One Hat Cyber Team

Edit File: descriptor.html

<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
 "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">

<html xmlns="http://www.w3.org/1999/xhtml">
 <head>
 <meta http-equiv="X-UA-Compatible" content="IE=Edge" />
 <meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
 <title>Descriptor HowTo Guide &#8212; Python 3.6.7 documentation</title>
 <link rel="stylesheet" href="../_static/pydoctheme.css" type="text/css" />
 <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
 <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
 <script type="text/javascript" src="../_static/jquery.js"></script>
 <script type="text/javascript" src="../_static/underscore.js"></script>
 <script type="text/javascript" src="../_static/doctools.js"></script>
 <script type="text/javascript" src="../_static/sidebar.js"></script>
 <link rel="search" type="application/opensearchdescription+xml"
 title="Search within Python 3.6.7 documentation"
 href="../_static/opensearch.xml"/>
 <link rel="author" title="About these documents" href="../about.html" />
 <link rel="index" title="Index" href="../genindex.html" />
 <link rel="search" title="Search" href="../search.html" />
 <link rel="copyright" title="Copyright" href="../copyright.html" />
 <link rel="next" title="Functional Programming HOWTO" href="functional.html" />
 <link rel="prev" title="Curses Programming with Python" href="curses.html" />
 <link rel="shortcut icon" type="image/png" href="../_static/py.png" />
 <link rel="canonical" href="https://docs.python.org/3/howto/descriptor.html" />
 
 <script type="text/javascript" src="../_static/copybutton.js"></script>

</head><body> 
 <div class="related" role="navigation" aria-label="related navigation">
 <h3>Navigation</h3>
 <ul>
 <li class="right" style="margin-right: 10px">
 <a href="../genindex.html" title="General Index"
 accesskey="I">index</a></li>
 <li class="right" >
 <a href="../py-modindex.html" title="Python Module Index"
 >modules</a> |</li>
 <li class="right" >
 <a href="functional.html" title="Functional Programming HOWTO"
 accesskey="N">next</a> |</li>
 <li class="right" >
 <a href="curses.html" title="Curses Programming with Python"
 accesskey="P">previous</a> |</li>
 <li><img src="../_static/py.png" alt=""
 style="vertical-align: middle; margin-top: -1px"/></li>
 <li><a href="https://www.python.org/">Python</a> &#187;</li>
 <li>
 <a href="../index.html">3.6.7 Documentation</a> &#187;
 </li>

<li class="nav-item nav-item-1"><a href="index.html" accesskey="U">Python HOWTOs</a> &#187;</li>
 <li class="right">

<div class="inline-search" style="display: none" role="search">
 <form class="inline-search" action="../search.html" method="get">
 <input placeholder="Quick search" type="text" name="q" />
 <input type="submit" value="Go" />
 <input type="hidden" name="check_keywords" value="yes" />
 <input type="hidden" name="area" value="default" />
 </form>
 </div>
 <script type="text/javascript">$('.inline-search').show(0);</script>
 |
 </li>

</ul>
 </div>

<div class="document">
 <div class="documentwrapper">
 <div class="bodywrapper">
 <div class="body" role="main">
 
 <div class="section" id="descriptor-howto-guide">
<h1><a class="toc-backref" href="#id1">Descriptor HowTo Guide</a><a class="headerlink" href="#descriptor-howto-guide" title="Permalink to this headline">¶</a></h1>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Author:</th><td class="field-body">Raymond Hettinger</td>
</tr>
<tr class="field-even field"><th class="field-name">Contact:</th><td class="field-body">&lt;python at rcn dot com&gt;</td>
</tr>
</tbody>
</table>
<div class="contents topic" id="contents">
Contents
<ul class="simple">
<li><a class="reference internal" href="#descriptor-howto-guide" id="id1">Descriptor HowTo Guide</a><ul>
<li><a class="reference internal" href="#abstract" id="id2">Abstract</a></li>
<li><a class="reference internal" href="#definition-and-introduction" id="id3">Definition and Introduction</a></li>
<li><a class="reference internal" href="#descriptor-protocol" id="id4">Descriptor Protocol</a></li>
<li><a class="reference internal" href="#invoking-descriptors" id="id5">Invoking Descriptors</a></li>
<li><a class="reference internal" href="#descriptor-example" id="id6">Descriptor Example</a></li>
<li><a class="reference internal" href="#properties" id="id7">Properties</a></li>
<li><a class="reference internal" href="#functions-and-methods" id="id8">Functions and Methods</a></li>
<li><a class="reference internal" href="#static-methods-and-class-methods" id="id9">Static Methods and Class Methods</a></li>
</ul>
</li>
</ul>
</div>
<div class="section" id="abstract">
<h2><a class="toc-backref" href="#id2">Abstract</a><a class="headerlink" href="#abstract" title="Permalink to this headline">¶</a></h2>
Defines descriptors, summarizes the protocol, and shows how descriptors are
called. Examines a custom descriptor and several built-in python descriptors
including functions, properties, static methods, and class methods. Shows how
each works by giving a pure Python equivalent and a sample application.
Learning about descriptors not only provides access to a larger toolset, it
creates a deeper understanding of how Python works and an appreciation for the
elegance of its design.
</div>
<div class="section" id="definition-and-introduction">
<h2><a class="toc-backref" href="#id3">Definition and Introduction</a><a class="headerlink" href="#definition-and-introduction" title="Permalink to this headline">¶</a></h2>
In general, a descriptor is an object attribute with “binding behavior”, one
whose attribute access has been overridden by methods in the descriptor
protocol. Those methods are <a class="reference internal" href="../reference/datamodel.html#object.__get__" title="object.__get__"><code class="xref py py-meth docutils literal notranslate">__get__()</code></a>, <a class="reference internal" href="../reference/datamodel.html#object.__set__" title="object.__set__"><code class="xref py py-meth docutils literal notranslate">__set__()</code></a>, and
<a class="reference internal" href="../reference/datamodel.html#object.__delete__" title="object.__delete__"><code class="xref py py-meth docutils literal notranslate">__delete__()</code></a>. If any of those methods are defined for an object, it is
said to be a descriptor.
The default behavior for attribute access is to get, set, or delete the
attribute from an object’s dictionary. For instance, <code class="docutils literal notranslate">a.x</code> has a lookup chain
starting with <code class="docutils literal notranslate">a.__dict__['x']</code>, then <code class="docutils literal notranslate">type(a).__dict__['x']</code>, and
continuing through the base classes of <code class="docutils literal notranslate">type(a)</code> excluding metaclasses. If the
looked-up value is an object defining one of the descriptor methods, then Python
may override the default behavior and invoke the descriptor method instead.
Where this occurs in the precedence chain depends on which descriptor methods
were defined.
Descriptors are a powerful, general purpose protocol. They are the mechanism
behind properties, methods, static methods, class methods, and <a class="reference internal" href="../library/functions.html#super" title="super"><code class="xref py py-func docutils literal notranslate">super()</code></a>.
They are used throughout Python itself to implement the new style classes
introduced in version 2.2. Descriptors simplify the underlying C-code and offer
a flexible set of new tools for everyday Python programs.
</div>
<div class="section" id="descriptor-protocol">
<h2><a class="toc-backref" href="#id4">Descriptor Protocol</a><a class="headerlink" href="#descriptor-protocol" title="Permalink to this headline">¶</a></h2>
<code class="docutils literal notranslate">descr.__get__(self, obj, type=None) --&gt; value</code>
<code class="docutils literal notranslate">descr.__set__(self, obj, value) --&gt; None</code>
<code class="docutils literal notranslate">descr.__delete__(self, obj) --&gt; None</code>
That is all there is to it. Define any of these methods and an object is
considered a descriptor and can override default behavior upon being looked up
as an attribute.
If an object defines both <a class="reference internal" href="../reference/datamodel.html#object.__get__" title="object.__get__"><code class="xref py py-meth docutils literal notranslate">__get__()</code></a> and <a class="reference internal" href="../reference/datamodel.html#object.__set__" title="object.__set__"><code class="xref py py-meth docutils literal notranslate">__set__()</code></a>, it is considered
a data descriptor. Descriptors that only define <a class="reference internal" href="../reference/datamodel.html#object.__get__" title="object.__get__"><code class="xref py py-meth docutils literal notranslate">__get__()</code></a> are called
non-data descriptors (they are typically used for methods but other uses are
possible).
Data and non-data descriptors differ in how overrides are calculated with
respect to entries in an instance’s dictionary. If an instance’s dictionary
has an entry with the same name as a data descriptor, the data descriptor
takes precedence. If an instance’s dictionary has an entry with the same
name as a non-data descriptor, the dictionary entry takes precedence.
To make a read-only data descriptor, define both <a class="reference internal" href="../reference/datamodel.html#object.__get__" title="object.__get__"><code class="xref py py-meth docutils literal notranslate">__get__()</code></a> and
<a class="reference internal" href="../reference/datamodel.html#object.__set__" title="object.__set__"><code class="xref py py-meth docutils literal notranslate">__set__()</code></a> with the <a class="reference internal" href="../reference/datamodel.html#object.__set__" title="object.__set__"><code class="xref py py-meth docutils literal notranslate">__set__()</code></a> raising an <a class="reference internal" href="../library/exceptions.html#AttributeError" title="AttributeError"><code class="xref py py-exc docutils literal notranslate">AttributeError</code></a> when
called. Defining the <a class="reference internal" href="../reference/datamodel.html#object.__set__" title="object.__set__"><code class="xref py py-meth docutils literal notranslate">__set__()</code></a> method with an exception raising
placeholder is enough to make it a data descriptor.
</div>
<div class="section" id="invoking-descriptors">
<h2><a class="toc-backref" href="#id5">Invoking Descriptors</a><a class="headerlink" href="#invoking-descriptors" title="Permalink to this headline">¶</a></h2>
A descriptor can be called directly by its method name. For example,
<code class="docutils literal notranslate">d.__get__(obj)</code>.
Alternatively, it is more common for a descriptor to be invoked automatically
upon attribute access. For example, <code class="docutils literal notranslate">obj.d</code> looks up <code class="docutils literal notranslate">d</code> in the dictionary
of <code class="docutils literal notranslate">obj</code>. If <code class="docutils literal notranslate">d</code> defines the method <a class="reference internal" href="../reference/datamodel.html#object.__get__" title="object.__get__"><code class="xref py py-meth docutils literal notranslate">__get__()</code></a>, then <code class="docutils literal notranslate">d.__get__(obj)</code>
is invoked according to the precedence rules listed below.
The details of invocation depend on whether <code class="docutils literal notranslate">obj</code> is an object or a class.
For objects, the machinery is in <a class="reference internal" href="../reference/datamodel.html#object.__getattribute__" title="object.__getattribute__"><code class="xref py py-meth docutils literal notranslate">object.__getattribute__()</code></a> which
transforms <code class="docutils literal notranslate">b.x</code> into <code class="docutils literal notranslate">type(b).__dict__['x'].__get__(b, type(b))</code>. The
implementation works through a precedence chain that gives data descriptors
priority over instance variables, instance variables priority over non-data
descriptors, and assigns lowest priority to <a class="reference internal" href="../reference/datamodel.html#object.__getattr__" title="object.__getattr__"><code class="xref py py-meth docutils literal notranslate">__getattr__()</code></a> if provided.
The full C implementation can be found in <a class="reference internal" href="../c-api/object.html#c.PyObject_GenericGetAttr" title="PyObject_GenericGetAttr"><code class="xref c c-func docutils literal notranslate">PyObject_GenericGetAttr()</code></a> in
<a class="reference external" href="https://github.com/python/cpython/tree/3.6/Objects/object.c">Objects/object.c</a>.
For classes, the machinery is in <code class="xref py py-meth docutils literal notranslate">type.__getattribute__()</code> which transforms
<code class="docutils literal notranslate">B.x</code> into <code class="docutils literal notranslate">B.__dict__['x'].__get__(None, B)</code>. In pure Python, it looks
like:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>def __getattribute__(self, key):
 &quot;Emulate type_getattro() in Objects/typeobject.c&quot;
 v = object.__getattribute__(self, key)
 if hasattr(v, &#39;__get__&#39;):
 return v.__get__(None, self)
 return v
</pre></div>
</div>
The important points to remember are:
<ul class="simple">
<li>descriptors are invoked by the <a class="reference internal" href="../reference/datamodel.html#object.__getattribute__" title="object.__getattribute__"><code class="xref py py-meth docutils literal notranslate">__getattribute__()</code></a> method</li>
<li>overriding <a class="reference internal" href="../reference/datamodel.html#object.__getattribute__" title="object.__getattribute__"><code class="xref py py-meth docutils literal notranslate">__getattribute__()</code></a> prevents automatic descriptor calls</li>
<li><a class="reference internal" href="../reference/datamodel.html#object.__getattribute__" title="object.__getattribute__"><code class="xref py py-meth docutils literal notranslate">object.__getattribute__()</code></a> and <code class="xref py py-meth docutils literal notranslate">type.__getattribute__()</code> make
different calls to <a class="reference internal" href="../reference/datamodel.html#object.__get__" title="object.__get__"><code class="xref py py-meth docutils literal notranslate">__get__()</code></a>.</li>
<li>data descriptors always override instance dictionaries.</li>
<li>non-data descriptors may be overridden by instance dictionaries.</li>
</ul>
The object returned by <code class="docutils literal notranslate">super()</code> also has a custom <a class="reference internal" href="../reference/datamodel.html#object.__getattribute__" title="object.__getattribute__"><code class="xref py py-meth docutils literal notranslate">__getattribute__()</code></a>
method for invoking descriptors. The call <code class="docutils literal notranslate">super(B, obj).m()</code> searches
<code class="docutils literal notranslate">obj.__class__.__mro__</code> for the base class <code class="docutils literal notranslate">A</code> immediately following <code class="docutils literal notranslate">B</code>
and then returns <code class="docutils literal notranslate">A.__dict__['m'].__get__(obj, B)</code>. If not a descriptor,
<code class="docutils literal notranslate">m</code> is returned unchanged. If not in the dictionary, <code class="docutils literal notranslate">m</code> reverts to a
search using <a class="reference internal" href="../reference/datamodel.html#object.__getattribute__" title="object.__getattribute__"><code class="xref py py-meth docutils literal notranslate">object.__getattribute__()</code></a>.
The implementation details are in <code class="xref c c-func docutils literal notranslate">super_getattro()</code> in
<a class="reference external" href="https://github.com/python/cpython/tree/3.6/Objects/typeobject.c">Objects/typeobject.c</a>. and a pure Python equivalent can be found in
<a class="reference external" href="https://www.python.org/download/releases/2.2.3/descrintro/#cooperation">Guido’s Tutorial</a>.
The details above show that the mechanism for descriptors is embedded in the
<a class="reference internal" href="../reference/datamodel.html#object.__getattribute__" title="object.__getattribute__"><code class="xref py py-meth docutils literal notranslate">__getattribute__()</code></a> methods for <a class="reference internal" href="../library/functions.html#object" title="object"><code class="xref py py-class docutils literal notranslate">object</code></a>, <a class="reference internal" href="../library/functions.html#type" title="type"><code class="xref py py-class docutils literal notranslate">type</code></a>, and
<a class="reference internal" href="../library/functions.html#super" title="super"><code class="xref py py-func docutils literal notranslate">super()</code></a>. Classes inherit this machinery when they derive from
<a class="reference internal" href="../library/functions.html#object" title="object"><code class="xref py py-class docutils literal notranslate">object</code></a> or if they have a meta-class providing similar functionality.
Likewise, classes can turn-off descriptor invocation by overriding
<a class="reference internal" href="../reference/datamodel.html#object.__getattribute__" title="object.__getattribute__"><code class="xref py py-meth docutils literal notranslate">__getattribute__()</code></a>.
</div>
<div class="section" id="descriptor-example">
<h2><a class="toc-backref" href="#id6">Descriptor Example</a><a class="headerlink" href="#descriptor-example" title="Permalink to this headline">¶</a></h2>
The following code creates a class whose objects are data descriptors which
print a message for each get or set. Overriding <a class="reference internal" href="../reference/datamodel.html#object.__getattribute__" title="object.__getattribute__"><code class="xref py py-meth docutils literal notranslate">__getattribute__()</code></a> is
alternate approach that could do this for every attribute. However, this
descriptor is useful for monitoring just a few chosen attributes:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>class RevealAccess(object):
 &quot;&quot;&quot;A data descriptor that sets and returns values
 normally and prints a message logging their access.
 &quot;&quot;&quot;

def __init__(self, initval=None, name=&#39;var&#39;):
 self.val = initval
 self.name = name

def __set__(self, obj, val):
 print(&#39;Updating&#39;, self.name)
 self.val = val

&gt;&gt;&gt; class MyClass(object):
... x = RevealAccess(10, &#39;var &quot;x&quot;&#39;)
... y = 5
...
&gt;&gt;&gt; m = MyClass()
&gt;&gt;&gt; m.x
Retrieving var &quot;x&quot;
10
&gt;&gt;&gt; m.x = 20
Updating var &quot;x&quot;
&gt;&gt;&gt; m.x
Retrieving var &quot;x&quot;
20
&gt;&gt;&gt; m.y
5
</pre></div>
</div>
The protocol is simple and offers exciting possibilities. Several use cases are
so common that they have been packaged into individual function calls.
Properties, bound methods, static methods, and class methods are all
based on the descriptor protocol.
</div>
<div class="section" id="properties">
<h2><a class="toc-backref" href="#id7">Properties</a><a class="headerlink" href="#properties" title="Permalink to this headline">¶</a></h2>
Calling <a class="reference internal" href="../library/functions.html#property" title="property"><code class="xref py py-func docutils literal notranslate">property()</code></a> is a succinct way of building a data descriptor that
triggers function calls upon access to an attribute. Its signature is:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>property(fget=None, fset=None, fdel=None, doc=None) -&gt; property attribute
</pre></div>
</div>
The documentation shows a typical use to define a managed attribute <code class="docutils literal notranslate">x</code>:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>class C(object):
 def getx(self): return self.__x
 def setx(self, value): self.__x = value
 def delx(self): del self.__x
 x = property(getx, setx, delx, &quot;I&#39;m the &#39;x&#39; property.&quot;)
</pre></div>
</div>
To see how <a class="reference internal" href="../library/functions.html#property" title="property"><code class="xref py py-func docutils literal notranslate">property()</code></a> is implemented in terms of the descriptor protocol,
here is a pure Python equivalent:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>class Property(object):
 &quot;Emulate PyProperty_Type() in Objects/descrobject.c&quot;

def __init__(self, fget=None, fset=None, fdel=None, doc=None):
 self.fget = fget
 self.fset = fset
 self.fdel = fdel
 if doc is None and fget is not None:
 doc = fget.__doc__
 self.__doc__ = doc

def __set__(self, obj, value):
 if self.fset is None:
 raise AttributeError(&quot;can&#39;t set attribute&quot;)
 self.fset(obj, value)

def __delete__(self, obj):
 if self.fdel is None:
 raise AttributeError(&quot;can&#39;t delete attribute&quot;)
 self.fdel(obj)

def getter(self, fget):
 return type(self)(fget, self.fset, self.fdel, self.__doc__)

def setter(self, fset):
 return type(self)(self.fget, fset, self.fdel, self.__doc__)

def deleter(self, fdel):
 return type(self)(self.fget, self.fset, fdel, self.__doc__)
</pre></div>
</div>
The <a class="reference internal" href="../library/functions.html#property" title="property"><code class="xref py py-func docutils literal notranslate">property()</code></a> builtin helps whenever a user interface has granted
attribute access and then subsequent changes require the intervention of a
method.
For instance, a spreadsheet class may grant access to a cell value through
<code class="docutils literal notranslate">Cell('b10').value</code>. Subsequent improvements to the program require the cell
to be recalculated on every access; however, the programmer does not want to
affect existing client code accessing the attribute directly. The solution is
to wrap access to the value attribute in a property data descriptor:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>class Cell(object):
 . . .
 def getvalue(self):
 &quot;Recalculate the cell before returning value&quot;
 self.recalc()
 return self._value
 value = property(getvalue)
</pre></div>
</div>
</div>
<div class="section" id="functions-and-methods">
<h2><a class="toc-backref" href="#id8">Functions and Methods</a><a class="headerlink" href="#functions-and-methods" title="Permalink to this headline">¶</a></h2>
Python’s object oriented features are built upon a function based environment.
Using non-data descriptors, the two are merged seamlessly.
Class dictionaries store methods as functions. In a class definition, methods
are written using <a class="reference internal" href="../reference/compound_stmts.html#def"><code class="xref std std-keyword docutils literal notranslate">def</code></a> or <a class="reference internal" href="../reference/expressions.html#lambda"><code class="xref std std-keyword docutils literal notranslate">lambda</code></a>, the usual tools for
creating functions. Methods only differ from regular functions in that the
first argument is reserved for the object instance. By Python convention, the
instance reference is called self but may be called this or any other
variable name.
To support method calls, functions include the <a class="reference internal" href="../reference/datamodel.html#object.__get__" title="object.__get__"><code class="xref py py-meth docutils literal notranslate">__get__()</code></a> method for
binding methods during attribute access. This means that all functions are
non-data descriptors which return bound methods when they are invoked from an
object. In pure python, it works like this:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>class Function(object):
 . . .
 def __get__(self, obj, objtype=None):
 &quot;Simulate func_descr_get() in Objects/funcobject.c&quot;
 if obj is None:
 return self
 return types.MethodType(self, obj)
</pre></div>
</div>
Running the interpreter shows how the function descriptor works in practice:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; class D(object):
... def f(self, x):
... return x
...
&gt;&gt;&gt; d = D()

# Access through the class dictionary does not invoke __get__.
# It just returns the underlying function object.
&gt;&gt;&gt; D.__dict__[&#39;f&#39;]
&lt;function D.f at 0x00C45070&gt;

# Dotted access from a class calls __get__() which just returns
# the underlying function unchanged.
&gt;&gt;&gt; D.f
&lt;function D.f at 0x00C45070&gt;

# The function has a __qualname__ attribute to support introspection
&gt;&gt;&gt; D.f.__qualname__
&#39;D.f&#39;

# Dotted access from an instance calls __get__() which returns the
# function wrapped in a bound method object
&gt;&gt;&gt; d.f
&lt;bound method D.f of &lt;__main__.D object at 0x00B18C90&gt;&gt;

# Internally, the bound method stores the underlying function,
# the bound instance, and the class of the bound instance.
&gt;&gt;&gt; d.f.__func__
&lt;function D.f at 0x1012e5ae8&gt;
&gt;&gt;&gt; d.f.__self__
&lt;__main__.D object at 0x1012e1f98&gt;
&gt;&gt;&gt; d.f.__class__
&lt;class &#39;method&#39;&gt;
</pre></div>
</div>
</div>
<div class="section" id="static-methods-and-class-methods">
<h2><a class="toc-backref" href="#id9">Static Methods and Class Methods</a><a class="headerlink" href="#static-methods-and-class-methods" title="Permalink to this headline">¶</a></h2>
Non-data descriptors provide a simple mechanism for variations on the usual
patterns of binding functions into methods.
To recap, functions have a <a class="reference internal" href="../reference/datamodel.html#object.__get__" title="object.__get__"><code class="xref py py-meth docutils literal notranslate">__get__()</code></a> method so that they can be converted
to a method when accessed as attributes. The non-data descriptor transforms an
<code class="docutils literal notranslate">obj.f(*args)</code> call into <code class="docutils literal notranslate">f(obj, *args)</code>. Calling <code class="docutils literal notranslate">klass.f(*args)</code>
becomes <code class="docutils literal notranslate">f(*args)</code>.
This chart summarizes the binding and its two most useful variants:
<blockquote>
<div><table border="1" class="docutils">
<colgroup>
<col width="30%" />
<col width="39%" />
<col width="32%" />
</colgroup>
<thead valign="bottom">
<tr class="row-odd"><th class="head">Transformation</th>
<th class="head">Called from an
Object</th>
<th class="head">Called from a
Class</th>
</tr>
</thead>
<tbody valign="top">
<tr class="row-even"><td>function</td>
<td>f(obj, *args)</td>
<td>f(*args)</td>
</tr>
<tr class="row-odd"><td>staticmethod</td>
<td>f(*args)</td>
<td>f(*args)</td>
</tr>
<tr class="row-even"><td>classmethod</td>
<td>f(type(obj), *args)</td>
<td>f(klass, *args)</td>
</tr>
</tbody>
</table>
</div></blockquote>
Static methods return the underlying function without changes. Calling either
<code class="docutils literal notranslate">c.f</code> or <code class="docutils literal notranslate">C.f</code> is the equivalent of a direct lookup into
<code class="docutils literal notranslate">object.__getattribute__(c, &quot;f&quot;)</code> or <code class="docutils literal notranslate">object.__getattribute__(C, &quot;f&quot;)</code>. As a
result, the function becomes identically accessible from either an object or a
class.
Good candidates for static methods are methods that do not reference the
<code class="docutils literal notranslate">self</code> variable.
For instance, a statistics package may include a container class for
experimental data. The class provides normal methods for computing the average,
mean, median, and other descriptive statistics that depend on the data. However,
there may be useful functions which are conceptually related but do not depend
on the data. For instance, <code class="docutils literal notranslate">erf(x)</code> is handy conversion routine that comes up
in statistical work but does not directly depend on a particular dataset.
It can be called either from an object or the class: <code class="docutils literal notranslate">s.erf(1.5) --&gt; .9332</code> or
<code class="docutils literal notranslate">Sample.erf(1.5) --&gt; .9332</code>.
Since staticmethods return the underlying function with no changes, the example
calls are unexciting:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; class E(object):
... def f(x):
... print(x)
... f = staticmethod(f)
...
&gt;&gt;&gt; print(E.f(3))
3
&gt;&gt;&gt; print(E().f(3))
3
</pre></div>
</div>
Using the non-data descriptor protocol, a pure Python version of
<a class="reference internal" href="../library/functions.html#staticmethod" title="staticmethod"><code class="xref py py-func docutils literal notranslate">staticmethod()</code></a> would look like this:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>class StaticMethod(object):
 &quot;Emulate PyStaticMethod_Type() in Objects/funcobject.c&quot;

def __init__(self, f):
 self.f = f

def __get__(self, obj, objtype=None):
 return self.f
</pre></div>
</div>
Unlike static methods, class methods prepend the class reference to the
argument list before calling the function. This format is the same
for whether the caller is an object or a class:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; class E(object):
... def f(klass, x):
... return klass.__name__, x
... f = classmethod(f)
...
&gt;&gt;&gt; print(E.f(3))
(&#39;E&#39;, 3)
&gt;&gt;&gt; print(E().f(3))
(&#39;E&#39;, 3)
</pre></div>
</div>
This behavior is useful whenever the function only needs to have a class
reference and does not care about any underlying data. One use for classmethods
is to create alternate class constructors. In Python 2.3, the classmethod
<a class="reference internal" href="../library/stdtypes.html#dict.fromkeys" title="dict.fromkeys"><code class="xref py py-func docutils literal notranslate">dict.fromkeys()</code></a> creates a new dictionary from a list of keys. The pure
Python equivalent is:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>class Dict(object):
 . . .
 def fromkeys(klass, iterable, value=None):
 &quot;Emulate dict_fromkeys() in Objects/dictobject.c&quot;
 d = klass()
 for key in iterable:
 d[key] = value
 return d
 fromkeys = classmethod(fromkeys)
</pre></div>
</div>
Now a new dictionary of unique keys can be constructed like this:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; Dict.fromkeys(&#39;abracadabra&#39;)
{&#39;a&#39;: None, &#39;r&#39;: None, &#39;b&#39;: None, &#39;c&#39;: None, &#39;d&#39;: None}
</pre></div>
</div>
Using the non-data descriptor protocol, a pure Python version of
<a class="reference internal" href="../library/functions.html#classmethod" title="classmethod"><code class="xref py py-func docutils literal notranslate">classmethod()</code></a> would look like this:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>class ClassMethod(object):
 &quot;Emulate PyClassMethod_Type() in Objects/funcobject.c&quot;

</div>
 </div>
 </div>
 <div class="sphinxsidebar" role="navigation" aria-label="main navigation">
 <div class="sphinxsidebarwrapper">
 <h3><a href="../contents.html">Table Of Contents</a></h3>
 <ul>
<li><a class="reference internal" href="#">Descriptor HowTo Guide</a><ul>
<li><a class="reference internal" href="#abstract">Abstract</a></li>
<li><a class="reference internal" href="#definition-and-introduction">Definition and Introduction</a></li>
<li><a class="reference internal" href="#descriptor-protocol">Descriptor Protocol</a></li>
<li><a class="reference internal" href="#invoking-descriptors">Invoking Descriptors</a></li>
<li><a class="reference internal" href="#descriptor-example">Descriptor Example</a></li>
<li><a class="reference internal" href="#properties">Properties</a></li>
<li><a class="reference internal" href="#functions-and-methods">Functions and Methods</a></li>
<li><a class="reference internal" href="#static-methods-and-class-methods">Static Methods and Class Methods</a></li>
</ul>
</li>
</ul>

<h4>Previous topic</h4>
 <a href="curses.html"
 title="previous chapter">Curses Programming with Python</a>
 <h4>Next topic</h4>
 <a href="functional.html"
 title="next chapter">Functional Programming HOWTO</a>
 <div role="note" aria-label="source link">
 <h3>This Page</h3>
 <ul class="this-page-menu">
 <li><a href="../bugs.html">Report a Bug</a></li>
 <li>
 <a href="https://github.com/python/cpython/blob/3.6/Doc/howto/descriptor.rst"
 rel="nofollow">Show Source
 </a>
 </li>
 </ul>
 </div>
 </div>
 </div>
 <div class="clearer"></div>
 </div> 
 <div class="related" role="navigation" aria-label="related navigation">
 <h3>Navigation</h3>
 <ul>
 <li class="right" style="margin-right: 10px">
 <a href="../genindex.html" title="General Index"
 >index</a></li>
 <li class="right" >
 <a href="../py-modindex.html" title="Python Module Index"
 >modules</a> |</li>
 <li class="right" >
 <a href="functional.html" title="Functional Programming HOWTO"
 >next</a> |</li>
 <li class="right" >
 <a href="curses.html" title="Curses Programming with Python"
 >previous</a> |</li>
 <li><img src="../_static/py.png" alt=""
 style="vertical-align: middle; margin-top: -1px"/></li>
 <li><a href="https://www.python.org/">Python</a> &#187;</li>
 <li>
 <a href="../index.html">3.6.7 Documentation</a> &#187;
 </li>

<li class="nav-item nav-item-1"><a href="index.html" >Python HOWTOs</a> &#187;</li>
 <li class="right">

</ul>
 </div> 
 <div class="footer">
 &copy; <a href="../copyright.html">Copyright</a> 2001-2023, Python Software Foundation.
 
 The Python Software Foundation is a non-profit corporation.
 <a href="https://www.python.org/psf/donations/">Please donate.</a>
 
 Last updated on Dec 18, 2023.
 <a href="../bugs.html">Found a bug</a>?
 
 Created using <a href="http://sphinx.pocoo.org/">Sphinx</a> 1.7.6.
 </div>

</body>
</html>