implement :class:`deque` slicing and |
| 467 | deletion. For example, a pure python implementation of ``del d[n]`` relies on |
| 468 | the :meth:`rotate` method to position elements to be popped:: |
| 469 | |
| 470 | def delete_nth(d, n): |
| 471 | d.rotate(-n) |
| 472 | d.popleft() |
| 473 | d.rotate(n) |
| 474 | |
| 475 | To implement :class:`deque` slicing, use a similar approach applying |
| 476 | :meth:`rotate` to bring a target element to the left side of the deque. Remove |
| 477 | old entries with :meth:`popleft`, add new entries with :meth:`extend`, and then |
| 478 | reverse the rotation. |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 479 | With minor variations on that approach, it is easy to implement Forth style |
| 480 | stack manipulations such as ``dup``, ``drop``, ``swap``, ``over``, ``pick``, |
| 481 | ``rot``, and ``roll``. |
| 482 | |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 483 | |
| 484 | :class:`defaultdict` objects |
| 485 | ---------------------------- |
| 486 | |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 487 | .. class:: defaultdict([default_factory[, ...]]) |
| 488 | |
| 489 | Returns a new dictionary-like object. :class:`defaultdict` is a subclass of the |
| 490 | builtin :class:`dict` class. It overrides one method and adds one writable |
| 491 | instance variable. The remaining functionality is the same as for the |
| 492 | :class:`dict` class and is not documented here. |
| 493 | |
| 494 | The first argument provides the initial value for the :attr:`default_factory` |
| 495 | attribute; it defaults to ``None``. All remaining arguments are treated the same |
| 496 | as if they were passed to the :class:`dict` constructor, including keyword |
| 497 | arguments. |
| 498 | |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 499 | |
Benjamin Peterson | e41251e | 2008-04-25 01:59:09 +0000 | [diff] [blame] | 500 | :class:`defaultdict` objects support the following method in addition to the |
| 501 | standard :class:`dict` operations: |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 502 | |
Benjamin Peterson | e41251e | 2008-04-25 01:59:09 +0000 | [diff] [blame] | 503 | .. method:: defaultdict.__missing__(key) |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 504 | |
Benjamin Peterson | 5478b47 | 2008-09-17 22:25:09 +0000 | [diff] [blame] | 505 | If the :attr:`default_factory` attribute is ``None``, this raises a |
Benjamin Peterson | e41251e | 2008-04-25 01:59:09 +0000 | [diff] [blame] | 506 | :exc:`KeyError` exception with the *key* as argument. |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 507 | |
Benjamin Peterson | e41251e | 2008-04-25 01:59:09 +0000 | [diff] [blame] | 508 | If :attr:`default_factory` is not ``None``, it is called without arguments |
| 509 | to provide a default value for the given *key*, this value is inserted in |
| 510 | the dictionary for the *key*, and returned. |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 511 | |
Benjamin Peterson | e41251e | 2008-04-25 01:59:09 +0000 | [diff] [blame] | 512 | If calling :attr:`default_factory` raises an exception this exception is |
| 513 | propagated unchanged. |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 514 | |
Benjamin Peterson | e41251e | 2008-04-25 01:59:09 +0000 | [diff] [blame] | 515 | This method is called by the :meth:`__getitem__` method of the |
| 516 | :class:`dict` class when the requested key is not found; whatever it |
| 517 | returns or raises is then returned or raised by :meth:`__getitem__`. |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 518 | |
| 519 | |
Benjamin Peterson | e41251e | 2008-04-25 01:59:09 +0000 | [diff] [blame] | 520 | :class:`defaultdict` objects support the following instance variable: |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 521 | |
Benjamin Peterson | e41251e | 2008-04-25 01:59:09 +0000 | [diff] [blame] | 522 | |
| 523 | .. attribute:: defaultdict.default_factory |
| 524 | |
| 525 | This attribute is used by the :meth:`__missing__` method; it is |
| 526 | initialized from the first argument to the constructor, if present, or to |
| 527 | ``None``, if absent. |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 528 | |
| 529 | |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 530 | :class:`defaultdict` Examples |
| 531 | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ |
| 532 | |
| 533 | Using :class:`list` as the :attr:`default_factory`, it is easy to group a |
Christian Heimes | fe337bf | 2008-03-23 21:54:12 +0000 | [diff] [blame] | 534 | sequence of key-value pairs into a dictionary of lists: |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 535 | |
| 536 | >>> s = [('yellow', 1), ('blue', 2), ('yellow', 3), ('blue', 4), ('red', 1)] |
| 537 | >>> d = defaultdict(list) |
| 538 | >>> for k, v in s: |
| 539 | ... d[k].append(v) |
| 540 | ... |
| 541 | >>> d.items() |
| 542 | [('blue', [2, 4]), ('red', [1]), ('yellow', [1, 3])] |
| 543 | |
| 544 | When each key is encountered for the first time, it is not already in the |
| 545 | mapping; so an entry is automatically created using the :attr:`default_factory` |
| 546 | function which returns an empty :class:`list`. The :meth:`list.append` |
| 547 | operation then attaches the value to the new list. When keys are encountered |
| 548 | again, the look-up proceeds normally (returning the list for that key) and the |
| 549 | :meth:`list.append` operation adds another value to the list. This technique is |
Christian Heimes | fe337bf | 2008-03-23 21:54:12 +0000 | [diff] [blame] | 550 | simpler and faster than an equivalent technique using :meth:`dict.setdefault`: |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 551 | |
| 552 | >>> d = {} |
| 553 | >>> for k, v in s: |
| 554 | ... d.setdefault(k, []).append(v) |
| 555 | ... |
| 556 | >>> d.items() |
| 557 | [('blue', [2, 4]), ('red', [1]), ('yellow', [1, 3])] |
| 558 | |
| 559 | Setting the :attr:`default_factory` to :class:`int` makes the |
| 560 | :class:`defaultdict` useful for counting (like a bag or multiset in other |
Christian Heimes | fe337bf | 2008-03-23 21:54:12 +0000 | [diff] [blame] | 561 | languages): |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 562 | |
| 563 | >>> s = 'mississippi' |
| 564 | >>> d = defaultdict(int) |
| 565 | >>> for k in s: |
| 566 | ... d[k] += 1 |
| 567 | ... |
| 568 | >>> d.items() |
| 569 | [('i', 4), ('p', 2), ('s', 4), ('m', 1)] |
| 570 | |
| 571 | When a letter is first encountered, it is missing from the mapping, so the |
| 572 | :attr:`default_factory` function calls :func:`int` to supply a default count of |
| 573 | zero. The increment operation then builds up the count for each letter. |
| 574 | |
| 575 | The function :func:`int` which always returns zero is just a special case of |
| 576 | constant functions. A faster and more flexible way to create constant functions |
| 577 | is to use a lambda function which can supply any constant value (not just |
Christian Heimes | fe337bf | 2008-03-23 21:54:12 +0000 | [diff] [blame] | 578 | zero): |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 579 | |
| 580 | >>> def constant_factory(value): |
| 581 | ... return lambda: value |
| 582 | >>> d = defaultdict(constant_factory('<missing>')) |
| 583 | >>> d.update(name='John', action='ran') |
| 584 | >>> '%(name)s %(action)s to %(object)s' % d |
| 585 | 'John ran to <missing>' |
| 586 | |
| 587 | Setting the :attr:`default_factory` to :class:`set` makes the |
Christian Heimes | fe337bf | 2008-03-23 21:54:12 +0000 | [diff] [blame] | 588 | :class:`defaultdict` useful for building a dictionary of sets: |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 589 | |
| 590 | >>> s = [('red', 1), ('blue', 2), ('red', 3), ('blue', 4), ('red', 1), ('blue', 4)] |
| 591 | >>> d = defaultdict(set) |
| 592 | >>> for k, v in s: |
| 593 | ... d[k].add(v) |
| 594 | ... |
| 595 | >>> d.items() |
| 596 | [('blue', set([2, 4])), ('red', set([1, 3]))] |
| 597 | |
| 598 | |
Guido van Rossum | 3d392eb | 2007-11-16 00:35:22 +0000 | [diff] [blame] | 599 | :func:`namedtuple` Factory Function for Tuples with Named Fields |
Christian Heimes | 790c823 | 2008-01-07 21:14:23 +0000 | [diff] [blame] | 600 | ---------------------------------------------------------------- |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 601 | |
Thomas Wouters | 1b7f891 | 2007-09-19 03:06:30 +0000 | [diff] [blame] | 602 | Named tuples assign meaning to each position in a tuple and allow for more readable, |
| 603 | self-documenting code. They can be used wherever regular tuples are used, and |
| 604 | they add the ability to access fields by name instead of position index. |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 605 | |
Georg Brandl | c2a4f4f | 2009-04-10 09:03:43 +0000 | [diff] [blame] | 606 | .. function:: namedtuple(typename, field_names, verbose=False, rename=False) |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 607 | |
| 608 | Returns a new tuple subclass named *typename*. The new subclass is used to |
Christian Heimes | c3f30c4 | 2008-02-22 16:37:40 +0000 | [diff] [blame] | 609 | create tuple-like objects that have fields accessible by attribute lookup as |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 610 | well as being indexable and iterable. Instances of the subclass also have a |
Benjamin Peterson | 4469d0c | 2008-11-30 22:46:23 +0000 | [diff] [blame] | 611 | helpful docstring (with typename and field_names) and a helpful :meth:`__repr__` |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 612 | method which lists the tuple contents in a ``name=value`` format. |
| 613 | |
Benjamin Peterson | 4469d0c | 2008-11-30 22:46:23 +0000 | [diff] [blame] | 614 | The *field_names* are a single string with each fieldname separated by whitespace |
| 615 | and/or commas, for example ``'x y'`` or ``'x, y'``. Alternatively, *field_names* |
Christian Heimes | 25bb783 | 2008-01-11 16:17:00 +0000 | [diff] [blame] | 616 | can be a sequence of strings such as ``['x', 'y']``. |
Georg Brandl | 9afde1c | 2007-11-01 20:32:30 +0000 | [diff] [blame] | 617 | |
| 618 | Any valid Python identifier may be used for a fieldname except for names |
Christian Heimes | 0449f63 | 2007-12-15 01:27:15 +0000 | [diff] [blame] | 619 | starting with an underscore. Valid identifiers consist of letters, digits, |
| 620 | and underscores but do not start with a digit or underscore and cannot be |
Georg Brandl | f694518 | 2008-02-01 11:56:49 +0000 | [diff] [blame] | 621 | a :mod:`keyword` such as *class*, *for*, *return*, *global*, *pass*, |
Georg Brandl | 9afde1c | 2007-11-01 20:32:30 +0000 | [diff] [blame] | 622 | or *raise*. |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 623 | |
Benjamin Peterson | a86f2c0 | 2009-02-10 02:41:10 +0000 | [diff] [blame] | 624 | If *rename* is true, invalid fieldnames are automatically replaced |
| 625 | with positional names. For example, ``['abc', 'def', 'ghi', 'abc']`` is |
Raymond Hettinger | 85737b8 | 2009-04-02 22:37:59 +0000 | [diff] [blame] | 626 | converted to ``['abc', '_1', 'ghi', '_3']``, eliminating the keyword |
Benjamin Peterson | a86f2c0 | 2009-02-10 02:41:10 +0000 | [diff] [blame] | 627 | ``def`` and the duplicate fieldname ``abc``. |
| 628 | |
Christian Heimes | 25bb783 | 2008-01-11 16:17:00 +0000 | [diff] [blame] | 629 | If *verbose* is true, the class definition is printed just before being built. |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 630 | |
Georg Brandl | 9afde1c | 2007-11-01 20:32:30 +0000 | [diff] [blame] | 631 | Named tuple instances do not have per-instance dictionaries, so they are |
Thomas Wouters | 8ce81f7 | 2007-09-20 18:22:40 +0000 | [diff] [blame] | 632 | lightweight and require no more memory than regular tuples. |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 633 | |
Raymond Hettinger | b62ad24 | 2009-03-02 22:16:43 +0000 | [diff] [blame] | 634 | .. versionchanged:: 3.1 |
Benjamin Peterson | a86f2c0 | 2009-02-10 02:41:10 +0000 | [diff] [blame] | 635 | added support for *rename*. |
| 636 | |
Christian Heimes | fe337bf | 2008-03-23 21:54:12 +0000 | [diff] [blame] | 637 | Example: |
| 638 | |
| 639 | .. doctest:: |
| 640 | :options: +NORMALIZE_WHITESPACE |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 641 | |
Guido van Rossum | 3d392eb | 2007-11-16 00:35:22 +0000 | [diff] [blame] | 642 | >>> Point = namedtuple('Point', 'x y', verbose=True) |
Thomas Wouters | 1b7f891 | 2007-09-19 03:06:30 +0000 | [diff] [blame] | 643 | class Point(tuple): |
| 644 | 'Point(x, y)' |
Christian Heimes | fe337bf | 2008-03-23 21:54:12 +0000 | [diff] [blame] | 645 | <BLANKLINE> |
Thomas Wouters | 1b7f891 | 2007-09-19 03:06:30 +0000 | [diff] [blame] | 646 | __slots__ = () |
Christian Heimes | fe337bf | 2008-03-23 21:54:12 +0000 | [diff] [blame] | 647 | <BLANKLINE> |
Christian Heimes | faf2f63 | 2008-01-06 16:59:19 +0000 | [diff] [blame] | 648 | _fields = ('x', 'y') |
Christian Heimes | fe337bf | 2008-03-23 21:54:12 +0000 | [diff] [blame] | 649 | <BLANKLINE> |
Raymond Hettinger | 089ba7f | 2009-05-27 00:38:24 +0000 | [diff] [blame] | 650 | def __new__(_cls, x, y): |
| 651 | return _tuple.__new__(_cls, (x, y)) |
Christian Heimes | fe337bf | 2008-03-23 21:54:12 +0000 | [diff] [blame] | 652 | <BLANKLINE> |
Christian Heimes | faf2f63 | 2008-01-06 16:59:19 +0000 | [diff] [blame] | 653 | @classmethod |
Christian Heimes | fe337bf | 2008-03-23 21:54:12 +0000 | [diff] [blame] | 654 | def _make(cls, iterable, new=tuple.__new__, len=len): |
Christian Heimes | faf2f63 | 2008-01-06 16:59:19 +0000 | [diff] [blame] | 655 | 'Make a new Point object from a sequence or iterable' |
Christian Heimes | fe337bf | 2008-03-23 21:54:12 +0000 | [diff] [blame] | 656 | result = new(cls, iterable) |
Christian Heimes | faf2f63 | 2008-01-06 16:59:19 +0000 | [diff] [blame] | 657 | if len(result) != 2: |
| 658 | raise TypeError('Expected 2 arguments, got %d' % len(result)) |
| 659 | return result |
Christian Heimes | fe337bf | 2008-03-23 21:54:12 +0000 | [diff] [blame] | 660 | <BLANKLINE> |
Thomas Wouters | 1b7f891 | 2007-09-19 03:06:30 +0000 | [diff] [blame] | 661 | def __repr__(self): |
| 662 | return 'Point(x=%r, y=%r)' % self |
Christian Heimes | fe337bf | 2008-03-23 21:54:12 +0000 | [diff] [blame] | 663 | <BLANKLINE> |
Raymond Hettinger | a4f52b1 | 2009-03-02 22:28:31 +0000 | [diff] [blame] | 664 | def _asdict(self): |
| 665 | 'Return a new OrderedDict which maps field names to their values' |
| 666 | return OrderedDict(zip(self._fields, self)) |
Christian Heimes | fe337bf | 2008-03-23 21:54:12 +0000 | [diff] [blame] | 667 | <BLANKLINE> |
Raymond Hettinger | 089ba7f | 2009-05-27 00:38:24 +0000 | [diff] [blame] | 668 | def _replace(_self, **kwds): |
Guido van Rossum | 3d392eb | 2007-11-16 00:35:22 +0000 | [diff] [blame] | 669 | 'Return a new Point object replacing specified fields with new values' |
Raymond Hettinger | 089ba7f | 2009-05-27 00:38:24 +0000 | [diff] [blame] | 670 | result = _self._make(map(kwds.pop, ('x', 'y'), _self)) |
Christian Heimes | faf2f63 | 2008-01-06 16:59:19 +0000 | [diff] [blame] | 671 | if kwds: |
| 672 | raise ValueError('Got unexpected field names: %r' % kwds.keys()) |
| 673 | return result |
Georg Brandl | 48310cd | 2009-01-03 21:18:54 +0000 | [diff] [blame] | 674 | <BLANKLINE> |
| 675 | def __getnewargs__(self): |
Benjamin Peterson | 4118174 | 2008-07-02 20:22:54 +0000 | [diff] [blame] | 676 | return tuple(self) |
Christian Heimes | fe337bf | 2008-03-23 21:54:12 +0000 | [diff] [blame] | 677 | <BLANKLINE> |
Raymond Hettinger | 089ba7f | 2009-05-27 00:38:24 +0000 | [diff] [blame] | 678 | x = _property(_itemgetter(0)) |
| 679 | y = _property(_itemgetter(1)) |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 680 | |
Thomas Wouters | 1b7f891 | 2007-09-19 03:06:30 +0000 | [diff] [blame] | 681 | >>> p = Point(11, y=22) # instantiate with positional or keyword arguments |
Christian Heimes | 99170a5 | 2007-12-19 02:07:34 +0000 | [diff] [blame] | 682 | >>> p[0] + p[1] # indexable like the plain tuple (11, 22) |
Thomas Wouters | 1b7f891 | 2007-09-19 03:06:30 +0000 | [diff] [blame] | 683 | 33 |
| 684 | >>> x, y = p # unpack like a regular tuple |
| 685 | >>> x, y |
| 686 | (11, 22) |
Christian Heimes | c3f30c4 | 2008-02-22 16:37:40 +0000 | [diff] [blame] | 687 | >>> p.x + p.y # fields also accessible by name |
Thomas Wouters | 1b7f891 | 2007-09-19 03:06:30 +0000 | [diff] [blame] | 688 | 33 |
| 689 | >>> p # readable __repr__ with a name=value style |
| 690 | Point(x=11, y=22) |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 691 | |
Thomas Wouters | 1b7f891 | 2007-09-19 03:06:30 +0000 | [diff] [blame] | 692 | Named tuples are especially useful for assigning field names to result tuples returned |
| 693 | by the :mod:`csv` or :mod:`sqlite3` modules:: |
| 694 | |
Guido van Rossum | 3d392eb | 2007-11-16 00:35:22 +0000 | [diff] [blame] | 695 | EmployeeRecord = namedtuple('EmployeeRecord', 'name, age, title, department, paygrade') |
Georg Brandl | 9afde1c | 2007-11-01 20:32:30 +0000 | [diff] [blame] | 696 | |
Thomas Wouters | 1b7f891 | 2007-09-19 03:06:30 +0000 | [diff] [blame] | 697 | import csv |
Christian Heimes | faf2f63 | 2008-01-06 16:59:19 +0000 | [diff] [blame] | 698 | for emp in map(EmployeeRecord._make, csv.reader(open("employees.csv", "rb"))): |
Thomas Wouters | 1b7f891 | 2007-09-19 03:06:30 +0000 | [diff] [blame] | 699 | print(emp.name, emp.title) |
| 700 | |
Georg Brandl | 9afde1c | 2007-11-01 20:32:30 +0000 | [diff] [blame] | 701 | import sqlite3 |
| 702 | conn = sqlite3.connect('/companydata') |
| 703 | cursor = conn.cursor() |
| 704 | cursor.execute('SELECT name, age, title, department, paygrade FROM employees') |
Christian Heimes | faf2f63 | 2008-01-06 16:59:19 +0000 | [diff] [blame] | 705 | for emp in map(EmployeeRecord._make, cursor.fetchall()): |
Christian Heimes | 0041223 | 2008-01-10 16:02:19 +0000 | [diff] [blame] | 706 | print(emp.name, emp.title) |
Georg Brandl | 9afde1c | 2007-11-01 20:32:30 +0000 | [diff] [blame] | 707 | |
Christian Heimes | 99170a5 | 2007-12-19 02:07:34 +0000 | [diff] [blame] | 708 | In addition to the methods inherited from tuples, named tuples support |
Christian Heimes | 2380ac7 | 2008-01-09 00:17:24 +0000 | [diff] [blame] | 709 | three additional methods and one attribute. To prevent conflicts with |
| 710 | field names, the method and attribute names start with an underscore. |
Christian Heimes | 99170a5 | 2007-12-19 02:07:34 +0000 | [diff] [blame] | 711 | |
Christian Heimes | 790c823 | 2008-01-07 21:14:23 +0000 | [diff] [blame] | 712 | .. method:: somenamedtuple._make(iterable) |
Christian Heimes | 99170a5 | 2007-12-19 02:07:34 +0000 | [diff] [blame] | 713 | |
Christian Heimes | faf2f63 | 2008-01-06 16:59:19 +0000 | [diff] [blame] | 714 | Class method that makes a new instance from an existing sequence or iterable. |
Christian Heimes | 99170a5 | 2007-12-19 02:07:34 +0000 | [diff] [blame] | 715 | |
Christian Heimes | fe337bf | 2008-03-23 21:54:12 +0000 | [diff] [blame] | 716 | .. doctest:: |
Thomas Wouters | 1b7f891 | 2007-09-19 03:06:30 +0000 | [diff] [blame] | 717 | |
Christian Heimes | faf2f63 | 2008-01-06 16:59:19 +0000 | [diff] [blame] | 718 | >>> t = [11, 22] |
| 719 | >>> Point._make(t) |
| 720 | Point(x=11, y=22) |
Thomas Wouters | 1b7f891 | 2007-09-19 03:06:30 +0000 | [diff] [blame] | 721 | |
Christian Heimes | 790c823 | 2008-01-07 21:14:23 +0000 | [diff] [blame] | 722 | .. method:: somenamedtuple._asdict() |
Georg Brandl | 9afde1c | 2007-11-01 20:32:30 +0000 | [diff] [blame] | 723 | |
Raymond Hettinger | a4f52b1 | 2009-03-02 22:28:31 +0000 | [diff] [blame] | 724 | Return a new :class:`OrderedDict` which maps field names to their corresponding |
| 725 | values:: |
Georg Brandl | 9afde1c | 2007-11-01 20:32:30 +0000 | [diff] [blame] | 726 | |
Christian Heimes | 0449f63 | 2007-12-15 01:27:15 +0000 | [diff] [blame] | 727 | >>> p._asdict() |
Raymond Hettinger | a4f52b1 | 2009-03-02 22:28:31 +0000 | [diff] [blame] | 728 | OrderedDict([('x', 11), ('y', 22)]) |
| 729 | |
Raymond Hettinger | a88e4da | 2009-03-03 05:12:27 +0000 | [diff] [blame] | 730 | .. versionchanged:: 3.1 |
Raymond Hettinger | a4f52b1 | 2009-03-02 22:28:31 +0000 | [diff] [blame] | 731 | Returns an :class:`OrderedDict` instead of a regular :class:`dict`. |
Christian Heimes | fe337bf | 2008-03-23 21:54:12 +0000 | [diff] [blame] | 732 | |
Christian Heimes | 790c823 | 2008-01-07 21:14:23 +0000 | [diff] [blame] | 733 | .. method:: somenamedtuple._replace(kwargs) |
Thomas Wouters | 1b7f891 | 2007-09-19 03:06:30 +0000 | [diff] [blame] | 734 | |
Christian Heimes | fe337bf | 2008-03-23 21:54:12 +0000 | [diff] [blame] | 735 | Return a new instance of the named tuple replacing specified fields with new |
| 736 | values: |
Thomas Wouters | 8ce81f7 | 2007-09-20 18:22:40 +0000 | [diff] [blame] | 737 | |
| 738 | :: |
Thomas Wouters | 1b7f891 | 2007-09-19 03:06:30 +0000 | [diff] [blame] | 739 | |
| 740 | >>> p = Point(x=11, y=22) |
Christian Heimes | 0449f63 | 2007-12-15 01:27:15 +0000 | [diff] [blame] | 741 | >>> p._replace(x=33) |
Thomas Wouters | 1b7f891 | 2007-09-19 03:06:30 +0000 | [diff] [blame] | 742 | Point(x=33, y=22) |
| 743 | |
Guido van Rossum | 3d392eb | 2007-11-16 00:35:22 +0000 | [diff] [blame] | 744 | >>> for partnum, record in inventory.items(): |
Christian Heimes | 454f37b | 2008-01-10 00:10:02 +0000 | [diff] [blame] | 745 | ... inventory[partnum] = record._replace(price=newprices[partnum], timestamp=time.now()) |
Thomas Wouters | 1b7f891 | 2007-09-19 03:06:30 +0000 | [diff] [blame] | 746 | |
Christian Heimes | 790c823 | 2008-01-07 21:14:23 +0000 | [diff] [blame] | 747 | .. attribute:: somenamedtuple._fields |
Thomas Wouters | 1b7f891 | 2007-09-19 03:06:30 +0000 | [diff] [blame] | 748 | |
Christian Heimes | 2380ac7 | 2008-01-09 00:17:24 +0000 | [diff] [blame] | 749 | Tuple of strings listing the field names. Useful for introspection |
Georg Brandl | 9afde1c | 2007-11-01 20:32:30 +0000 | [diff] [blame] | 750 | and for creating new named tuple types from existing named tuples. |
Thomas Wouters | 8ce81f7 | 2007-09-20 18:22:40 +0000 | [diff] [blame] | 751 | |
Christian Heimes | fe337bf | 2008-03-23 21:54:12 +0000 | [diff] [blame] | 752 | .. doctest:: |
Thomas Wouters | 1b7f891 | 2007-09-19 03:06:30 +0000 | [diff] [blame] | 753 | |
Christian Heimes | 0449f63 | 2007-12-15 01:27:15 +0000 | [diff] [blame] | 754 | >>> p._fields # view the field names |
Thomas Wouters | 1b7f891 | 2007-09-19 03:06:30 +0000 | [diff] [blame] | 755 | ('x', 'y') |
Thomas Wouters | 1b7f891 | 2007-09-19 03:06:30 +0000 | [diff] [blame] | 756 | |
Guido van Rossum | 3d392eb | 2007-11-16 00:35:22 +0000 | [diff] [blame] | 757 | >>> Color = namedtuple('Color', 'red green blue') |
Christian Heimes | 0449f63 | 2007-12-15 01:27:15 +0000 | [diff] [blame] | 758 | >>> Pixel = namedtuple('Pixel', Point._fields + Color._fields) |
Thomas Wouters | 1b7f891 | 2007-09-19 03:06:30 +0000 | [diff] [blame] | 759 | >>> Pixel(11, 22, 128, 255, 0) |
Christian Heimes | 454f37b | 2008-01-10 00:10:02 +0000 | [diff] [blame] | 760 | Pixel(x=11, y=22, red=128, green=255, blue=0) |
Georg Brandl | 116aa62 | 2007-08-15 14:28:22 +0000 | [diff] [blame] | 761 | |
Christian Heimes | 0449f63 | 2007-12-15 01:27:15 +0000 | [diff] [blame] | 762 | To retrieve a field whose name is stored in a string, use the :func:`getattr` |
Christian Heimes | fe337bf | 2008-03-23 21:54:12 +0000 | [diff] [blame] | 763 | function: |
Christian Heimes | 0449f63 | 2007-12-15 01:27:15 +0000 | [diff] [blame] | 764 | |
| 765 | >>> getattr(p, 'x') |
| 766 | 11 |
| 767 | |
Raymond Hettinger | 651453a | 2009-02-11 00:20:02 +0000 | [diff] [blame] | 768 | To convert a dictionary to a named tuple, use the double-star-operator |
| 769 | (as described in :ref:`tut-unpacking-arguments`): |
Christian Heimes | 99170a5 | 2007-12-19 02:07:34 +0000 | [diff] [blame] | 770 | |
| 771 | >>> d = {'x': 11, 'y': 22} |
| 772 | >>> Point(**d) |
| 773 | Point(x=11, y=22) |
| 774 | |
Guido van Rossum | 3d392eb | 2007-11-16 00:35:22 +0000 | [diff] [blame] | 775 | Since a named tuple is a regular Python class, it is easy to add or change |
Christian Heimes | 043d6f6 | 2008-01-07 17:19:16 +0000 | [diff] [blame] | 776 | functionality with a subclass. Here is how to add a calculated field and |
Christian Heimes | fe337bf | 2008-03-23 21:54:12 +0000 | [diff] [blame] | 777 | a fixed-width print format: |
Guido van Rossum | 3d392eb | 2007-11-16 00:35:22 +0000 | [diff] [blame] | 778 | |
Christian Heimes | 043d6f6 | 2008-01-07 17:19:16 +0000 | [diff] [blame] | 779 | >>> class Point(namedtuple('Point', 'x y')): |
Christian Heimes | 25bb783 | 2008-01-11 16:17:00 +0000 | [diff] [blame] | 780 | ... __slots__ = () |
Christian Heimes | 454f37b | 2008-01-10 00:10:02 +0000 | [diff] [blame] | 781 | ... @property |
| 782 | ... def hypot(self): |
| 783 | ... return (self.x ** 2 + self.y ** 2) ** 0.5 |
| 784 | ... def __str__(self): |
Christian Heimes | 25bb783 | 2008-01-11 16:17:00 +0000 | [diff] [blame] | 785 | ... return 'Point: x=%6.3f y=%6.3f hypot=%6.3f' % (self.x, self.y, self.hypot) |
Guido van Rossum | 3d392eb | 2007-11-16 00:35:22 +0000 | [diff] [blame] | 786 | |
Georg Brandl | 0df7979 | 2008-10-04 18:33:26 +0000 | [diff] [blame] | 787 | >>> for p in Point(3, 4), Point(14, 5/7): |
Christian Heimes | 0041223 | 2008-01-10 16:02:19 +0000 | [diff] [blame] | 788 | ... print(p) |
Christian Heimes | 25bb783 | 2008-01-11 16:17:00 +0000 | [diff] [blame] | 789 | Point: x= 3.000 y= 4.000 hypot= 5.000 |
| 790 | Point: x=14.000 y= 0.714 hypot=14.018 |
Christian Heimes | 043d6f6 | 2008-01-07 17:19:16 +0000 | [diff] [blame] | 791 | |
Christian Heimes | af98da1 | 2008-01-27 15:18:18 +0000 | [diff] [blame] | 792 | The subclass shown above sets ``__slots__`` to an empty tuple. This keeps |
Christian Heimes | 679db4a | 2008-01-18 09:56:22 +0000 | [diff] [blame] | 793 | keep memory requirements low by preventing the creation of instance dictionaries. |
| 794 | |
Christian Heimes | 2380ac7 | 2008-01-09 00:17:24 +0000 | [diff] [blame] | 795 | |
| 796 | Subclassing is not useful for adding new, stored fields. Instead, simply |
Christian Heimes | fe337bf | 2008-03-23 21:54:12 +0000 | [diff] [blame] | 797 | create a new named tuple type from the :attr:`_fields` attribute: |
Christian Heimes | 2380ac7 | 2008-01-09 00:17:24 +0000 | [diff] [blame] | 798 | |
Christian Heimes | 25bb783 | 2008-01-11 16:17:00 +0000 | [diff] [blame] | 799 | >>> Point3D = namedtuple('Point3D', Point._fields + ('z',)) |
Christian Heimes | 2380ac7 | 2008-01-09 00:17:24 +0000 | [diff] [blame] | 800 | |
| 801 | Default values can be implemented by using :meth:`_replace` to |
Christian Heimes | fe337bf | 2008-03-23 21:54:12 +0000 | [diff] [blame] | 802 | customize a prototype instance: |
Guido van Rossum | 3d392eb | 2007-11-16 00:35:22 +0000 | [diff] [blame] | 803 | |
| 804 | >>> Account = namedtuple('Account', 'owner balance transaction_count') |
Christian Heimes | 587c2bf | 2008-01-19 16:21:02 +0000 | [diff] [blame] | 805 | >>> default_account = Account('<owner name>', 0.0, 0) |
| 806 | >>> johns_account = default_account._replace(owner='John') |
Guido van Rossum | 3d392eb | 2007-11-16 00:35:22 +0000 | [diff] [blame] | 807 | |
Christian Heimes | e4ca815 | 2008-05-08 17:18:53 +0000 | [diff] [blame] | 808 | Enumerated constants can be implemented with named tuples, but it is simpler |
| 809 | and more efficient to use a simple class declaration: |
| 810 | |
| 811 | >>> Status = namedtuple('Status', 'open pending closed')._make(range(3)) |
| 812 | >>> Status.open, Status.pending, Status.closed |
| 813 | (0, 1, 2) |
| 814 | >>> class Status: |
| 815 | ... open, pending, closed = range(3) |
| 816 | |
Raymond Hettinger | 651453a | 2009-02-11 00:20:02 +0000 | [diff] [blame] | 817 | .. seealso:: |
Thomas Wouters | 47b49bf | 2007-08-30 22:15:33 +0000 | [diff] [blame] | 818 | |
Raymond Hettinger | 651453a | 2009-02-11 00:20:02 +0000 | [diff] [blame] | 819 | `Named tuple recipe <http://code.activestate.com/recipes/500261/>`_ |
| 820 | adapted for Python 2.4. |
Raymond Hettinger | e4c96ad | 2008-02-06 01:23:58 +0000 | [diff] [blame] | 821 | |
| 822 | |
Raymond Hettinger | 2d32f63 | 2009-03-02 21:24:57 +0000 | [diff] [blame] | 823 | :class:`OrderedDict` objects |
| 824 | ---------------------------- |
| 825 | |
| 826 | Ordered dictionaries are just like regular dictionaries but they remember the |
| 827 | order that items were inserted. When iterating over an ordered dictionary, |
| 828 | the items are returned in the order their keys were first added. |
| 829 | |
| 830 | .. class:: OrderedDict([items]) |
| 831 | |
| 832 | Return an instance of a dict subclass, supporting the usual :class:`dict` |
| 833 | methods. An *OrderedDict* is a dict that remembers the order that keys |
| 834 | were first inserted. If a new entry overwrites an existing entry, the |
| 835 | original insertion position is left unchanged. Deleting an entry and |
| 836 | reinserting it will move it to the end. |
| 837 | |
Benjamin Peterson | d45bf58 | 2009-03-02 21:44:54 +0000 | [diff] [blame] | 838 | .. versionadded:: 3.1 |
Raymond Hettinger | 2d32f63 | 2009-03-02 21:24:57 +0000 | [diff] [blame] | 839 | |
Raymond Hettinger | dc879f0 | 2009-03-19 20:30:56 +0000 | [diff] [blame] | 840 | .. method:: OrderedDict.popitem(last=True) |
| 841 | |
| 842 | The :meth:`popitem` method for ordered dictionaries returns and removes |
| 843 | a (key, value) pair. The pairs are returned in LIFO order if *last* is |
| 844 | true or FIFO order if false. |
Raymond Hettinger | 2d32f63 | 2009-03-02 21:24:57 +0000 | [diff] [blame] | 845 | |
Raymond Hettinger | e909150 | 2009-05-19 17:40:07 +0000 | [diff] [blame] | 846 | In addition to the usual mapping methods, ordered dictionaries also support |
| 847 | reverse iteration using :func:`reversed`. |
| 848 | |
Raymond Hettinger | 2d32f63 | 2009-03-02 21:24:57 +0000 | [diff] [blame] | 849 | Equality tests between :class:`OrderedDict` objects are order-sensitive |
| 850 | and are implemented as ``list(od1.items())==list(od2.items())``. |
| 851 | Equality tests between :class:`OrderedDict` objects and other |
| 852 | :class:`Mapping` objects are order-insensitive like regular dictionaries. |
| 853 | This allows :class:`OrderedDict` objects to be substituted anywhere a |
| 854 | regular dictionary is used. |
| 855 | |
Raymond Hettinger | 3618078 | 2009-04-09 22:34:23 +0000 | [diff] [blame] | 856 | The :class:`OrderedDict` constructor and :meth:`update` method both accept |
| 857 | keyword arguments, but their order is lost because Python's function call |
| 858 | semantics pass-in keyword arguments using a regular unordered dictionary. |
| 859 | |
Raymond Hettinger | dc879f0 | 2009-03-19 20:30:56 +0000 | [diff] [blame] | 860 | .. seealso:: |
| 861 | |
| 862 | `Equivalent OrderedDict recipe <http://code.activestate.com/recipes/576693/>`_ |
| 863 | that runs on Python 2.4 or later. |
| 864 | |
Raymond Hettinger | e4c96ad | 2008-02-06 01:23:58 +0000 | [diff] [blame] | 865 | |
| 866 | :class:`UserDict` objects |
Mark Summerfield | 8f2d006 | 2008-02-06 13:30:44 +0000 | [diff] [blame] | 867 | ------------------------- |
Raymond Hettinger | e4c96ad | 2008-02-06 01:23:58 +0000 | [diff] [blame] | 868 | |
Georg Brandl | 48310cd | 2009-01-03 21:18:54 +0000 | [diff] [blame] | 869 | The class, :class:`UserDict` acts as a wrapper around dictionary objects. |
| 870 | The need for this class has been partially supplanted by the ability to |
Raymond Hettinger | e4c96ad | 2008-02-06 01:23:58 +0000 | [diff] [blame] | 871 | subclass directly from :class:`dict`; however, this class can be easier |
| 872 | to work with because the underlying dictionary is accessible as an |
| 873 | attribute. |
| 874 | |
| 875 | .. class:: UserDict([initialdata]) |
| 876 | |
| 877 | Class that simulates a dictionary. The instance's contents are kept in a |
| 878 | regular dictionary, which is accessible via the :attr:`data` attribute of |
| 879 | :class:`UserDict` instances. If *initialdata* is provided, :attr:`data` is |
| 880 | initialized with its contents; note that a reference to *initialdata* will not |
| 881 | be kept, allowing it be used for other purposes. |
| 882 | |
Georg Brandl | 48310cd | 2009-01-03 21:18:54 +0000 | [diff] [blame] | 883 | In addition to supporting the methods and operations of mappings, |
Raymond Hettinger | ebcee3f | 2008-02-06 19:54:00 +0000 | [diff] [blame] | 884 | :class:`UserDict` instances provide the following attribute: |
Raymond Hettinger | e4c96ad | 2008-02-06 01:23:58 +0000 | [diff] [blame] | 885 | |
| 886 | .. attribute:: UserDict.data |
| 887 | |
| 888 | A real dictionary used to store the contents of the :class:`UserDict` class. |
Raymond Hettinger | 53dbe39 | 2008-02-12 20:03:09 +0000 | [diff] [blame] | 889 | |
| 890 | |
| 891 | |
| 892 | :class:`UserList` objects |
| 893 | ------------------------- |
| 894 | |
| 895 | This class acts as a wrapper around list objects. It is a useful base class |
Georg Brandl | 48310cd | 2009-01-03 21:18:54 +0000 | [diff] [blame] | 896 | for your own list-like classes which can inherit from them and override |
Raymond Hettinger | 53dbe39 | 2008-02-12 20:03:09 +0000 | [diff] [blame] | 897 | existing methods or add new ones. In this way, one can add new behaviors to |
| 898 | lists. |
| 899 | |
Georg Brandl | 48310cd | 2009-01-03 21:18:54 +0000 | [diff] [blame] | 900 | The need for this class has been partially supplanted by the ability to |
Raymond Hettinger | 53dbe39 | 2008-02-12 20:03:09 +0000 | [diff] [blame] | 901 | subclass directly from :class:`list`; however, this class can be easier |
| 902 | to work with because the underlying list is accessible as an attribute. |
| 903 | |
| 904 | .. class:: UserList([list]) |
| 905 | |
| 906 | Class that simulates a list. The instance's contents are kept in a regular |
| 907 | list, which is accessible via the :attr:`data` attribute of :class:`UserList` |
| 908 | instances. The instance's contents are initially set to a copy of *list*, |
| 909 | defaulting to the empty list ``[]``. *list* can be any iterable, for |
| 910 | example a real Python list or a :class:`UserList` object. |
| 911 | |
Georg Brandl | 48310cd | 2009-01-03 21:18:54 +0000 | [diff] [blame] | 912 | In addition to supporting the methods and operations of mutable sequences, |
Raymond Hettinger | 53dbe39 | 2008-02-12 20:03:09 +0000 | [diff] [blame] | 913 | :class:`UserList` instances provide the following attribute: |
| 914 | |
| 915 | .. attribute:: UserList.data |
| 916 | |
Georg Brandl | 48310cd | 2009-01-03 21:18:54 +0000 | [diff] [blame] | 917 | A real :class:`list` object used to store the contents of the |
Raymond Hettinger | 53dbe39 | 2008-02-12 20:03:09 +0000 | [diff] [blame] | 918 | :class:`UserList` class. |
| 919 | |
| 920 | **Subclassing requirements:** Subclasses of :class:`UserList` are expect to |
| 921 | offer a constructor which can be called with either no arguments or one |
| 922 | argument. List operations which return a new sequence attempt to create an |
| 923 | instance of the actual implementation class. To do so, it assumes that the |
| 924 | constructor can be called with a single parameter, which is a sequence object |
| 925 | used as a data source. |
| 926 | |
| 927 | If a derived class does not wish to comply with this requirement, all of the |
| 928 | special methods supported by this class will need to be overridden; please |
| 929 | consult the sources for information about the methods which need to be provided |
| 930 | in that case. |
Raymond Hettinger | b3a65f8 | 2008-02-21 22:11:37 +0000 | [diff] [blame] | 931 | |
| 932 | :class:`UserString` objects |
Christian Heimes | c3f30c4 | 2008-02-22 16:37:40 +0000 | [diff] [blame] | 933 | --------------------------- |
Raymond Hettinger | b3a65f8 | 2008-02-21 22:11:37 +0000 | [diff] [blame] | 934 | |
Georg Brandl | 48310cd | 2009-01-03 21:18:54 +0000 | [diff] [blame] | 935 | The class, :class:`UserString` acts as a wrapper around string objects. |
| 936 | The need for this class has been partially supplanted by the ability to |
Raymond Hettinger | b3a65f8 | 2008-02-21 22:11:37 +0000 | [diff] [blame] | 937 | subclass directly from :class:`str`; however, this class can be easier |
| 938 | to work with because the underlying string is accessible as an |
| 939 | attribute. |
| 940 | |
| 941 | .. class:: UserString([sequence]) |
| 942 | |
| 943 | Class that simulates a string or a Unicode string object. The instance's |
Georg Brandl | 48310cd | 2009-01-03 21:18:54 +0000 | [diff] [blame] | 944 | content is kept in a regular string object, which is accessible via the |
| 945 | :attr:`data` attribute of :class:`UserString` instances. The instance's |
Raymond Hettinger | b3a65f8 | 2008-02-21 22:11:37 +0000 | [diff] [blame] | 946 | contents are initially set to a copy of *sequence*. The *sequence* can |
| 947 | be an instance of :class:`bytes`, :class:`str`, :class:`UserString` (or a |
| 948 | subclass) or an arbitrary sequence which can be converted into a string using |
| 949 | the built-in :func:`str` function. |