blob: a09d3cf7255d3eaf1ad5b89fe222385c019f88d5 [file] [log] [blame]
Georg Brandl8ec7f652007-08-15 14:28:01 +00001:mod:`functools` --- Higher order functions and operations on callable objects
2==============================================================================
3
4.. module:: functools
5 :synopsis: Higher order functions and operations on callable objects.
6.. moduleauthor:: Peter Harris <scav@blueyonder.co.uk>
7.. moduleauthor:: Raymond Hettinger <python@rcn.com>
8.. moduleauthor:: Nick Coghlan <ncoghlan@gmail.com>
9.. sectionauthor:: Peter Harris <scav@blueyonder.co.uk>
10
11
12.. versionadded:: 2.5
13
14The :mod:`functools` module is for higher-order functions: functions that act on
15or return other functions. In general, any callable object can be treated as a
16function for the purposes of this module.
17
Georg Brandlae0ee8a2007-08-28 08:29:08 +000018The :mod:`functools` module defines the following functions:
19
20
21.. function:: reduce(function, iterable[, initializer])
22
23 This is the same function as :func:`reduce`. It is made available in this module
24 to allow writing code more forward-compatible with Python 3.
25
26 .. versionadded:: 2.6
Georg Brandl8ec7f652007-08-15 14:28:01 +000027
28
29.. function:: partial(func[,*args][, **keywords])
30
31 Return a new :class:`partial` object which when called will behave like *func*
32 called with the positional arguments *args* and keyword arguments *keywords*. If
33 more arguments are supplied to the call, they are appended to *args*. If
34 additional keyword arguments are supplied, they extend and override *keywords*.
35 Roughly equivalent to::
36
37 def partial(func, *args, **keywords):
38 def newfunc(*fargs, **fkeywords):
39 newkeywords = keywords.copy()
40 newkeywords.update(fkeywords)
41 return func(*(args + fargs), **newkeywords)
42 newfunc.func = func
43 newfunc.args = args
44 newfunc.keywords = keywords
45 return newfunc
46
47 The :func:`partial` is used for partial function application which "freezes"
48 some portion of a function's arguments and/or keywords resulting in a new object
49 with a simplified signature. For example, :func:`partial` can be used to create
50 a callable that behaves like the :func:`int` function where the *base* argument
Georg Brandle8f1b002008-03-22 22:04:10 +000051 defaults to two:
Georg Brandl8ec7f652007-08-15 14:28:01 +000052
Georg Brandle8f1b002008-03-22 22:04:10 +000053 >>> from functools import partial
Georg Brandl8ec7f652007-08-15 14:28:01 +000054 >>> basetwo = partial(int, base=2)
55 >>> basetwo.__doc__ = 'Convert base 2 string to an int.'
56 >>> basetwo('10010')
57 18
58
59
60.. function:: update_wrapper(wrapper, wrapped[, assigned][, updated])
61
62 Update a *wrapper* function to look like the *wrapped* function. The optional
63 arguments are tuples to specify which attributes of the original function are
64 assigned directly to the matching attributes on the wrapper function and which
65 attributes of the wrapper function are updated with the corresponding attributes
66 from the original function. The default values for these arguments are the
67 module level constants *WRAPPER_ASSIGNMENTS* (which assigns to the wrapper
68 function's *__name__*, *__module__* and *__doc__*, the documentation string) and
69 *WRAPPER_UPDATES* (which updates the wrapper function's *__dict__*, i.e. the
70 instance dictionary).
71
Georg Brandl584265b2007-12-02 14:58:50 +000072 The main intended use for this function is in :term:`decorator` functions which
73 wrap the decorated function and return the wrapper. If the wrapper function is
74 not updated, the metadata of the returned function will reflect the wrapper
Georg Brandl8ec7f652007-08-15 14:28:01 +000075 definition rather than the original function definition, which is typically less
76 than helpful.
77
78
79.. function:: wraps(wrapped[, assigned][, updated])
80
81 This is a convenience function for invoking ``partial(update_wrapper,
82 wrapped=wrapped, assigned=assigned, updated=updated)`` as a function decorator
Georg Brandle8f1b002008-03-22 22:04:10 +000083 when defining a wrapper function. For example:
Georg Brandl8ec7f652007-08-15 14:28:01 +000084
Georg Brandle8f1b002008-03-22 22:04:10 +000085 >>> from functools import wraps
Georg Brandl8ec7f652007-08-15 14:28:01 +000086 >>> def my_decorator(f):
87 ... @wraps(f)
88 ... def wrapper(*args, **kwds):
89 ... print 'Calling decorated function'
90 ... return f(*args, **kwds)
91 ... return wrapper
92 ...
93 >>> @my_decorator
94 ... def example():
95 ... """Docstring"""
96 ... print 'Called example function'
97 ...
98 >>> example()
99 Calling decorated function
100 Called example function
101 >>> example.__name__
102 'example'
103 >>> example.__doc__
104 'Docstring'
105
106 Without the use of this decorator factory, the name of the example function
107 would have been ``'wrapper'``, and the docstring of the original :func:`example`
108 would have been lost.
109
110
111.. _partial-objects:
112
113:class:`partial` Objects
114------------------------
115
116:class:`partial` objects are callable objects created by :func:`partial`. They
117have three read-only attributes:
118
119
120.. attribute:: partial.func
121
122 A callable object or function. Calls to the :class:`partial` object will be
123 forwarded to :attr:`func` with new arguments and keywords.
124
125
126.. attribute:: partial.args
127
128 The leftmost positional arguments that will be prepended to the positional
129 arguments provided to a :class:`partial` object call.
130
131
132.. attribute:: partial.keywords
133
134 The keyword arguments that will be supplied when the :class:`partial` object is
135 called.
136
137:class:`partial` objects are like :class:`function` objects in that they are
138callable, weak referencable, and can have attributes. There are some important
139differences. For instance, the :attr:`__name__` and :attr:`__doc__` attributes
140are not created automatically. Also, :class:`partial` objects defined in
141classes behave like static methods and do not transform into bound methods
142during instance attribute look-up.
143