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
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
|
<span id="abc-abstract-base-classes"></span><h1>abc — Abstract Base Classes</h1> <p><strong>Source code:</strong> <a class="reference external" href="https://github.com/python/cpython/tree/3.12/Lib/abc.py">Lib/abc.py</a></p> <p>This module provides the infrastructure for defining <a class="reference internal" href="../glossary#term-abstract-base-class"><span class="xref std std-term">abstract base classes</span></a> (ABCs) in Python, as outlined in <span class="target" id="index-0"></span><a class="pep reference external" href="https://peps.python.org/pep-3119/"><strong>PEP 3119</strong></a>; see the PEP for why this was added to Python. (See also <span class="target" id="index-1"></span><a class="pep reference external" href="https://peps.python.org/pep-3141/"><strong>PEP 3141</strong></a> and the <a class="reference internal" href="numbers#module-numbers" title="numbers: Numeric abstract base classes (Complex, Real, Integral, etc.)."><code>numbers</code></a> module regarding a type hierarchy for numbers based on ABCs.)</p> <p>The <a class="reference internal" href="collections#module-collections" title="collections: Container datatypes"><code>collections</code></a> module has some concrete classes that derive from ABCs; these can, of course, be further derived. In addition, the <a class="reference internal" href="collections.abc#module-collections.abc" title="collections.abc: Abstract base classes for containers"><code>collections.abc</code></a> submodule has some ABCs that can be used to test whether a class or instance provides a particular interface, for example, if it is <a class="reference internal" href="../glossary#term-hashable"><span class="xref std std-term">hashable</span></a> or if it is a <a class="reference internal" href="../glossary#term-mapping"><span class="xref std std-term">mapping</span></a>.</p> <p>This module provides the metaclass <a class="reference internal" href="#abc.ABCMeta" title="abc.ABCMeta"><code>ABCMeta</code></a> for defining ABCs and a helper class <a class="reference internal" href="#abc.ABC" title="abc.ABC"><code>ABC</code></a> to alternatively define ABCs through inheritance:</p> <dl class="py class"> <dt class="sig sig-object py" id="abc.ABC">
<code>class abc.ABC</code> </dt> <dd>
<p>A helper class that has <a class="reference internal" href="#abc.ABCMeta" title="abc.ABCMeta"><code>ABCMeta</code></a> as its metaclass. With this class, an abstract base class can be created by simply deriving from <code>ABC</code> avoiding sometimes confusing metaclass usage, for example:</p> <pre data-language="python">from abc import ABC
class MyABC(ABC):
pass
</pre> <p>Note that the type of <code>ABC</code> is still <a class="reference internal" href="#abc.ABCMeta" title="abc.ABCMeta"><code>ABCMeta</code></a>, therefore inheriting from <code>ABC</code> requires the usual precautions regarding metaclass usage, as multiple inheritance may lead to metaclass conflicts. One may also define an abstract base class by passing the metaclass keyword and using <code>ABCMeta</code> directly, for example:</p> <pre data-language="python">from abc import ABCMeta
class MyABC(metaclass=ABCMeta):
pass
</pre> <div class="versionadded"> <p><span class="versionmodified added">New in version 3.4.</span></p> </div> </dd>
</dl> <dl class="py class"> <dt class="sig sig-object py" id="abc.ABCMeta">
<code>class abc.ABCMeta</code> </dt> <dd>
<p>Metaclass for defining Abstract Base Classes (ABCs).</p> <p>Use this metaclass to create an ABC. An ABC can be subclassed directly, and then acts as a mix-in class. You can also register unrelated concrete classes (even built-in classes) and unrelated ABCs as “virtual subclasses” – these and their descendants will be considered subclasses of the registering ABC by the built-in <a class="reference internal" href="functions#issubclass" title="issubclass"><code>issubclass()</code></a> function, but the registering ABC won’t show up in their MRO (Method Resolution Order) nor will method implementations defined by the registering ABC be callable (not even via <a class="reference internal" href="functions#super" title="super"><code>super()</code></a>). <a class="footnote-reference brackets" href="#id2" id="id1">1</a></p> <p>Classes created with a metaclass of <code>ABCMeta</code> have the following method:</p> <dl class="py method"> <dt class="sig sig-object py" id="abc.ABCMeta.register">
<code>register(subclass)</code> </dt> <dd>
<p>Register <em>subclass</em> as a “virtual subclass” of this ABC. For example:</p> <pre data-language="python">from abc import ABC
class MyABC(ABC):
pass
MyABC.register(tuple)
assert issubclass(tuple, MyABC)
assert isinstance((), MyABC)
</pre> <div class="versionchanged"> <p><span class="versionmodified changed">Changed in version 3.3: </span>Returns the registered subclass, to allow usage as a class decorator.</p> </div> <div class="versionchanged"> <p><span class="versionmodified changed">Changed in version 3.4: </span>To detect calls to <code>register()</code>, you can use the <a class="reference internal" href="#abc.get_cache_token" title="abc.get_cache_token"><code>get_cache_token()</code></a> function.</p> </div> </dd>
</dl> <p>You can also override this method in an abstract base class:</p> <dl class="py method"> <dt class="sig sig-object py" id="abc.ABCMeta.__subclasshook__">
<code>__subclasshook__(subclass)</code> </dt> <dd>
<p>(Must be defined as a class method.)</p> <p>Check whether <em>subclass</em> is considered a subclass of this ABC. This means that you can customize the behavior of <a class="reference internal" href="functions#issubclass" title="issubclass"><code>issubclass()</code></a> further without the need to call <a class="reference internal" href="#abc.ABCMeta.register" title="abc.ABCMeta.register"><code>register()</code></a> on every class you want to consider a subclass of the ABC. (This class method is called from the <a class="reference internal" href="../reference/datamodel#class.__subclasscheck__" title="class.__subclasscheck__"><code>__subclasscheck__()</code></a> method of the ABC.)</p> <p>This method should return <code>True</code>, <code>False</code> or <code>NotImplemented</code>. If it returns <code>True</code>, the <em>subclass</em> is considered a subclass of this ABC. If it returns <code>False</code>, the <em>subclass</em> is not considered a subclass of this ABC, even if it would normally be one. If it returns <code>NotImplemented</code>, the subclass check is continued with the usual mechanism.</p> </dd>
</dl> <p>For a demonstration of these concepts, look at this example ABC definition:</p> <pre data-language="python">class Foo:
def __getitem__(self, index):
...
def __len__(self):
...
def get_iterator(self):
return iter(self)
class MyIterable(ABC):
@abstractmethod
def __iter__(self):
while False:
yield None
def get_iterator(self):
return self.__iter__()
@classmethod
def __subclasshook__(cls, C):
if cls is MyIterable:
if any("__iter__" in B.__dict__ for B in C.__mro__):
return True
return NotImplemented
MyIterable.register(Foo)
</pre> <p>The ABC <code>MyIterable</code> defines the standard iterable method, <a class="reference internal" href="stdtypes#iterator.__iter__" title="iterator.__iter__"><code>__iter__()</code></a>, as an abstract method. The implementation given here can still be called from subclasses. The <code>get_iterator()</code> method is also part of the <code>MyIterable</code> abstract base class, but it does not have to be overridden in non-abstract derived classes.</p> <p>The <a class="reference internal" href="#abc.ABCMeta.__subclasshook__" title="abc.ABCMeta.__subclasshook__"><code>__subclasshook__()</code></a> class method defined here says that any class that has an <a class="reference internal" href="stdtypes#iterator.__iter__" title="iterator.__iter__"><code>__iter__()</code></a> method in its <a class="reference internal" href="stdtypes#object.__dict__" title="object.__dict__"><code>__dict__</code></a> (or in that of one of its base classes, accessed via the <a class="reference internal" href="stdtypes#class.__mro__" title="class.__mro__"><code>__mro__</code></a> list) is considered a <code>MyIterable</code> too.</p> <p>Finally, the last line makes <code>Foo</code> a virtual subclass of <code>MyIterable</code>, even though it does not define an <a class="reference internal" href="stdtypes#iterator.__iter__" title="iterator.__iter__"><code>__iter__()</code></a> method (it uses the old-style iterable protocol, defined in terms of <a class="reference internal" href="../reference/datamodel#object.__len__" title="object.__len__"><code>__len__()</code></a> and <a class="reference internal" href="../reference/datamodel#object.__getitem__" title="object.__getitem__"><code>__getitem__()</code></a>). Note that this will not make <code>get_iterator</code> available as a method of <code>Foo</code>, so it is provided separately.</p> </dd>
</dl> <p>The <code>abc</code> module also provides the following decorator:</p> <dl class="py function"> <dt class="sig sig-object py" id="abc.abstractmethod">
<code>@abc.abstractmethod</code> </dt> <dd>
<p>A decorator indicating abstract methods.</p> <p>Using this decorator requires that the class’s metaclass is <a class="reference internal" href="#abc.ABCMeta" title="abc.ABCMeta"><code>ABCMeta</code></a> or is derived from it. A class that has a metaclass derived from <code>ABCMeta</code> cannot be instantiated unless all of its abstract methods and properties are overridden. The abstract methods can be called using any of the normal ‘super’ call mechanisms. <code>abstractmethod()</code> may be used to declare abstract methods for properties and descriptors.</p> <p>Dynamically adding abstract methods to a class, or attempting to modify the abstraction status of a method or class once it is created, are only supported using the <a class="reference internal" href="#abc.update_abstractmethods" title="abc.update_abstractmethods"><code>update_abstractmethods()</code></a> function. The <code>abstractmethod()</code> only affects subclasses derived using regular inheritance; “virtual subclasses” registered with the ABC’s <a class="reference internal" href="#abc.ABCMeta.register" title="abc.ABCMeta.register"><code>register()</code></a> method are not affected.</p> <p>When <code>abstractmethod()</code> is applied in combination with other method descriptors, it should be applied as the innermost decorator, as shown in the following usage examples:</p> <pre data-language="python">class C(ABC):
@abstractmethod
def my_abstract_method(self, arg1):
...
@classmethod
@abstractmethod
def my_abstract_classmethod(cls, arg2):
...
@staticmethod
@abstractmethod
def my_abstract_staticmethod(arg3):
...
@property
@abstractmethod
def my_abstract_property(self):
...
@my_abstract_property.setter
@abstractmethod
def my_abstract_property(self, val):
...
@abstractmethod
def _get_x(self):
...
@abstractmethod
def _set_x(self, val):
...
x = property(_get_x, _set_x)
</pre> <p>In order to correctly interoperate with the abstract base class machinery, the descriptor must identify itself as abstract using <code>__isabstractmethod__</code>. In general, this attribute should be <code>True</code> if any of the methods used to compose the descriptor are abstract. For example, Python’s built-in <a class="reference internal" href="functions#property" title="property"><code>property</code></a> does the equivalent of:</p> <pre data-language="python">class Descriptor:
...
@property
def __isabstractmethod__(self):
return any(getattr(f, '__isabstractmethod__', False) for
f in (self._fget, self._fset, self._fdel))
</pre> <div class="admonition note"> <p class="admonition-title">Note</p> <p>Unlike Java abstract methods, these abstract methods may have an implementation. This implementation can be called via the <a class="reference internal" href="functions#super" title="super"><code>super()</code></a> mechanism from the class that overrides it. This could be useful as an end-point for a super-call in a framework that uses cooperative multiple-inheritance.</p> </div> </dd>
</dl> <p>The <code>abc</code> module also supports the following legacy decorators:</p> <dl class="py function"> <dt class="sig sig-object py" id="abc.abstractclassmethod">
<code>@abc.abstractclassmethod</code> </dt> <dd>
<div class="versionadded"> <p><span class="versionmodified added">New in version 3.2.</span></p> </div> <div class="deprecated"> <p><span class="versionmodified deprecated">Deprecated since version 3.3: </span>It is now possible to use <a class="reference internal" href="functions#classmethod" title="classmethod"><code>classmethod</code></a> with <a class="reference internal" href="#abc.abstractmethod" title="abc.abstractmethod"><code>abstractmethod()</code></a>, making this decorator redundant.</p> </div> <p>A subclass of the built-in <a class="reference internal" href="functions#classmethod" title="classmethod"><code>classmethod()</code></a>, indicating an abstract classmethod. Otherwise it is similar to <a class="reference internal" href="#abc.abstractmethod" title="abc.abstractmethod"><code>abstractmethod()</code></a>.</p> <p>This special case is deprecated, as the <a class="reference internal" href="functions#classmethod" title="classmethod"><code>classmethod()</code></a> decorator is now correctly identified as abstract when applied to an abstract method:</p> <pre data-language="python">class C(ABC):
@classmethod
@abstractmethod
def my_abstract_classmethod(cls, arg):
...
</pre> </dd>
</dl> <dl class="py function"> <dt class="sig sig-object py" id="abc.abstractstaticmethod">
<code>@abc.abstractstaticmethod</code> </dt> <dd>
<div class="versionadded"> <p><span class="versionmodified added">New in version 3.2.</span></p> </div> <div class="deprecated"> <p><span class="versionmodified deprecated">Deprecated since version 3.3: </span>It is now possible to use <a class="reference internal" href="functions#staticmethod" title="staticmethod"><code>staticmethod</code></a> with <a class="reference internal" href="#abc.abstractmethod" title="abc.abstractmethod"><code>abstractmethod()</code></a>, making this decorator redundant.</p> </div> <p>A subclass of the built-in <a class="reference internal" href="functions#staticmethod" title="staticmethod"><code>staticmethod()</code></a>, indicating an abstract staticmethod. Otherwise it is similar to <a class="reference internal" href="#abc.abstractmethod" title="abc.abstractmethod"><code>abstractmethod()</code></a>.</p> <p>This special case is deprecated, as the <a class="reference internal" href="functions#staticmethod" title="staticmethod"><code>staticmethod()</code></a> decorator is now correctly identified as abstract when applied to an abstract method:</p> <pre data-language="python">class C(ABC):
@staticmethod
@abstractmethod
def my_abstract_staticmethod(arg):
...
</pre> </dd>
</dl> <dl class="py function"> <dt class="sig sig-object py" id="abc.abstractproperty">
<code>@abc.abstractproperty</code> </dt> <dd>
<div class="deprecated"> <p><span class="versionmodified deprecated">Deprecated since version 3.3: </span>It is now possible to use <a class="reference internal" href="functions#property" title="property"><code>property</code></a>, <a class="reference internal" href="functions#property.getter" title="property.getter"><code>property.getter()</code></a>, <a class="reference internal" href="functions#property.setter" title="property.setter"><code>property.setter()</code></a> and <a class="reference internal" href="functions#property.deleter" title="property.deleter"><code>property.deleter()</code></a> with <a class="reference internal" href="#abc.abstractmethod" title="abc.abstractmethod"><code>abstractmethod()</code></a>, making this decorator redundant.</p> </div> <p>A subclass of the built-in <a class="reference internal" href="functions#property" title="property"><code>property()</code></a>, indicating an abstract property.</p> <p>This special case is deprecated, as the <a class="reference internal" href="functions#property" title="property"><code>property()</code></a> decorator is now correctly identified as abstract when applied to an abstract method:</p> <pre data-language="python">class C(ABC):
@property
@abstractmethod
def my_abstract_property(self):
...
</pre> <p>The above example defines a read-only property; you can also define a read-write abstract property by appropriately marking one or more of the underlying methods as abstract:</p> <pre data-language="python">class C(ABC):
@property
def x(self):
...
@x.setter
@abstractmethod
def x(self, val):
...
</pre> <p>If only some components are abstract, only those components need to be updated to create a concrete property in a subclass:</p> <pre data-language="python">class D(C):
@C.x.setter
def x(self, val):
...
</pre> </dd>
</dl> <p>The <code>abc</code> module also provides the following functions:</p> <dl class="py function"> <dt class="sig sig-object py" id="abc.get_cache_token">
<code>abc.get_cache_token()</code> </dt> <dd>
<p>Returns the current abstract base class cache token.</p> <p>The token is an opaque object (that supports equality testing) identifying the current version of the abstract base class cache for virtual subclasses. The token changes with every call to <a class="reference internal" href="#abc.ABCMeta.register" title="abc.ABCMeta.register"><code>ABCMeta.register()</code></a> on any ABC.</p> <div class="versionadded"> <p><span class="versionmodified added">New in version 3.4.</span></p> </div> </dd>
</dl> <dl class="py function"> <dt class="sig sig-object py" id="abc.update_abstractmethods">
<code>abc.update_abstractmethods(cls)</code> </dt> <dd>
<p>A function to recalculate an abstract class’s abstraction status. This function should be called if a class’s abstract methods have been implemented or changed after it was created. Usually, this function should be called from within a class decorator.</p> <p>Returns <em>cls</em>, to allow usage as a class decorator.</p> <p>If <em>cls</em> is not an instance of <a class="reference internal" href="#abc.ABCMeta" title="abc.ABCMeta"><code>ABCMeta</code></a>, does nothing.</p> <div class="admonition note"> <p class="admonition-title">Note</p> <p>This function assumes that <em>cls</em>’s superclasses are already updated. It does not update any subclasses.</p> </div> <div class="versionadded"> <p><span class="versionmodified added">New in version 3.10.</span></p> </div> </dd>
</dl> <h4 class="rubric">Footnotes</h4> <dl class="footnote brackets"> <dt class="label" id="id2">
<code>1</code> </dt> <dd>
<p>C++ programmers should note that Python’s virtual base class concept is not the same as C++’s.</p> </dd> </dl> <div class="_attribution">
<p class="_attribution-p">
© 2001–2023 Python Software Foundation<br>Licensed under the PSF License.<br>
<a href="https://docs.python.org/3.12/library/abc.html" class="_attribution-link">https://docs.python.org/3.12/library/abc.html</a>
</p>
</div>
|
