^^&&^^

Edit File: 2.4.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="Content-Type" content="text/html; charset=utf-8" />
 
 <title>What’s New in Python 2.4 &mdash; Python 2.7.5 documentation</title>
 
 <link rel="stylesheet" href="../_static/default.css" type="text/css" />
 <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
 
 <script type="text/javascript">
 var DOCUMENTATION_OPTIONS = {
 URL_ROOT: '../',
 VERSION: '2.7.5',
 COLLAPSE_INDEX: false,
 FILE_SUFFIX: '.html',
 HAS_SOURCE: true
 };
 </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 2.7.5 documentation"
 href="../_static/opensearch.xml"/>
 <link rel="author" title="About these documents" href="../about.html" />
 <link rel="copyright" title="Copyright" href="../copyright.html" />
 <link rel="top" title="Python 2.7.5 documentation" href="../index.html" />
 <link rel="up" title="What’s New in Python" href="index.html" />
 <link rel="next" title="What’s New in Python 2.3" href="2.3.html" />
 <link rel="prev" title="What’s New in Python 2.5" href="2.5.html" />
 <link rel="shortcut icon" type="image/png" href="../_static/py.png" />
 <script type="text/javascript" src="../_static/copybutton.js"></script>

</head>
 <body>
 <div class="related">
 <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="2.3.html" title="What’s New in Python 2.3"
 accesskey="N">next</a> |</li>
 <li class="right" >
 <a href="2.5.html" title="What’s New in Python 2.5"
 accesskey="P">previous</a> |</li>
 <li><img src="../_static/py.png" alt=""
 style="vertical-align: middle; margin-top: -1px"/></li>
 <li><a href="http://www.python.org/">Python</a> &raquo;</li>
 <li>
 <a href="../index.html">Python 2.7.5 documentation</a> &raquo;
 </li>

<li><a href="index.html" accesskey="U">What&#8217;s New in Python</a> &raquo;</li> 
 </ul>
 </div>

<div class="document">
 <div class="documentwrapper">
 <div class="bodywrapper">
 <div class="body">
 
 <div class="section" id="what-s-new-in-python-2-4">
<h1>What&#8217;s New in Python 2.4<a class="headerlink" href="#what-s-new-in-python-2-4" 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">A.M. Kuchling</td>
</tr>
</tbody>
</table>
This article explains the new features in Python 2.4.1, released on March 30,
2005.
Python 2.4 is a medium-sized release. It doesn&#8217;t introduce as many changes as
the radical Python 2.2, but introduces more features than the conservative 2.3
release. The most significant new language features are function decorators and
generator expressions; most other changes are to the standard library.
According to the CVS change logs, there were 481 patches applied and 502 bugs
fixed between Python 2.3 and 2.4. Both figures are likely to be underestimates.
This article doesn&#8217;t attempt to provide a complete specification of every single
new feature, but instead provides a brief introduction to each feature. For
full details, you should refer to the documentation for Python 2.4, such as the
Python Library Reference and the Python Reference Manual. Often you will be
referred to the PEP for a particular new feature for explanations of the
implementation and design rationale.
<div class="section" id="pep-218-built-in-set-objects">
<h2>PEP 218: Built-In Set Objects<a class="headerlink" href="#pep-218-built-in-set-objects" title="Permalink to this headline">¶</a></h2>
Python 2.3 introduced the <a class="reference internal" href="../library/sets.html#module-sets" title="sets: Implementation of sets of unique elements. (deprecated)"><tt class="xref py py-mod docutils literal">sets</tt></a> module. C implementations of set data
types have now been added to the Python core as two new built-in types,
<tt class="xref py py-func docutils literal">set(iterable)()</tt> and <tt class="xref py py-func docutils literal">frozenset(iterable)()</tt>. They provide high speed
operations for membership testing, for eliminating duplicates from sequences,
and for mathematical operations like unions, intersections, differences, and
symmetric differences.
<div class="highlight-python"><div class="highlight"><pre>&gt;&gt;&gt; a = set(&#39;abracadabra&#39;) # form a set from a string
&gt;&gt;&gt; &#39;z&#39; in a # fast membership testing
False
&gt;&gt;&gt; a # unique letters in a
set([&#39;a&#39;, &#39;r&#39;, &#39;b&#39;, &#39;c&#39;, &#39;d&#39;])
&gt;&gt;&gt; &#39;&#39;.join(a) # convert back into a string
&#39;arbcd&#39;

&gt;&gt;&gt; b = set(&#39;alacazam&#39;) # form a second set
&gt;&gt;&gt; a - b # letters in a but not in b
set([&#39;r&#39;, &#39;d&#39;, &#39;b&#39;])
&gt;&gt;&gt; a | b # letters in either a or b
set([&#39;a&#39;, &#39;c&#39;, &#39;r&#39;, &#39;d&#39;, &#39;b&#39;, &#39;m&#39;, &#39;z&#39;, &#39;l&#39;])
&gt;&gt;&gt; a &amp; b # letters in both a and b
set([&#39;a&#39;, &#39;c&#39;])
&gt;&gt;&gt; a ^ b # letters in a or b but not both
set([&#39;r&#39;, &#39;d&#39;, &#39;b&#39;, &#39;m&#39;, &#39;z&#39;, &#39;l&#39;])

&gt;&gt;&gt; a.add(&#39;z&#39;) # add a new element
&gt;&gt;&gt; a.update(&#39;wxy&#39;) # add multiple new elements
&gt;&gt;&gt; a
set([&#39;a&#39;, &#39;c&#39;, &#39;b&#39;, &#39;d&#39;, &#39;r&#39;, &#39;w&#39;, &#39;y&#39;, &#39;x&#39;, &#39;z&#39;])
&gt;&gt;&gt; a.remove(&#39;x&#39;) # take one element out
&gt;&gt;&gt; a
set([&#39;a&#39;, &#39;c&#39;, &#39;b&#39;, &#39;d&#39;, &#39;r&#39;, &#39;w&#39;, &#39;y&#39;, &#39;z&#39;])
</pre></div>
</div>
The <a class="reference internal" href="../library/stdtypes.html#frozenset" title="frozenset"><tt class="xref py py-func docutils literal">frozenset()</tt></a> type is an immutable version of <a class="reference internal" href="../library/stdtypes.html#set" title="set"><tt class="xref py py-func docutils literal">set()</tt></a>. Since it is
immutable and hashable, it may be used as a dictionary key or as a member of
another set.
The <a class="reference internal" href="../library/sets.html#module-sets" title="sets: Implementation of sets of unique elements. (deprecated)"><tt class="xref py py-mod docutils literal">sets</tt></a> module remains in the standard library, and may be useful if you
wish to subclass the <tt class="xref py py-class docutils literal">Set</tt> or <tt class="xref py py-class docutils literal">ImmutableSet</tt> classes. There are
currently no plans to deprecate the module.
<div class="admonition-see-also admonition seealso">
See also
<dl class="last docutils">
<dt><a class="pep reference external" href="http://www.python.org/dev/peps/pep-0218">PEP 218</a> - Adding a Built-In Set Object Type</dt>
<dd>Originally proposed by Greg Wilson and ultimately implemented by Raymond
Hettinger.</dd>
</dl>
</div>
</div>
<div class="section" id="pep-237-unifying-long-integers-and-integers">
<h2>PEP 237: Unifying Long Integers and Integers<a class="headerlink" href="#pep-237-unifying-long-integers-and-integers" title="Permalink to this headline">¶</a></h2>
The lengthy transition process for this PEP, begun in Python 2.2, takes another
step forward in Python 2.4. In 2.3, certain integer operations that would
behave differently after int/long unification triggered <a class="reference internal" href="../library/exceptions.html#exceptions.FutureWarning" title="exceptions.FutureWarning"><tt class="xref py py-exc docutils literal">FutureWarning</tt></a>
warnings and returned values limited to 32 or 64 bits (depending on your
platform). In 2.4, these expressions no longer produce a warning and instead
produce a different result that&#8217;s usually a long integer.
The problematic expressions are primarily left shifts and lengthy hexadecimal
and octal constants. For example, <tt class="docutils literal">2 &lt;&lt; 32</tt> results in a warning in 2.3,
evaluating to 0 on 32-bit platforms. In Python 2.4, this expression now returns
the correct answer, 8589934592.
<div class="admonition-see-also admonition seealso">
See also
<dl class="last docutils">
<dt><a class="pep reference external" href="http://www.python.org/dev/peps/pep-0237">PEP 237</a> - Unifying Long Integers and Integers</dt>
<dd>Original PEP written by Moshe Zadka and GvR. The changes for 2.4 were
implemented by Kalle Svensson.</dd>
</dl>
</div>
</div>
<div class="section" id="pep-289-generator-expressions">
<h2>PEP 289: Generator Expressions<a class="headerlink" href="#pep-289-generator-expressions" title="Permalink to this headline">¶</a></h2>
The iterator feature introduced in Python 2.2 and the <a class="reference internal" href="../library/itertools.html#module-itertools" title="itertools: Functions creating iterators for efficient looping."><tt class="xref py py-mod docutils literal">itertools</tt></a> module
make it easier to write programs that loop through large data sets without
having the entire data set in memory at one time. List comprehensions don&#8217;t fit
into this picture very well because they produce a Python list object containing
all of the items. This unavoidably pulls all of the objects into memory, which
can be a problem if your data set is very large. When trying to write a
functionally-styled program, it would be natural to write something like:
<div class="highlight-python"><div class="highlight"><pre>links = [link for link in get_all_links() if not link.followed]
for link in links:
 ...
</pre></div>
</div>
instead of
<div class="highlight-python"><div class="highlight"><pre>for link in get_all_links():
 if link.followed:
 continue
 ...
</pre></div>
</div>
The first form is more concise and perhaps more readable, but if you&#8217;re dealing
with a large number of link objects you&#8217;d have to write the second form to avoid
having all link objects in memory at the same time.
Generator expressions work similarly to list comprehensions but don&#8217;t
materialize the entire list; instead they create a generator that will return
elements one by one. The above example could be written as:
<div class="highlight-python"><div class="highlight"><pre>links = (link for link in get_all_links() if not link.followed)
for link in links:
 ...
</pre></div>
</div>
Generator expressions always have to be written inside parentheses, as in the
above example. The parentheses signalling a function call also count, so if you
want to create an iterator that will be immediately passed to a function you
could write:
<div class="highlight-python"><div class="highlight"><pre>print sum(obj.count for obj in list_all_objects())
</pre></div>
</div>
Generator expressions differ from list comprehensions in various small ways.
Most notably, the loop variable (obj in the above example) is not accessible
outside of the generator expression. List comprehensions leave the variable
assigned to its last value; future versions of Python will change this, making
list comprehensions match generator expressions in this respect.
<div class="admonition-see-also admonition seealso">
See also
<dl class="last docutils">
<dt><a class="pep reference external" href="http://www.python.org/dev/peps/pep-0289">PEP 289</a> - Generator Expressions</dt>
<dd>Proposed by Raymond Hettinger and implemented by Jiwon Seo with early efforts
steered by Hye-Shik Chang.</dd>
</dl>
</div>
</div>
<div class="section" id="pep-292-simpler-string-substitutions">
<h2>PEP 292: Simpler String Substitutions<a class="headerlink" href="#pep-292-simpler-string-substitutions" title="Permalink to this headline">¶</a></h2>
Some new classes in the standard library provide an alternative mechanism for
substituting variables into strings; this style of substitution may be better
for applications where untrained users need to edit templates.
The usual way of substituting variables by name is the <tt class="docutils literal">%</tt> operator:
<div class="highlight-python"><div class="highlight"><pre>&gt;&gt;&gt; &#39;%(page)i: %(title)s&#39; % {&#39;page&#39;:2, &#39;title&#39;: &#39;The Best of Times&#39;}
&#39;2: The Best of Times&#39;
</pre></div>
</div>
When writing the template string, it can be easy to forget the <tt class="docutils literal">i</tt> or <tt class="docutils literal">s</tt>
after the closing parenthesis. This isn&#8217;t a big problem if the template is in a
Python module, because you run the code, get an &#8220;Unsupported format character&#8221;
<a class="reference internal" href="../library/exceptions.html#exceptions.ValueError" title="exceptions.ValueError"><tt class="xref py py-exc docutils literal">ValueError</tt></a>, and fix the problem. However, consider an application such
as Mailman where template strings or translations are being edited by users who
aren&#8217;t aware of the Python language. The format string&#8217;s syntax is complicated
to explain to such users, and if they make a mistake, it&#8217;s difficult to provide
helpful feedback to them.
PEP 292 adds a <tt class="xref py py-class docutils literal">Template</tt> class to the <a class="reference internal" href="../library/string.html#module-string" title="string: Common string operations."><tt class="xref py py-mod docutils literal">string</tt></a> module that uses
<tt class="docutils literal">$</tt> to indicate a substitution:
<div class="highlight-python"><div class="highlight"><pre>&gt;&gt;&gt; import string
&gt;&gt;&gt; t = string.Template(&#39;$page: $title&#39;)
&gt;&gt;&gt; t.substitute({&#39;page&#39;:2, &#39;title&#39;: &#39;The Best of Times&#39;})
&#39;2: The Best of Times&#39;
</pre></div>
</div>
If a key is missing from the dictionary, the <tt class="xref py py-meth docutils literal">substitute()</tt> method will
raise a <a class="reference internal" href="../library/exceptions.html#exceptions.KeyError" title="exceptions.KeyError"><tt class="xref py py-exc docutils literal">KeyError</tt></a>. There&#8217;s also a <tt class="xref py py-meth docutils literal">safe_substitute()</tt> method that
ignores missing keys:
<div class="highlight-python"><div class="highlight"><pre>&gt;&gt;&gt; t = string.Template(&#39;$page: $title&#39;)
&gt;&gt;&gt; t.safe_substitute({&#39;page&#39;:3})
&#39;3: $title&#39;
</pre></div>
</div>
<div class="admonition-see-also admonition seealso">
See also
<dl class="last docutils">
<dt><a class="pep reference external" href="http://www.python.org/dev/peps/pep-0292">PEP 292</a> - Simpler String Substitutions</dt>
<dd>Written and implemented by Barry Warsaw.</dd>
</dl>
</div>
</div>
<div class="section" id="pep-318-decorators-for-functions-and-methods">
<h2>PEP 318: Decorators for Functions and Methods<a class="headerlink" href="#pep-318-decorators-for-functions-and-methods" title="Permalink to this headline">¶</a></h2>
Python 2.2 extended Python&#8217;s object model by adding static methods and class
methods, but it didn&#8217;t extend Python&#8217;s syntax to provide any new way of defining
static or class methods. Instead, you had to write a <a class="reference internal" href="../reference/compound_stmts.html#def"><tt class="xref std std-keyword docutils literal">def</tt></a> statement
in the usual way, and pass the resulting method to a <a class="reference internal" href="../library/functions.html#staticmethod" title="staticmethod"><tt class="xref py py-func docutils literal">staticmethod()</tt></a> or
<a class="reference internal" href="../library/functions.html#classmethod" title="classmethod"><tt class="xref py py-func docutils literal">classmethod()</tt></a> function that would wrap up the function as a method of the
new type. Your code would look like this:
<div class="highlight-python"><div class="highlight"><pre>class C:
 def meth (cls):
 ...

meth = classmethod(meth) # Rebind name to wrapped-up class method
</pre></div>
</div>
If the method was very long, it would be easy to miss or forget the
<a class="reference internal" href="../library/functions.html#classmethod" title="classmethod"><tt class="xref py py-func docutils literal">classmethod()</tt></a> invocation after the function body.
The intention was always to add some syntax to make such definitions more
readable, but at the time of 2.2&#8217;s release a good syntax was not obvious. Today
a good syntax still isn&#8217;t obvious but users are asking for easier access to
the feature; a new syntactic feature has been added to meet this need.
The new feature is called &#8220;function decorators&#8221;. The name comes from the idea
that <a class="reference internal" href="../library/functions.html#classmethod" title="classmethod"><tt class="xref py py-func docutils literal">classmethod()</tt></a>, <a class="reference internal" href="../library/functions.html#staticmethod" title="staticmethod"><tt class="xref py py-func docutils literal">staticmethod()</tt></a>, and friends are storing
additional information on a function object; they&#8217;re decorating functions with
more details.
The notation borrows from Java and uses the <tt class="docutils literal">'&#64;'</tt> character as an indicator.
Using the new syntax, the example above would be written:
<div class="highlight-python"><div class="highlight"><pre>class C:

@classmethod
 def meth (cls):
 ...
</pre></div>
</div>
The <tt class="docutils literal">&#64;classmethod</tt> is shorthand for the <tt class="docutils literal">meth=classmethod(meth)</tt> assignment.
More generally, if you have the following:
<div class="highlight-python"><div class="highlight"><pre>@A
@B
@C
def f ():
 ...
</pre></div>
</div>
It&#8217;s equivalent to the following pre-decorator code:
<div class="highlight-python"><div class="highlight"><pre>def f(): ...
f = A(B(C(f)))
</pre></div>
</div>
Decorators must come on the line before a function definition, one decorator per
line, and can&#8217;t be on the same line as the def statement, meaning that <tt class="docutils literal">&#64;A def
f(): ...</tt> is illegal. You can only decorate function definitions, either at
the module level or inside a class; you can&#8217;t decorate class definitions.
A decorator is just a function that takes the function to be decorated as an
argument and returns either the same function or some new object. The return
value of the decorator need not be callable (though it typically is), unless
further decorators will be applied to the result. It&#8217;s easy to write your own
decorators. The following simple example just sets an attribute on the function
object:
<div class="highlight-python"><div class="highlight"><pre>&gt;&gt;&gt; def deco(func):
... func.attr = &#39;decorated&#39;
... return func
...
&gt;&gt;&gt; @deco
... def f(): pass
...
&gt;&gt;&gt; f
&lt;function f at 0x402ef0d4&gt;
&gt;&gt;&gt; f.attr
&#39;decorated&#39;
&gt;&gt;&gt;
</pre></div>
</div>
As a slightly more realistic example, the following decorator checks that the
supplied argument is an integer:
<div class="highlight-python"><div class="highlight"><pre>def require_int (func):
 def wrapper (arg):
 assert isinstance(arg, int)
 return func(arg)

return wrapper

@require_int
def p1 (arg):
 print arg

@require_int
def p2(arg):
 print arg*2
</pre></div>
</div>
An example in <a class="pep reference external" href="http://www.python.org/dev/peps/pep-0318">PEP 318</a> contains a fancier version of this idea that lets you
both specify the required type and check the returned type.
Decorator functions can take arguments. If arguments are supplied, your
decorator function is called with only those arguments and must return a new
decorator function; this function must take a single function and return a
function, as previously described. In other words, <tt class="docutils literal">&#64;A &#64;B &#64;C(args)</tt> becomes:
<div class="highlight-python"><div class="highlight"><pre>def f(): ...
_deco = C(args)
f = A(B(_deco(f)))
</pre></div>
</div>
Getting this right can be slightly brain-bending, but it&#8217;s not too difficult.
A small related change makes the <tt class="xref py py-attr docutils literal">func_name</tt> attribute of functions
writable. This attribute is used to display function names in tracebacks, so
decorators should change the name of any new function that&#8217;s constructed and
returned.
<div class="admonition-see-also admonition seealso">
See also
<dl class="last docutils">
<dt><a class="pep reference external" href="http://www.python.org/dev/peps/pep-0318">PEP 318</a> - Decorators for Functions, Methods and Classes</dt>
<dd>Written by Kevin D. Smith, Jim Jewett, and Skip Montanaro. Several people
wrote patches implementing function decorators, but the one that was actually
checked in was patch #979728, written by Mark Russell.</dd>
<dt><a class="reference external" href="http://www.python.org/moin/PythonDecoratorLibrary">http://www.python.org/moin/PythonDecoratorLibrary</a></dt>
<dd>This Wiki page contains several examples of decorators.</dd>
</dl>
</div>
</div>
<div class="section" id="pep-322-reverse-iteration">
<h2>PEP 322: Reverse Iteration<a class="headerlink" href="#pep-322-reverse-iteration" title="Permalink to this headline">¶</a></h2>
A new built-in function, <tt class="xref py py-func docutils literal">reversed(seq)()</tt>, takes a sequence and returns an
iterator that loops over the elements of the sequence in reverse order.
<div class="highlight-python"><div class="highlight"><pre>&gt;&gt;&gt; for i in reversed(xrange(1,4)):
... print i
...
3
2
1
</pre></div>
</div>
Compared to extended slicing, such as <tt class="docutils literal">range(1,4)[::-1]</tt>, <a class="reference internal" href="../library/functions.html#reversed" title="reversed"><tt class="xref py py-func docutils literal">reversed()</tt></a> is
easier to read, runs faster, and uses substantially less memory.
Note that <a class="reference internal" href="../library/functions.html#reversed" title="reversed"><tt class="xref py py-func docutils literal">reversed()</tt></a> only accepts sequences, not arbitrary iterators. If
you want to reverse an iterator, first convert it to a list with <a class="reference internal" href="../library/functions.html#list" title="list"><tt class="xref py py-func docutils literal">list()</tt></a>.
<div class="highlight-python"><div class="highlight"><pre>&gt;&gt;&gt; input = open(&#39;/etc/passwd&#39;, &#39;r&#39;)
&gt;&gt;&gt; for line in reversed(list(input)):
... print line
...
root:*:0:0:System Administrator:/var/root:/bin/tcsh
 ...
</pre></div>
</div>
<div class="admonition-see-also admonition seealso">
See also
<dl class="last docutils">
<dt><a class="pep reference external" href="http://www.python.org/dev/peps/pep-0322">PEP 322</a> - Reverse Iteration</dt>
<dd>Written and implemented by Raymond Hettinger.</dd>
</dl>
</div>
</div>
<div class="section" id="pep-324-new-subprocess-module">
<h2>PEP 324: New subprocess Module<a class="headerlink" href="#pep-324-new-subprocess-module" title="Permalink to this headline">¶</a></h2>
The standard library provides a number of ways to execute a subprocess, offering
different features and different levels of complexity.
<tt class="xref py py-func docutils literal">os.system(command)()</tt> is easy to use, but slow (it runs a shell process
which executes the command) and dangerous (you have to be careful about escaping
the shell&#8217;s metacharacters). The <a class="reference internal" href="../library/popen2.html#module-popen2" title="popen2: Subprocesses with accessible standard I/O streams. (deprecated)"><tt class="xref py py-mod docutils literal">popen2</tt></a> module offers classes that can
capture standard output and standard error from the subprocess, but the naming
is confusing. The <a class="reference internal" href="../library/subprocess.html#module-subprocess" title="subprocess: Subprocess management."><tt class="xref py py-mod docutils literal">subprocess</tt></a> module cleans this up, providing a unified
interface that offers all the features you might need.
Instead of <a class="reference internal" href="../library/popen2.html#module-popen2" title="popen2: Subprocesses with accessible standard I/O streams. (deprecated)"><tt class="xref py py-mod docutils literal">popen2</tt></a>&#8216;s collection of classes, <a class="reference internal" href="../library/subprocess.html#module-subprocess" title="subprocess: Subprocess management."><tt class="xref py py-mod docutils literal">subprocess</tt></a> contains a
single class called <tt class="xref py py-class docutils literal">Popen</tt> whose constructor supports a number of
different keyword arguments.
<div class="highlight-python"><pre>class Popen(args, bufsize=0, executable=None,
 stdin=None, stdout=None, stderr=None,
 preexec_fn=None, close_fds=False, shell=False,
 cwd=None, env=None, universal_newlines=False,
 startupinfo=None, creationflags=0):</pre>
</div>
args is commonly a sequence of strings that will be the arguments to the
program executed as the subprocess. (If the shell argument is true, args
can be a string which will then be passed on to the shell for interpretation,
just as <a class="reference internal" href="../library/os.html#os.system" title="os.system"><tt class="xref py py-func docutils literal">os.system()</tt></a> does.)
stdin, stdout, and stderr specify what the subprocess&#8217;s input, output, and
error streams will be. You can provide a file object or a file descriptor, or
you can use the constant <tt class="docutils literal">subprocess.PIPE</tt> to create a pipe between the
subprocess and the parent.
The constructor has a number of handy options:
<ul class="simple">
<li>close_fds requests that all file descriptors be closed before running the
subprocess.</li>
<li>cwd specifies the working directory in which the subprocess will be executed
(defaulting to whatever the parent&#8217;s working directory is).</li>
<li>env is a dictionary specifying environment variables.</li>
<li>preexec_fn is a function that gets called before the child is started.</li>
<li>universal_newlines opens the child&#8217;s input and output using Python&#8217;s
<a class="reference internal" href="../glossary.html#term-universal-newlines">universal newlines</a> feature.</li>
</ul>
Once you&#8217;ve created the <tt class="xref py py-class docutils literal">Popen</tt> instance, you can call its <tt class="xref py py-meth docutils literal">wait()</tt>
method to pause until the subprocess has exited, <tt class="xref py py-meth docutils literal">poll()</tt> to check if it&#8217;s
exited without pausing, or <tt class="xref py py-meth docutils literal">communicate(data)()</tt> to send the string data
to the subprocess&#8217;s standard input. <tt class="xref py py-meth docutils literal">communicate(data)()</tt> then reads any
data that the subprocess has sent to its standard output or standard error,
returning a tuple <tt class="docutils literal">(stdout_data, stderr_data)</tt>.
<tt class="xref py py-func docutils literal">call()</tt> is a shortcut that passes its arguments along to the <tt class="xref py py-class docutils literal">Popen</tt>
constructor, waits for the command to complete, and returns the status code of
the subprocess. It can serve as a safer analog to <a class="reference internal" href="../library/os.html#os.system" title="os.system"><tt class="xref py py-func docutils literal">os.system()</tt></a>:
<div class="highlight-python"><div class="highlight"><pre>sts = subprocess.call([&#39;dpkg&#39;, &#39;-i&#39;, &#39;/tmp/new-package.deb&#39;])
if sts == 0:
 # Success
 ...
else:
 # dpkg returned an error
 ...
</pre></div>
</div>
The command is invoked without use of the shell. If you really do want to use
the shell, you can add <tt class="docutils literal">shell=True</tt> as a keyword argument and provide a string
instead of a sequence:
<div class="highlight-python"><div class="highlight"><pre>sts = subprocess.call(&#39;dpkg -i /tmp/new-package.deb&#39;, shell=True)
</pre></div>
</div>
The PEP takes various examples of shell and Python code and shows how they&#8217;d be
translated into Python code that uses <a class="reference internal" href="../library/subprocess.html#module-subprocess" title="subprocess: Subprocess management."><tt class="xref py py-mod docutils literal">subprocess</tt></a>. Reading this section
of the PEP is highly recommended.
<div class="admonition-see-also admonition seealso">
See also
<dl class="last docutils">
<dt><a class="pep reference external" href="http://www.python.org/dev/peps/pep-0324">PEP 324</a> - subprocess - New process module</dt>
<dd>Written and implemented by Peter Åstrand, with assistance from Fredrik Lundh and
others.</dd>
</dl>
</div>
</div>
<div class="section" id="pep-327-decimal-data-type">
<h2>PEP 327: Decimal Data Type<a class="headerlink" href="#pep-327-decimal-data-type" title="Permalink to this headline">¶</a></h2>
Python has always supported floating-point (FP) numbers, based on the underlying
C <tt class="xref c c-type docutils literal">double</tt> type, as a data type. However, while most programming
languages provide a floating-point type, many people (even programmers) are
unaware that floating-point numbers don&#8217;t represent certain decimal fractions
accurately. The new <tt class="xref py py-class docutils literal">Decimal</tt> type can represent these fractions
accurately, up to a user-specified precision limit.
<div class="section" id="why-is-decimal-needed">
<h3>Why is Decimal needed?<a class="headerlink" href="#why-is-decimal-needed" title="Permalink to this headline">¶</a></h3>
The limitations arise from the representation used for floating-point numbers.
FP numbers are made up of three components:
<ul class="simple">
<li>The sign, which is positive or negative.</li>
<li>The mantissa, which is a single-digit binary number followed by a fractional
part. For example, <tt class="docutils literal">1.01</tt> in base-2 notation is <tt class="docutils literal">1 + 0/2 + 1/4</tt>, or 1.25 in
decimal notation.</li>
<li>The exponent, which tells where the decimal point is located in the number
represented.</li>
</ul>
For example, the number 1.25 has positive sign, a mantissa value of 1.01 (in
binary), and an exponent of 0 (the decimal point doesn&#8217;t need to be shifted).
The number 5 has the same sign and mantissa, but the exponent is 2 because the
mantissa is multiplied by 4 (2 to the power of the exponent 2); 1.25 * 4 equals
5.
Modern systems usually provide floating-point support that conforms to a
standard called IEEE 754. C&#8217;s <tt class="xref c c-type docutils literal">double</tt> type is usually implemented as a
64-bit IEEE 754 number, which uses 52 bits of space for the mantissa. This
means that numbers can only be specified to 52 bits of precision. If you&#8217;re
trying to represent numbers whose expansion repeats endlessly, the expansion is
cut off after 52 bits. Unfortunately, most software needs to produce output in
base 10, and common fractions in base 10 are often repeating decimals in binary.
For example, 1.1 decimal is binary <tt class="docutils literal">1.0001100110011 ...</tt>; .1 = 1/16 + 1/32 +
1/256 plus an infinite number of additional terms. IEEE 754 has to chop off
that infinitely repeated decimal after 52 digits, so the representation is
slightly inaccurate.
Sometimes you can see this inaccuracy when the number is printed:
<div class="highlight-python"><div class="highlight"><pre>&gt;&gt;&gt; 1.1
1.1000000000000001
</pre></div>
</div>
The inaccuracy isn&#8217;t always visible when you print the number because the FP-to-
decimal-string conversion is provided by the C library, and most C libraries try
to produce sensible output. Even if it&#8217;s not displayed, however, the inaccuracy
is still there and subsequent operations can magnify the error.
For many applications this doesn&#8217;t matter. If I&#8217;m plotting points and
displaying them on my monitor, the difference between 1.1 and 1.1000000000000001
is too small to be visible. Reports often limit output to a certain number of
decimal places, and if you round the number to two or three or even eight
decimal places, the error is never apparent. However, for applications where it
does matter, it&#8217;s a lot of work to implement your own custom arithmetic
routines.
Hence, the <tt class="xref py py-class docutils literal">Decimal</tt> type was created.
</div>
<div class="section" id="the-decimal-type">
<h3>The <tt class="xref py py-class docutils literal">Decimal</tt> type<a class="headerlink" href="#the-decimal-type" title="Permalink to this headline">¶</a></h3>
A new module, <a class="reference internal" href="../library/decimal.html#module-decimal" title="decimal: Implementation of the General Decimal Arithmetic Specification."><tt class="xref py py-mod docutils literal">decimal</tt></a>, was added to Python&#8217;s standard library. It
contains two classes, <tt class="xref py py-class docutils literal">Decimal</tt> and <tt class="xref py py-class docutils literal">Context</tt>. <tt class="xref py py-class docutils literal">Decimal</tt>
instances represent numbers, and <tt class="xref py py-class docutils literal">Context</tt> instances are used to wrap up
various settings such as the precision and default rounding mode.
<tt class="xref py py-class docutils literal">Decimal</tt> instances are immutable, like regular Python integers and FP
numbers; once it&#8217;s been created, you can&#8217;t change the value an instance
represents. <tt class="xref py py-class docutils literal">Decimal</tt> instances can be created from integers or
strings:
<div class="highlight-python"><div class="highlight"><pre>&gt;&gt;&gt; import decimal
&gt;&gt;&gt; decimal.Decimal(1972)
Decimal(&quot;1972&quot;)
&gt;&gt;&gt; decimal.Decimal(&quot;1.1&quot;)
Decimal(&quot;1.1&quot;)
</pre></div>
</div>
You can also provide tuples containing the sign, the mantissa represented as a
tuple of decimal digits, and the exponent:
<div class="highlight-python"><div class="highlight"><pre>&gt;&gt;&gt; decimal.Decimal((1, (1, 4, 7, 5), -2))
Decimal(&quot;-14.75&quot;)
</pre></div>
</div>
Cautionary note: the sign bit is a Boolean value, so 0 is positive and 1 is
negative.
Converting from floating-point numbers poses a bit of a problem: should the FP
number representing 1.1 turn into the decimal number for exactly 1.1, or for 1.1
plus whatever inaccuracies are introduced? The decision was to dodge the issue
and leave such a conversion out of the API. Instead, you should convert the
floating-point number into a string using the desired precision and pass the
string to the <tt class="xref py py-class docutils literal">Decimal</tt> constructor:
<div class="highlight-python"><div class="highlight"><pre>&gt;&gt;&gt; f = 1.1
&gt;&gt;&gt; decimal.Decimal(str(f))
Decimal(&quot;1.1&quot;)
&gt;&gt;&gt; decimal.Decimal(&#39;%.12f&#39; % f)
Decimal(&quot;1.100000000000&quot;)
</pre></div>
</div>
Once you have <tt class="xref py py-class docutils literal">Decimal</tt> instances, you can perform the usual mathematical
operations on them. One limitation: exponentiation requires an integer
exponent:
<div class="highlight-python"><div class="highlight"><pre>&gt;&gt;&gt; a = decimal.Decimal(&#39;35.72&#39;)
&gt;&gt;&gt; b = decimal.Decimal(&#39;1.73&#39;)
&gt;&gt;&gt; a+b
Decimal(&quot;37.45&quot;)
&gt;&gt;&gt; a-b
Decimal(&quot;33.99&quot;)
&gt;&gt;&gt; a*b
Decimal(&quot;61.7956&quot;)
&gt;&gt;&gt; a/b
Decimal(&quot;20.64739884393063583815028902&quot;)
&gt;&gt;&gt; a ** 2
Decimal(&quot;1275.9184&quot;)
&gt;&gt;&gt; a**b
Traceback (most recent call last):
 ...
decimal.InvalidOperation: x ** (non-integer)
</pre></div>
</div>
You can combine <tt class="xref py py-class docutils literal">Decimal</tt> instances with integers, but not with floating-
point numbers:
<div class="highlight-python"><div class="highlight"><pre>&gt;&gt;&gt; a + 4
Decimal(&quot;39.72&quot;)
&gt;&gt;&gt; a + 4.5
Traceback (most recent call last):
 ...
TypeError: You can interact Decimal only with int, long or Decimal data types.
&gt;&gt;&gt;
</pre></div>
</div>
<tt class="xref py py-class docutils literal">Decimal</tt> numbers can be used with the <a class="reference internal" href="../library/math.html#module-math" title="math: Mathematical functions (sin() etc.)."><tt class="xref py py-mod docutils literal">math</tt></a> and <a class="reference internal" href="../library/cmath.html#module-cmath" title="cmath: Mathematical functions for complex numbers."><tt class="xref py py-mod docutils literal">cmath</tt></a>
modules, but note that they&#8217;ll be immediately converted to floating-point
numbers before the operation is performed, resulting in a possible loss of
precision and accuracy. You&#8217;ll also get back a regular floating-point number
and not a <tt class="xref py py-class docutils literal">Decimal</tt>.
<div class="highlight-python"><div class="highlight"><pre>&gt;&gt;&gt; import math, cmath
&gt;&gt;&gt; d = decimal.Decimal(&#39;123456789012.345&#39;)
&gt;&gt;&gt; math.sqrt(d)
351364.18288201344
&gt;&gt;&gt; cmath.sqrt(-d)
351364.18288201344j
</pre></div>
</div>
<tt class="xref py py-class docutils literal">Decimal</tt> instances have a <tt class="xref py py-meth docutils literal">sqrt()</tt> method that returns a
<tt class="xref py py-class docutils literal">Decimal</tt>, but if you need other things such as trigonometric functions
you&#8217;ll have to implement them.
<div class="highlight-python"><div class="highlight"><pre>&gt;&gt;&gt; d.sqrt()
Decimal(&quot;351364.1828820134592177245001&quot;)
</pre></div>
</div>
</div>
<div class="section" id="the-context-type">
<h3>The <tt class="xref py py-class docutils literal">Context</tt> type<a class="headerlink" href="#the-context-type" title="Permalink to this headline">¶</a></h3>
Instances of the <tt class="xref py py-class docutils literal">Context</tt> class encapsulate several settings for
decimal operations:
<ul class="simple">
<li><tt class="xref py py-attr docutils literal">prec</tt> is the precision, the number of decimal places.</li>
<li><tt class="xref py py-attr docutils literal">rounding</tt> specifies the rounding mode. The <a class="reference internal" href="../library/decimal.html#module-decimal" title="decimal: Implementation of the General Decimal Arithmetic Specification."><tt class="xref py py-mod docutils literal">decimal</tt></a> module has
constants for the various possibilities: <tt class="xref py py-const docutils literal">ROUND_DOWN</tt>,
<tt class="xref py py-const docutils literal">ROUND_CEILING</tt>, <tt class="xref py py-const docutils literal">ROUND_HALF_EVEN</tt>, and various others.</li>
<li><tt class="xref py py-attr docutils literal">traps</tt> is a dictionary specifying what happens on encountering certain
error conditions: either an exception is raised or a value is returned. Some
examples of error conditions are division by zero, loss of precision, and
overflow.</li>
</ul>
There&#8217;s a thread-local default context available by calling <tt class="xref py py-func docutils literal">getcontext()</tt>;
you can change the properties of this context to alter the default precision,
rounding, or trap handling. The following example shows the effect of changing
the precision of the default context:
<div class="highlight-python"><div class="highlight"><pre>&gt;&gt;&gt; decimal.getcontext().prec
28
&gt;&gt;&gt; decimal.Decimal(1) / decimal.Decimal(7)
Decimal(&quot;0.1428571428571428571428571429&quot;)
&gt;&gt;&gt; decimal.getcontext().prec = 9
&gt;&gt;&gt; decimal.Decimal(1) / decimal.Decimal(7)
Decimal(&quot;0.142857143&quot;)
</pre></div>
</div>
The default action for error conditions is selectable; the module can either
return a special value such as infinity or not-a-number, or exceptions can be
raised:
<div class="highlight-python"><div class="highlight"><pre>&gt;&gt;&gt; decimal.Decimal(1) / decimal.Decimal(0)
Traceback (most recent call last):
 ...
decimal.DivisionByZero: x / 0
&gt;&gt;&gt; decimal.getcontext().traps[decimal.DivisionByZero] = False
&gt;&gt;&gt; decimal.Decimal(1) / decimal.Decimal(0)
Decimal(&quot;Infinity&quot;)
&gt;&gt;&gt;
</pre></div>
</div>
The <tt class="xref py py-class docutils literal">Context</tt> instance also has various methods for formatting numbers
such as <tt class="xref py py-meth docutils literal">to_eng_string()</tt> and <tt class="xref py py-meth docutils literal">to_sci_string()</tt>.
For more information, see the documentation for the <a class="reference internal" href="../library/decimal.html#module-decimal" title="decimal: Implementation of the General Decimal Arithmetic Specification."><tt class="xref py py-mod docutils literal">decimal</tt></a> module, which
includes a quick-start tutorial and a reference.
<div class="admonition-see-also admonition seealso">
See also
<dl class="last docutils">
<dt><a class="pep reference external" href="http://www.python.org/dev/peps/pep-0327">PEP 327</a> - Decimal Data Type</dt>
<dd>Written by Facundo Batista and implemented by Facundo Batista, Eric Price,
Raymond Hettinger, Aahz, and Tim Peters.</dd>
<dt><a class="reference external" href="http://www.lahey.com/float.htm">http://www.lahey.com/float.htm</a></dt>
<dd>The article uses Fortran code to illustrate many of the problems that floating-
point inaccuracy can cause.</dd>
<dt><a class="reference external" href="http://www2.hursley.ibm.com/decimal/">http://www2.hursley.ibm.com/decimal/</a></dt>
<dd>A description of a decimal-based representation. This representation is being
proposed as a standard, and underlies the new Python decimal type. Much of this
material was written by Mike Cowlishaw, designer of the Rexx language.</dd>
</dl>
</div>
</div>
</div>
<div class="section" id="pep-328-multi-line-imports">
<h2>PEP 328: Multi-line Imports<a class="headerlink" href="#pep-328-multi-line-imports" title="Permalink to this headline">¶</a></h2>
One language change is a small syntactic tweak aimed at making it easier to
import many names from a module. In a <tt class="docutils literal">from module import names</tt> statement,
names is a sequence of names separated by commas. If the sequence is very
long, you can either write multiple imports from the same module, or you can use
backslashes to escape the line endings like this:
<div class="highlight-python"><div class="highlight"><pre>from SimpleXMLRPCServer import SimpleXMLRPCServer,\
 SimpleXMLRPCRequestHandler,\
 CGIXMLRPCRequestHandler,\
 resolve_dotted_attribute
</pre></div>
</div>
The syntactic change in Python 2.4 simply allows putting the names within
parentheses. Python ignores newlines within a parenthesized expression, so the
backslashes are no longer needed:
<div class="highlight-python"><div class="highlight"><pre>from SimpleXMLRPCServer import (SimpleXMLRPCServer,
 SimpleXMLRPCRequestHandler,
 CGIXMLRPCRequestHandler,
 resolve_dotted_attribute)
</pre></div>
</div>
The PEP also proposes that all <a class="reference internal" href="../reference/simple_stmts.html#import"><tt class="xref std std-keyword docutils literal">import</tt></a> statements be absolute imports,
with a leading <tt class="docutils literal">.</tt> character to indicate a relative import. This part of the
PEP was not implemented for Python 2.4, but was completed for Python 2.5.
<div class="admonition-see-also admonition seealso">
See also
<dl class="last docutils">
<dt><a class="pep reference external" href="http://www.python.org/dev/peps/pep-0328">PEP 328</a> - Imports: Multi-Line and Absolute/Relative</dt>
<dd>Written by Aahz. Multi-line imports were implemented by Dima Dorfman.</dd>
</dl>
</div>
</div>
<div class="section" id="pep-331-locale-independent-float-string-conversions">
<h2>PEP 331: Locale-Independent Float/String Conversions<a class="headerlink" href="#pep-331-locale-independent-float-string-conversions" title="Permalink to this headline">¶</a></h2>
The <a class="reference internal" href="../library/locale.html#module-locale" title="locale: Internationalization services."><tt class="xref py py-mod docutils literal">locale</tt></a> modules lets Python software select various conversions and
display conventions that are localized to a particular country or language.
However, the module was careful to not change the numeric locale because various
functions in Python&#8217;s implementation required that the numeric locale remain set
to the <tt class="docutils literal">'C'</tt> locale. Often this was because the code was using the C
library&#8217;s <tt class="xref c c-func docutils literal">atof()</tt> function.
Not setting the numeric locale caused trouble for extensions that used third-
party C libraries, however, because they wouldn&#8217;t have the correct locale set.
The motivating example was GTK+, whose user interface widgets weren&#8217;t displaying
numbers in the current locale.
The solution described in the PEP is to add three new functions to the Python
API that perform ASCII-only conversions, ignoring the locale setting:
<ul class="simple">
<li><tt class="xref c c-func docutils literal">PyOS_ascii_strtod(str, ptr)()</tt> and <tt class="xref c c-func docutils literal">PyOS_ascii_atof(str, ptr)()</tt>
both convert a string to a C <tt class="xref c c-type docutils literal">double</tt>.</li>
<li><tt class="xref c c-func docutils literal">PyOS_ascii_formatd(buffer, buf_len, format, d)()</tt> converts a
<tt class="xref c c-type docutils literal">double</tt> to an ASCII string.</li>
</ul>
The code for these functions came from the GLib library
(<a class="reference external" href="http://library.gnome.org/devel/glib/stable/">http://library.gnome.org/devel/glib/stable/</a>), whose developers kindly
relicensed the relevant functions and donated them to the Python Software
Foundation. The <a class="reference internal" href="../library/locale.html#module-locale" title="locale: Internationalization services."><tt class="xref py py-mod docutils literal">locale</tt></a> module can now change the numeric locale,
letting extensions such as GTK+ produce the correct results.
<div class="admonition-see-also admonition seealso">
See also
<dl class="last docutils">
<dt><a class="pep reference external" href="http://www.python.org/dev/peps/pep-0331">PEP 331</a> - Locale-Independent Float/String Conversions</dt>
<dd>Written by Christian R. Reis, and implemented by Gustavo Carneiro.</dd>
</dl>
</div>
</div>
<div class="section" id="other-language-changes">
<h2>Other Language Changes<a class="headerlink" href="#other-language-changes" title="Permalink to this headline">¶</a></h2>
Here are all of the changes that Python 2.4 makes to the core Python language.
<ul>
<li>Decorators for functions and methods were added (<a class="pep reference external" href="http://www.python.org/dev/peps/pep-0318">PEP 318</a>).
</li>
<li>Built-in <a class="reference internal" href="../library/stdtypes.html#set" title="set"><tt class="xref py py-func docutils literal">set()</tt></a> and <a class="reference internal" href="../library/stdtypes.html#frozenset" title="frozenset"><tt class="xref py py-func docutils literal">frozenset()</tt></a> types were added (<a class="pep reference external" href="http://www.python.org/dev/peps/pep-0218">PEP 218</a>).
Other new built-ins include the <tt class="xref py py-func docutils literal">reversed(seq)()</tt> function (<a class="pep reference external" href="http://www.python.org/dev/peps/pep-0322">PEP 322</a>).
</li>
<li>Generator expressions were added (<a class="pep reference external" href="http://www.python.org/dev/peps/pep-0289">PEP 289</a>).
</li>
<li>Certain numeric expressions no longer return values restricted to 32 or 64
bits (<a class="pep reference external" href="http://www.python.org/dev/peps/pep-0237">PEP 237</a>).
</li>
<li>You can now put parentheses around the list of names in a <tt class="docutils literal">from module import
names</tt> statement (<a class="pep reference external" href="http://www.python.org/dev/peps/pep-0328">PEP 328</a>).
</li>
<li>The <a class="reference internal" href="../library/stdtypes.html#dict.update" title="dict.update"><tt class="xref py py-meth docutils literal">dict.update()</tt></a> method now accepts the same argument forms as the
<a class="reference internal" href="../library/stdtypes.html#dict" title="dict"><tt class="xref py py-class docutils literal">dict</tt></a> constructor. This includes any mapping, any iterable of key/value
pairs, and keyword arguments. (Contributed by Raymond Hettinger.)
</li>
<li>The string methods <tt class="xref py py-meth docutils literal">ljust()</tt>, <tt class="xref py py-meth docutils literal">rjust()</tt>, and <tt class="xref py py-meth docutils literal">center()</tt> now take
an optional argument for specifying a fill character other than a space.
(Contributed by Raymond Hettinger.)
</li>
<li>Strings also gained an <tt class="xref py py-meth docutils literal">rsplit()</tt> method that works like the <tt class="xref py py-meth docutils literal">split()</tt>
method but splits from the end of the string. (Contributed by Sean
Reifschneider.)
<div class="highlight-python"><div class="highlight"><pre>&gt;&gt;&gt; &#39;www.python.org&#39;.split(&#39;.&#39;, 1)
[&#39;www&#39;, &#39;python.org&#39;]
&#39;www.python.org&#39;.rsplit(&#39;.&#39;, 1)
[&#39;www.python&#39;, &#39;org&#39;]
</pre></div>
</div>
</li>
<li>Three keyword parameters, cmp, key, and reverse, were added to the
<tt class="xref py py-meth docutils literal">sort()</tt> method of lists. These parameters make some common usages of
<tt class="xref py py-meth docutils literal">sort()</tt> simpler. All of these parameters are optional.
For the cmp parameter, the value should be a comparison function that takes
two parameters and returns -1, 0, or +1 depending on how the parameters compare.
This function will then be used to sort the list. Previously this was the only
parameter that could be provided to <tt class="xref py py-meth docutils literal">sort()</tt>.
key should be a single-parameter function that takes a list element and
returns a comparison key for the element. The list is then sorted using the
comparison keys. The following example sorts a list case-insensitively:
<div class="highlight-python"><div class="highlight"><pre>&gt;&gt;&gt; L = [&#39;A&#39;, &#39;b&#39;, &#39;c&#39;, &#39;D&#39;]
&gt;&gt;&gt; L.sort() # Case-sensitive sort
&gt;&gt;&gt; L
[&#39;A&#39;, &#39;D&#39;, &#39;b&#39;, &#39;c&#39;]
&gt;&gt;&gt; # Using &#39;key&#39; parameter to sort list
&gt;&gt;&gt; L.sort(key=lambda x: x.lower())
&gt;&gt;&gt; L
[&#39;A&#39;, &#39;b&#39;, &#39;c&#39;, &#39;D&#39;]
&gt;&gt;&gt; # Old-fashioned way
&gt;&gt;&gt; L.sort(cmp=lambda x,y: cmp(x.lower(), y.lower()))
&gt;&gt;&gt; L
[&#39;A&#39;, &#39;b&#39;, &#39;c&#39;, &#39;D&#39;]
</pre></div>
</div>
The last example, which uses the cmp parameter, is the old way to perform a
case-insensitive sort. It works but is slower than using a key parameter.
Using key calls <tt class="xref py py-meth docutils literal">lower()</tt> method once for each element in the list while
using cmp will call it twice for each comparison, so using key saves on
invocations of the <tt class="xref py py-meth docutils literal">lower()</tt> method.
For simple key functions and comparison functions, it is often possible to avoid
a <a class="reference internal" href="../reference/expressions.html#lambda"><tt class="xref std std-keyword docutils literal">lambda</tt></a> expression by using an unbound method instead. For example,
the above case-insensitive sort is best written as:
<div class="highlight-python"><div class="highlight"><pre>&gt;&gt;&gt; L.sort(key=str.lower)
&gt;&gt;&gt; L
[&#39;A&#39;, &#39;b&#39;, &#39;c&#39;, &#39;D&#39;]
</pre></div>
</div>
Finally, the reverse parameter takes a Boolean value. If the value is true,
the list will be sorted into reverse order. Instead of <tt class="docutils literal">L.sort() ;
L.reverse()</tt>, you can now write <tt class="docutils literal">L.sort(reverse=True)</tt>.
The results of sorting are now guaranteed to be stable. This means that two
entries with equal keys will be returned in the same order as they were input.
For example, you can sort a list of people by name, and then sort the list by
age, resulting in a list sorted by age where people with the same age are in
name-sorted order.
(All changes to <tt class="xref py py-meth docutils literal">sort()</tt> contributed by Raymond Hettinger.)
</li>
<li>There is a new built-in function <tt class="xref py py-func docutils literal">sorted(iterable)()</tt> that works like the
in-place <tt class="xref py py-meth docutils literal">list.sort()</tt> method but can be used in expressions. The
differences are:
</li>
<li>the input may be any iterable;
</li>
<li>a newly formed copy is sorted, leaving the original intact; and
</li>
<li>the expression returns the new sorted copy
<div class="highlight-python"><div class="highlight"><pre>&gt;&gt;&gt; L = [9,7,8,3,2,4,1,6,5]
&gt;&gt;&gt; [10+i for i in sorted(L)] # usable in a list comprehension
[11, 12, 13, 14, 15, 16, 17, 18, 19]
&gt;&gt;&gt; L # original is left unchanged
[9,7,8,3,2,4,1,6,5]
&gt;&gt;&gt; sorted(&#39;Monty Python&#39;) # any iterable may be an input
[&#39; &#39;, &#39;M&#39;, &#39;P&#39;, &#39;h&#39;, &#39;n&#39;, &#39;n&#39;, &#39;o&#39;, &#39;o&#39;, &#39;t&#39;, &#39;t&#39;, &#39;y&#39;, &#39;y&#39;]

&gt;&gt;&gt; # List the contents of a dict sorted by key values
&gt;&gt;&gt; colormap = dict(red=1, blue=2, green=3, black=4, yellow=5)
&gt;&gt;&gt; for k, v in sorted(colormap.iteritems()):
... print k, v
...
black 4
blue 2
green 3
red 1
yellow 5
</pre></div>
</div>
(Contributed by Raymond Hettinger.)
</li>
<li>Integer operations will no longer trigger an <tt class="xref py py-exc docutils literal">OverflowWarning</tt>. The
<tt class="xref py py-exc docutils literal">OverflowWarning</tt> warning will disappear in Python 2.5.
</li>
<li>The interpreter gained a new switch, <a class="reference internal" href="../using/cmdline.html#cmdoption-m">-m</a>, that takes a name, searches
for the corresponding module on <tt class="docutils literal">sys.path</tt>, and runs the module as a script.
For example, you can now run the Python profiler with <tt class="docutils literal">python -m profile</tt>.
(Contributed by Nick Coghlan.)
</li>
<li>The <tt class="xref py py-func docutils literal">eval(expr, globals, locals)()</tt> and <tt class="xref py py-func docutils literal">execfile(filename, globals,
locals)()</tt> functions and the <a class="reference internal" href="../reference/simple_stmts.html#exec"><tt class="xref std std-keyword docutils literal">exec</tt></a> statement now accept any mapping type
for the locals parameter. Previously this had to be a regular Python
dictionary. (Contributed by Raymond Hettinger.)
</li>
<li>The <a class="reference internal" href="../library/functions.html#zip" title="zip"><tt class="xref py py-func docutils literal">zip()</tt></a> built-in function and <a class="reference internal" href="../library/itertools.html#itertools.izip" title="itertools.izip"><tt class="xref py py-func docutils literal">itertools.izip()</tt></a> now return an
empty list if called with no arguments. Previously they raised a
<a class="reference internal" href="../library/exceptions.html#exceptions.TypeError" title="exceptions.TypeError"><tt class="xref py py-exc docutils literal">TypeError</tt></a> exception. This makes them more suitable for use with variable
length argument lists:
<div class="highlight-python"><div class="highlight"><pre>&gt;&gt;&gt; def transpose(array):
... return zip(*array)
...
&gt;&gt;&gt; transpose([(1,2,3), (4,5,6)])
[(1, 4), (2, 5), (3, 6)]
&gt;&gt;&gt; transpose([])
[]
</pre></div>
</div>
(Contributed by Raymond Hettinger.)
</li>
<li>Encountering a failure while importing a module no longer leaves a partially-
initialized module object in <tt class="docutils literal">sys.modules</tt>. The incomplete module object left
behind would fool further imports of the same module into succeeding, leading to
confusing errors. (Fixed by Tim Peters.)
</li>
<li><a class="reference internal" href="../library/constants.html#None" title="None"><tt class="xref py py-const docutils literal">None</tt></a> is now a constant; code that binds a new value to the name
<tt class="docutils literal">None</tt> is now a syntax error. (Contributed by Raymond Hettinger.)
</li>
</ul>
<div class="section" id="optimizations">
<h3>Optimizations<a class="headerlink" href="#optimizations" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li>The inner loops for list and tuple slicing were optimized and now run about
one-third faster. The inner loops for dictionaries were also optimized,
resulting in performance boosts for <tt class="xref py py-meth docutils literal">keys()</tt>, <tt class="xref py py-meth docutils literal">values()</tt>, <tt class="xref py py-meth docutils literal">items()</tt>,
<tt class="xref py py-meth docutils literal">iterkeys()</tt>, <tt class="xref py py-meth docutils literal">itervalues()</tt>, and <tt class="xref py py-meth docutils literal">iteritems()</tt>. (Contributed by
Raymond Hettinger.)</li>
<li>The machinery for growing and shrinking lists was optimized for speed and for
space efficiency. Appending and popping from lists now runs faster due to more
efficient code paths and less frequent use of the underlying system
<tt class="xref c c-func docutils literal">realloc()</tt>. List comprehensions also benefit. <tt class="xref py py-meth docutils literal">list.extend()</tt> was
also optimized and no longer converts its argument into a temporary list before
extending the base list. (Contributed by Raymond Hettinger.)</li>
<li><a class="reference internal" href="../library/functions.html#list" title="list"><tt class="xref py py-func docutils literal">list()</tt></a>, <a class="reference internal" href="../library/functions.html#tuple" title="tuple"><tt class="xref py py-func docutils literal">tuple()</tt></a>, <a class="reference internal" href="../library/functions.html#map" title="map"><tt class="xref py py-func docutils literal">map()</tt></a>, <a class="reference internal" href="../library/functions.html#filter" title="filter"><tt class="xref py py-func docutils literal">filter()</tt></a>, and <a class="reference internal" href="../library/functions.html#zip" title="zip"><tt class="xref py py-func docutils literal">zip()</tt></a> now
run several times faster with non-sequence arguments that supply a
<a class="reference internal" href="../reference/datamodel.html#object.__len__" title="object.__len__"><tt class="xref py py-meth docutils literal">__len__()</tt></a> method. (Contributed by Raymond Hettinger.)</li>
<li>The methods <tt class="xref py py-meth docutils literal">list.__getitem__()</tt>, <tt class="xref py py-meth docutils literal">dict.__getitem__()</tt>, and
<tt class="xref py py-meth docutils literal">dict.__contains__()</tt> are now implemented as <tt class="xref py py-class docutils literal">method_descriptor</tt>
objects rather than <tt class="xref py py-class docutils literal">wrapper_descriptor</tt> objects. This form of access
doubles their performance and makes them more suitable for use as arguments to
functionals: <tt class="docutils literal">map(mydict.__getitem__, keylist)</tt>. (Contributed by Raymond
Hettinger.)</li>
<li>Added a new opcode, <tt class="docutils literal">LIST_APPEND</tt>, that simplifies the generated bytecode
for list comprehensions and speeds them up by about a third. (Contributed by
Raymond Hettinger.)</li>
<li>The peephole bytecode optimizer has been improved to produce shorter, faster
bytecode; remarkably, the resulting bytecode is more readable. (Enhanced by
Raymond Hettinger.)</li>
<li>String concatenations in statements of the form <tt class="docutils literal">s = s + &quot;abc&quot;</tt> and <tt class="docutils literal">s +=
&quot;abc&quot;</tt> are now performed more efficiently in certain circumstances. This
optimization won&#8217;t be present in other Python implementations such as Jython, so
you shouldn&#8217;t rely on it; using the <tt class="xref py py-meth docutils literal">join()</tt> method of strings is still
recommended when you want to efficiently glue a large number of strings
together. (Contributed by Armin Rigo.)</li>
</ul>
The net result of the 2.4 optimizations is that Python 2.4 runs the pystone
benchmark around 5% faster than Python 2.3 and 35% faster than Python 2.2.
(pystone is not a particularly good benchmark, but it&#8217;s the most commonly used
measurement of Python&#8217;s performance. Your own applications may show greater or
smaller benefits from Python 2.4.)
</div>
</div>
<div class="section" id="new-improved-and-deprecated-modules">
<h2>New, Improved, and Deprecated Modules<a class="headerlink" href="#new-improved-and-deprecated-modules" title="Permalink to this headline">¶</a></h2>
As usual, Python&#8217;s standard library received a number of enhancements and bug
fixes. Here&#8217;s a partial list of the most notable changes, sorted alphabetically
by module name. Consult the <tt class="file docutils literal">Misc/NEWS</tt> file in the source tree for a more
complete list of changes, or look through the CVS logs for all the details.
<ul>
<li>The <a class="reference internal" href="../library/asyncore.html#module-asyncore" title="asyncore: A base class for developing asynchronous socket handling services."><tt class="xref py py-mod docutils literal">asyncore</tt></a> module&#8217;s <tt class="xref py py-func docutils literal">loop()</tt> function now has a count parameter
that lets you perform a limited number of passes through the polling loop. The
default is still to loop forever.
</li>
<li>The <a class="reference internal" href="../library/base64.html#module-base64" title="base64: RFC 3548: Base16, Base32, Base64 Data Encodings"><tt class="xref py py-mod docutils literal">base64</tt></a> module now has more complete RFC 3548 support for Base64,
Base32, and Base16 encoding and decoding, including optional case folding and
optional alternative alphabets. (Contributed by Barry Warsaw.)
</li>
<li>The <a class="reference internal" href="../library/bisect.html#module-bisect" title="bisect: Array bisection algorithms for binary searching."><tt class="xref py py-mod docutils literal">bisect</tt></a> module now has an underlying C implementation for improved
performance. (Contributed by Dmitry Vasiliev.)
</li>
<li>The CJKCodecs collections of East Asian codecs, maintained by Hye-Shik Chang,
was integrated into 2.4. The new encodings are:
</li>
<li>Chinese (PRC): gb2312, gbk, gb18030, big5hkscs, hz
</li>
<li>Chinese (ROC): big5, cp950
</li>
<li><dl class="first docutils">
<dt>Japanese: cp932, euc-jis-2004, euc-jp, euc-jisx0213, iso-2022-jp,</dt>
<dd>iso-2022-jp-1, iso-2022-jp-2, iso-2022-jp-3, iso-2022-jp-ext, iso-2022-jp-2004,
shift-jis, shift-jisx0213, shift-jis-2004
</dd>
</dl>
</li>
<li>Korean: cp949, euc-kr, johab, iso-2022-kr
</li>
<li>Some other new encodings were added: HP Roman8, ISO_8859-11, ISO_8859-16,
PCTP-154, and TIS-620.
</li>
<li>The UTF-8 and UTF-16 codecs now cope better with receiving partial input.
Previously the <tt class="xref py py-class docutils literal">StreamReader</tt> class would try to read more data, making
it impossible to resume decoding from the stream. The <tt class="xref py py-meth docutils literal">read()</tt> method will
now return as much data as it can and future calls will resume decoding where
previous ones left off. (Implemented by Walter Dörwald.)
</li>
<li>There is a new <a class="reference internal" href="../library/collections.html#module-collections" title="collections: High-performance datatypes"><tt class="xref py py-mod docutils literal">collections</tt></a> module for various specialized collection
datatypes. Currently it contains just one type, <tt class="xref py py-class docutils literal">deque</tt>, a double-
ended queue that supports efficiently adding and removing elements from either
end:
<div class="highlight-python"><div class="highlight"><pre>&gt;&gt;&gt; from collections import deque
&gt;&gt;&gt; d = deque(&#39;ghi&#39;) # make a new deque with three items
&gt;&gt;&gt; d.append(&#39;j&#39;) # add a new entry to the right side
&gt;&gt;&gt; d.appendleft(&#39;f&#39;) # add a new entry to the left side
&gt;&gt;&gt; d # show the representation of the deque
deque([&#39;f&#39;, &#39;g&#39;, &#39;h&#39;, &#39;i&#39;, &#39;j&#39;])
&gt;&gt;&gt; d.pop() # return and remove the rightmost item
&#39;j&#39;
&gt;&gt;&gt; d.popleft() # return and remove the leftmost item
&#39;f&#39;
&gt;&gt;&gt; list(d) # list the contents of the deque
[&#39;g&#39;, &#39;h&#39;, &#39;i&#39;]
&gt;&gt;&gt; &#39;h&#39; in d # search the deque
True
</pre></div>
</div>
Several modules, such as the <a class="reference internal" href="../library/queue.html#module-Queue" title="Queue: A synchronized queue class."><tt class="xref py py-mod docutils literal">Queue</tt></a> and <a class="reference internal" href="../library/threading.html#module-threading" title="threading: Higher-level threading interface."><tt class="xref py py-mod docutils literal">threading</tt></a> modules, now take
advantage of <a class="reference internal" href="../library/collections.html#collections.deque" title="collections.deque"><tt class="xref py py-class docutils literal">collections.deque</tt></a> for improved performance. (Contributed
by Raymond Hettinger.)
</li>
<li>The <a class="reference internal" href="../library/configparser.html#module-ConfigParser" title="ConfigParser: Configuration file parser."><tt class="xref py py-mod docutils literal">ConfigParser</tt></a> classes have been enhanced slightly. The <tt class="xref py py-meth docutils literal">read()</tt>
method now returns a list of the files that were successfully parsed, and the
<a class="reference internal" href="../library/stdtypes.html#set" title="set"><tt class="xref py py-meth docutils literal">set()</tt></a> method raises <a class="reference internal" href="../library/exceptions.html#exceptions.TypeError" title="exceptions.TypeError"><tt class="xref py py-exc docutils literal">TypeError</tt></a> if passed a value argument that
isn&#8217;t a string. (Contributed by John Belmonte and David Goodger.)
</li>
<li>The <a class="reference internal" href="../library/curses.html#module-curses" title="curses: An interface to the curses library, providing portable terminal handling. (Unix)"><tt class="xref py py-mod docutils literal">curses</tt></a> module now supports the ncurses extension
<tt class="xref py py-func docutils literal">use_default_colors()</tt>. On platforms where the terminal supports
transparency, this makes it possible to use a transparent background.
(Contributed by Jörg Lehmann.)
</li>
<li>The <a class="reference internal" href="../library/difflib.html#module-difflib" title="difflib: Helpers for computing differences between objects."><tt class="xref py py-mod docutils literal">difflib</tt></a> module now includes an <tt class="xref py py-class docutils literal">HtmlDiff</tt> class that creates
an HTML table showing a side by side comparison of two versions of a text.
(Contributed by Dan Gass.)
</li>
<li>The <a class="reference internal" href="../library/email.html#module-email" title="email: Package supporting the parsing, manipulating, and generating email messages, including MIME documents."><tt class="xref py py-mod docutils literal">email</tt></a> package was updated to version 3.0, which dropped various
deprecated APIs and removes support for Python versions earlier than 2.3. The
3.0 version of the package uses a new incremental parser for MIME messages,
available in the <tt class="xref py py-mod docutils literal">email.FeedParser</tt> module. The new parser doesn&#8217;t require
reading the entire message into memory, and doesn&#8217;t raise exceptions if a
message is malformed; instead it records any problems in the <tt class="xref py py-attr docutils literal">defect</tt>
attribute of the message. (Developed by Anthony Baxter, Barry Warsaw, Thomas
Wouters, and others.)
</li>
<li>The <a class="reference internal" href="../library/heapq.html#module-heapq" title="heapq: Heap queue algorithm (a.k.a. priority queue)."><tt class="xref py py-mod docutils literal">heapq</tt></a> module has been converted to C. The resulting tenfold
improvement in speed makes the module suitable for handling high volumes of
data. In addition, the module has two new functions <tt class="xref py py-func docutils literal">nlargest()</tt> and
<tt class="xref py py-func docutils literal">nsmallest()</tt> that use heaps to find the N largest or smallest values in a
dataset without the expense of a full sort. (Contributed by Raymond Hettinger.)
</li>
<li>The <a class="reference internal" href="../library/httplib.html#module-httplib" title="httplib: HTTP and HTTPS protocol client (requires sockets)."><tt class="xref py py-mod docutils literal">httplib</tt></a> module now contains constants for HTTP status codes defined
in various HTTP-related RFC documents. Constants have names such as
<tt class="xref py py-const docutils literal">OK</tt>, <tt class="xref py py-const docutils literal">CREATED</tt>, <tt class="xref py py-const docutils literal">CONTINUE</tt>, and
<tt class="xref py py-const docutils literal">MOVED_PERMANENTLY</tt>; use pydoc to get a full list. (Contributed by
Andrew Eland.)
</li>
<li>The <a class="reference internal" href="../library/imaplib.html#module-imaplib" title="imaplib: IMAP4 protocol client (requires sockets)."><tt class="xref py py-mod docutils literal">imaplib</tt></a> module now supports IMAP&#8217;s THREAD command (contributed by
Yves Dionne) and new <tt class="xref py py-meth docutils literal">deleteacl()</tt> and <tt class="xref py py-meth docutils literal">myrights()</tt> methods (contributed
by Arnaud Mazin).
</li>
<li>The <a class="reference internal" href="../library/itertools.html#module-itertools" title="itertools: Functions creating iterators for efficient looping."><tt class="xref py py-mod docutils literal">itertools</tt></a> module gained a <tt class="xref py py-func docutils literal">groupby(iterable[, *func*])()</tt>
function. iterable is something that can be iterated over to return a stream
of elements, and the optional func parameter is a function that takes an
element and returns a key value; if omitted, the key is simply the element
itself. <tt class="xref py py-func docutils literal">groupby()</tt> then groups the elements into subsequences which have
matching values of the key, and returns a series of 2-tuples containing the key
value and an iterator over the subsequence.
Here&#8217;s an example to make this clearer. The key function simply returns
whether a number is even or odd, so the result of <tt class="xref py py-func docutils literal">groupby()</tt> is to return
consecutive runs of odd or even numbers.
<div class="highlight-python"><div class="highlight"><pre>&gt;&gt;&gt; import itertools
&gt;&gt;&gt; L = [2, 4, 6, 7, 8, 9, 11, 12, 14]
&gt;&gt;&gt; for key_val, it in itertools.groupby(L, lambda x: x % 2):
... print key_val, list(it)
...
0 [2, 4, 6]
1 [7]
0 [8]
1 [9, 11]
0 [12, 14]
&gt;&gt;&gt;
</pre></div>
</div>
<tt class="xref py py-func docutils literal">groupby()</tt> is typically used with sorted input. The logic for
<tt class="xref py py-func docutils literal">groupby()</tt> is similar to the Unix <tt class="docutils literal">uniq</tt> filter which makes it handy for
eliminating, counting, or identifying duplicate elements:
<div class="highlight-python"><div class="highlight"><pre>&gt;&gt;&gt; word = &#39;abracadabra&#39;
&gt;&gt;&gt; letters = sorted(word) # Turn string into a sorted list of letters
&gt;&gt;&gt; letters
[&#39;a&#39;, &#39;a&#39;, &#39;a&#39;, &#39;a&#39;, &#39;a&#39;, &#39;b&#39;, &#39;b&#39;, &#39;c&#39;, &#39;d&#39;, &#39;r&#39;, &#39;r&#39;]
&gt;&gt;&gt; for k, g in itertools.groupby(letters):
... print k, list(g)
...
a [&#39;a&#39;, &#39;a&#39;, &#39;a&#39;, &#39;a&#39;, &#39;a&#39;]
b [&#39;b&#39;, &#39;b&#39;]
c [&#39;c&#39;]
d [&#39;d&#39;]
r [&#39;r&#39;, &#39;r&#39;]
&gt;&gt;&gt; # List unique letters
&gt;&gt;&gt; [k for k, g in groupby(letters)]
[&#39;a&#39;, &#39;b&#39;, &#39;c&#39;, &#39;d&#39;, &#39;r&#39;]
&gt;&gt;&gt; # Count letter occurrences
&gt;&gt;&gt; [(k, len(list(g))) for k, g in groupby(letters)]
[(&#39;a&#39;, 5), (&#39;b&#39;, 2), (&#39;c&#39;, 1), (&#39;d&#39;, 1), (&#39;r&#39;, 2)]
</pre></div>
</div>
(Contributed by Hye-Shik Chang.)
</li>
<li><a class="reference internal" href="../library/itertools.html#module-itertools" title="itertools: Functions creating iterators for efficient looping."><tt class="xref py py-mod docutils literal">itertools</tt></a> also gained a function named <tt class="xref py py-func docutils literal">tee(iterator, N)()</tt> that
returns N independent iterators that replicate iterator. If N is omitted,
the default is 2.
<div class="highlight-python"><div class="highlight"><pre>&gt;&gt;&gt; L = [1,2,3]
&gt;&gt;&gt; i1, i2 = itertools.tee(L)
&gt;&gt;&gt; i1,i2
(&lt;itertools.tee object at 0x402c2080&gt;, &lt;itertools.tee object at 0x402c2090&gt;)
&gt;&gt;&gt; list(i1) # Run the first iterator to exhaustion
[1, 2, 3]
&gt;&gt;&gt; list(i2) # Run the second iterator to exhaustion
[1, 2, 3]
</pre></div>
</div>
Note that <tt class="xref py py-func docutils literal">tee()</tt> has to keep copies of the values returned by the
iterator; in the worst case, it may need to keep all of them. This should
therefore be used carefully if the leading iterator can run far ahead of the
trailing iterator in a long stream of inputs. If the separation is large, then
you might as well use <a class="reference internal" href="../library/functions.html#list" title="list"><tt class="xref py py-func docutils literal">list()</tt></a> instead. When the iterators track closely
with one another, <tt class="xref py py-func docutils literal">tee()</tt> is ideal. Possible applications include
bookmarking, windowing, or lookahead iterators. (Contributed by Raymond
Hettinger.)
</li>
<li>A number of functions were added to the <a class="reference internal" href="../library/locale.html#module-locale" title="locale: Internationalization services."><tt class="xref py py-mod docutils literal">locale</tt></a> module, such as
<tt class="xref py py-func docutils literal">bind_textdomain_codeset()</tt> to specify a particular encoding and a family of
<tt class="xref py py-func docutils literal">l*gettext()</tt> functions that return messages in the chosen encoding.
(Contributed by Gustavo Niemeyer.)
</li>
<li>Some keyword arguments were added to the <a class="reference internal" href="../library/logging.html#module-logging" title="logging: Flexible event logging system for applications."><tt class="xref py py-mod docutils literal">logging</tt></a> package&#8217;s
<tt class="xref py py-func docutils literal">basicConfig()</tt> function to simplify log configuration. The default
behavior is to log messages to standard error, but various keyword arguments can
be specified to log to a particular file, change the logging format, or set the
logging level. For example:
<div class="highlight-python"><div class="highlight"><pre>import logging
logging.basicConfig(filename=&#39;/var/log/application.log&#39;,
 level=0, # Log all messages
 format=&#39;%(levelname):%(process):%(thread):%(message)&#39;)
</pre></div>
</div>
Other additions to the <a class="reference internal" href="../library/logging.html#module-logging" title="logging: Flexible event logging system for applications."><tt class="xref py py-mod docutils literal">logging</tt></a> package include a <tt class="xref py py-meth docutils literal">log(level, msg)()</tt>
convenience method, as well as a <tt class="xref py py-class docutils literal">TimedRotatingFileHandler</tt> class that
rotates its log files at a timed interval. The module already had
<tt class="xref py py-class docutils literal">RotatingFileHandler</tt>, which rotated logs once the file exceeded a
certain size. Both classes derive from a new <tt class="xref py py-class docutils literal">BaseRotatingHandler</tt> class
that can be used to implement other rotating handlers.
(Changes implemented by Vinay Sajip.)
</li>
<li>The <a class="reference internal" href="../library/marshal.html#module-marshal" title="marshal: Convert Python objects to streams of bytes and back (with different constraints)."><tt class="xref py py-mod docutils literal">marshal</tt></a> module now shares interned strings on unpacking a data
structure. This may shrink the size of certain pickle strings, but the primary
effect is to make <tt class="file docutils literal">.pyc</tt> files significantly smaller. (Contributed by
Martin von Löwis.)
</li>
<li>The <a class="reference internal" href="../library/nntplib.html#module-nntplib" title="nntplib: NNTP protocol client (requires sockets)."><tt class="xref py py-mod docutils literal">nntplib</tt></a> module&#8217;s <tt class="xref py py-class docutils literal">NNTP</tt> class gained <tt class="xref py py-meth docutils literal">description()</tt> and
<tt class="xref py py-meth docutils literal">descriptions()</tt> methods to retrieve newsgroup descriptions for a single
group or for a range of groups. (Contributed by Jürgen A. Erhard.)
</li>
<li>Two new functions were added to the <a class="reference internal" href="../library/operator.html#module-operator" title="operator: Functions corresponding to the standard operators."><tt class="xref py py-mod docutils literal">operator</tt></a> module,
<tt class="xref py py-func docutils literal">attrgetter(attr)()</tt> and <tt class="xref py py-func docutils literal">itemgetter(index)()</tt>. Both functions return
callables that take a single argument and return the corresponding attribute or
item; these callables make excellent data extractors when used with <a class="reference internal" href="../library/functions.html#map" title="map"><tt class="xref py py-func docutils literal">map()</tt></a>
or <a class="reference internal" href="../library/functions.html#sorted" title="sorted"><tt class="xref py py-func docutils literal">sorted()</tt></a>. For example:
<div class="highlight-python"><div class="highlight"><pre>&gt;&gt;&gt; L = [(&#39;c&#39;, 2), (&#39;d&#39;, 1), (&#39;a&#39;, 4), (&#39;b&#39;, 3)]
&gt;&gt;&gt; map(operator.itemgetter(0), L)
[&#39;c&#39;, &#39;d&#39;, &#39;a&#39;, &#39;b&#39;]
&gt;&gt;&gt; map(operator.itemgetter(1), L)
[2, 1, 4, 3]
&gt;&gt;&gt; sorted(L, key=operator.itemgetter(1)) # Sort list by second tuple item
[(&#39;d&#39;, 1), (&#39;c&#39;, 2), (&#39;b&#39;, 3), (&#39;a&#39;, 4)]
</pre></div>
</div>
(Contributed by Raymond Hettinger.)
</li>
<li>The <a class="reference internal" href="../library/optparse.html#module-optparse" title="optparse: Command-line option parsing library. (deprecated)"><tt class="xref py py-mod docutils literal">optparse</tt></a> module was updated in various ways. The module now passes
its messages through <a class="reference internal" href="../library/gettext.html#gettext.gettext" title="gettext.gettext"><tt class="xref py py-func docutils literal">gettext.gettext()</tt></a>, making it possible to
internationalize Optik&#8217;s help and error messages. Help messages for options can
now include the string <tt class="docutils literal">'%default'</tt>, which will be replaced by the option&#8217;s
default value. (Contributed by Greg Ward.)
</li>
<li>The long-term plan is to deprecate the <a class="reference internal" href="../library/rfc822.html#module-rfc822" title="rfc822: Parse 2822 style mail messages. (deprecated)"><tt class="xref py py-mod docutils literal">rfc822</tt></a> module in some future
Python release in favor of the <a class="reference internal" href="../library/email.html#module-email" title="email: Package supporting the parsing, manipulating, and generating email messages, including MIME documents."><tt class="xref py py-mod docutils literal">email</tt></a> package. To this end, the
<tt class="xref py py-func docutils literal">email.Utils.formatdate()</tt> function has been changed to make it usable as a
replacement for <tt class="xref py py-func docutils literal">rfc822.formatdate()</tt>. You may want to write new e-mail
processing code with this in mind. (Change implemented by Anthony Baxter.)
</li>
<li>A new <tt class="xref py py-func docutils literal">urandom(n)()</tt> function was added to the <a class="reference internal" href="../library/os.html#module-os" title="os: Miscellaneous operating system interfaces."><tt class="xref py py-mod docutils literal">os</tt></a> module, returning
a string containing n bytes of random data. This function provides access to
platform-specific sources of randomness such as <tt class="file docutils literal">/dev/urandom</tt> on Linux or
the Windows CryptoAPI. (Contributed by Trevor Perrin.)
</li>
<li>Another new function: <tt class="xref py py-func docutils literal">os.path.lexists(path)()</tt> returns true if the file
specified by path exists, whether or not it&#8217;s a symbolic link. This differs
from the existing <tt class="xref py py-func docutils literal">os.path.exists(path)()</tt> function, which returns false if
path is a symlink that points to a destination that doesn&#8217;t exist.
(Contributed by Beni Cherniavsky.)
</li>
<li>A new <tt class="xref py py-func docutils literal">getsid()</tt> function was added to the <a class="reference internal" href="../library/posix.html#module-posix" title="posix: The most common POSIX system calls (normally used via module os). (Unix)"><tt class="xref py py-mod docutils literal">posix</tt></a> module that
underlies the <a class="reference internal" href="../library/os.html#module-os" title="os: Miscellaneous operating system interfaces."><tt class="xref py py-mod docutils literal">os</tt></a> module. (Contributed by J. Raynor.)
</li>
<li>The <a class="reference internal" href="../library/poplib.html#module-poplib" title="poplib: POP3 protocol client (requires sockets)."><tt class="xref py py-mod docutils literal">poplib</tt></a> module now supports POP over SSL. (Contributed by Hector
Urtubia.)
</li>
<li>The <a class="reference internal" href="../library/profile.html#module-profile" title="profile: Python source profiler."><tt class="xref py py-mod docutils literal">profile</tt></a> module can now profile C extension functions. (Contributed
by Nick Bastin.)
</li>
<li>The <a class="reference internal" href="../library/random.html#module-random" title="random: Generate pseudo-random numbers with various common distributions."><tt class="xref py py-mod docutils literal">random</tt></a> module has a new method called <tt class="xref py py-meth docutils literal">getrandbits(N)()</tt> that
returns a long integer N bits in length. The existing <tt class="xref py py-meth docutils literal">randrange()</tt>
method now uses <tt class="xref py py-meth docutils literal">getrandbits()</tt> where appropriate, making generation of
arbitrarily large random numbers more efficient. (Contributed by Raymond
Hettinger.)
</li>
<li>The regular expression language accepted by the <a class="reference internal" href="../library/re.html#module-re" title="re: Regular expression operations."><tt class="xref py py-mod docutils literal">re</tt></a> module was extended
with simple conditional expressions, written as <tt class="docutils literal">(?(group)A|B)</tt>. group is
either a numeric group ID or a group name defined with <tt class="docutils literal">(?P&lt;group&gt;...)</tt>
earlier in the expression. If the specified group matched, the regular
expression pattern A will be tested against the string; if the group didn&#8217;t
match, the pattern B will be used instead. (Contributed by Gustavo Niemeyer.)
</li>
<li>The <a class="reference internal" href="../library/re.html#module-re" title="re: Regular expression operations."><tt class="xref py py-mod docutils literal">re</tt></a> module is also no longer recursive, thanks to a massive amount
of work by Gustavo Niemeyer. In a recursive regular expression engine, certain
patterns result in a large amount of C stack space being consumed, and it was
possible to overflow the stack. For example, if you matched a 30000-byte string
of <tt class="docutils literal">a</tt> characters against the expression <tt class="docutils literal">(a|b)+</tt>, one stack frame was
consumed per character. Python 2.3 tried to check for stack overflow and raise
a <a class="reference internal" href="../library/exceptions.html#exceptions.RuntimeError" title="exceptions.RuntimeError"><tt class="xref py py-exc docutils literal">RuntimeError</tt></a> exception, but certain patterns could sidestep the
checking and if you were unlucky Python could segfault. Python 2.4&#8217;s regular
expression engine can match this pattern without problems.
</li>
<li>The <a class="reference internal" href="../library/signal.html#module-signal" title="signal: Set handlers for asynchronous events."><tt class="xref py py-mod docutils literal">signal</tt></a> module now performs tighter error-checking on the parameters
to the <a class="reference internal" href="../library/signal.html#signal.signal" title="signal.signal"><tt class="xref py py-func docutils literal">signal.signal()</tt></a> function. For example, you can&#8217;t set a handler on
the <tt class="xref py py-const docutils literal">SIGKILL</tt> signal; previous versions of Python would quietly accept
this, but 2.4 will raise a <a class="reference internal" href="../library/exceptions.html#exceptions.RuntimeError" title="exceptions.RuntimeError"><tt class="xref py py-exc docutils literal">RuntimeError</tt></a> exception.
</li>
<li>Two new functions were added to the <a class="reference internal" href="../library/socket.html#module-socket" title="socket: Low-level networking interface."><tt class="xref py py-mod docutils literal">socket</tt></a> module. <tt class="xref py py-func docutils literal">socketpair()</tt>
returns a pair of connected sockets and <tt class="xref py py-func docutils literal">getservbyport(port)()</tt> looks up the
service name for a given port number. (Contributed by Dave Cole and Barry
Warsaw.)
</li>
<li>The <a class="reference internal" href="../library/sys.html#sys.exitfunc" title="sys.exitfunc"><tt class="xref py py-func docutils literal">sys.exitfunc()</tt></a> function has been deprecated. Code should be using
the existing <a class="reference internal" href="../library/atexit.html#module-atexit" title="atexit: Register and execute cleanup functions."><tt class="xref py py-mod docutils literal">atexit</tt></a> module, which correctly handles calling multiple exit
functions. Eventually <a class="reference internal" href="../library/sys.html#sys.exitfunc" title="sys.exitfunc"><tt class="xref py py-func docutils literal">sys.exitfunc()</tt></a> will become a purely internal
interface, accessed only by <a class="reference internal" href="../library/atexit.html#module-atexit" title="atexit: Register and execute cleanup functions."><tt class="xref py py-mod docutils literal">atexit</tt></a>.
</li>
<li>The <a class="reference internal" href="../library/tarfile.html#module-tarfile" title="tarfile: Read and write tar-format archive files."><tt class="xref py py-mod docutils literal">tarfile</tt></a> module now generates GNU-format tar files by default.
(Contributed by Lars Gustaebel.)
</li>
<li>The <a class="reference internal" href="../library/threading.html#module-threading" title="threading: Higher-level threading interface."><tt class="xref py py-mod docutils literal">threading</tt></a> module now has an elegantly simple way to support
thread-local data. The module contains a <tt class="xref py py-class docutils literal">local</tt> class whose attribute
values are local to different threads.
<div class="highlight-python"><div class="highlight"><pre>import threading

data = threading.local()
data.number = 42
data.url = (&#39;www.python.org&#39;, 80)
</pre></div>
</div>
Other threads can assign and retrieve their own values for the <tt class="xref py py-attr docutils literal">number</tt>
and <tt class="xref py py-attr docutils literal">url</tt> attributes. You can subclass <tt class="xref py py-class docutils literal">local</tt> to initialize
attributes or to add methods. (Contributed by Jim Fulton.)
</li>
<li>The <a class="reference internal" href="../library/timeit.html#module-timeit" title="timeit: Measure the execution time of small code snippets."><tt class="xref py py-mod docutils literal">timeit</tt></a> module now automatically disables periodic garbage
collection during the timing loop. This change makes consecutive timings more
comparable. (Contributed by Raymond Hettinger.)
</li>
<li>The <a class="reference internal" href="../library/weakref.html#module-weakref" title="weakref: Support for weak references and weak dictionaries."><tt class="xref py py-mod docutils literal">weakref</tt></a> module now supports a wider variety of objects including
Python functions, class instances, sets, frozensets, deques, arrays, files,
sockets, and regular expression pattern objects. (Contributed by Raymond
Hettinger.)
</li>
<li>The <a class="reference internal" href="../library/xmlrpclib.html#module-xmlrpclib" title="xmlrpclib: XML-RPC client access."><tt class="xref py py-mod docutils literal">xmlrpclib</tt></a> module now supports a multi-call extension for
transmitting multiple XML-RPC calls in a single HTTP operation. (Contributed by
Brian Quinlan.)
</li>
<li>The <tt class="xref py py-mod docutils literal">mpz</tt>, <tt class="xref py py-mod docutils literal">rotor</tt>, and <tt class="xref py py-mod docutils literal">xreadlines</tt> modules have been
removed.
</li>
</ul>
<div class="section" id="cookielib">
<h3>cookielib<a class="headerlink" href="#cookielib" title="Permalink to this headline">¶</a></h3>
The <a class="reference internal" href="../library/cookielib.html#module-cookielib" title="cookielib: Classes for automatic handling of HTTP cookies."><tt class="xref py py-mod docutils literal">cookielib</tt></a> library supports client-side handling for HTTP cookies,
mirroring the <a class="reference internal" href="../library/cookie.html#module-Cookie" title="Cookie: Support for HTTP state management (cookies)."><tt class="xref py py-mod docutils literal">Cookie</tt></a> module&#8217;s server-side cookie support. Cookies are
stored in cookie jars; the library transparently stores cookies offered by the
web server in the cookie jar, and fetches the cookie from the jar when
connecting to the server. As in web browsers, policy objects control whether
cookies are accepted or not.
In order to store cookies across sessions, two implementations of cookie jars
are provided: one that stores cookies in the Netscape format so applications can
use the Mozilla or Lynx cookie files, and one that stores cookies in the same
format as the Perl libwww library.
<a class="reference internal" href="../library/urllib2.html#module-urllib2" title="urllib2: Next generation URL opening library."><tt class="xref py py-mod docutils literal">urllib2</tt></a> has been changed to interact with <a class="reference internal" href="../library/cookielib.html#module-cookielib" title="cookielib: Classes for automatic handling of HTTP cookies."><tt class="xref py py-mod docutils literal">cookielib</tt></a>:
<tt class="xref py py-class docutils literal">HTTPCookieProcessor</tt> manages a cookie jar that is used when accessing
URLs.
This module was contributed by John J. Lee.
</div>
<div class="section" id="doctest">
<h3>doctest<a class="headerlink" href="#doctest" title="Permalink to this headline">¶</a></h3>
The <a class="reference internal" href="../library/doctest.html#module-doctest" title="doctest: Test pieces of code within docstrings."><tt class="xref py py-mod docutils literal">doctest</tt></a> module underwent considerable refactoring thanks to Edward
Loper and Tim Peters. Testing can still be as simple as running
<a class="reference internal" href="../library/doctest.html#doctest.testmod" title="doctest.testmod"><tt class="xref py py-func docutils literal">doctest.testmod()</tt></a>, but the refactorings allow customizing the module&#8217;s
operation in various ways
The new <tt class="xref py py-class docutils literal">DocTestFinder</tt> class extracts the tests from a given object&#8217;s
docstrings:
<div class="highlight-python"><div class="highlight"><pre>def f (x, y):
 &quot;&quot;&quot;&gt;&gt;&gt; f(2,2)
4
&gt;&gt;&gt; f(3,2)
6
 &quot;&quot;&quot;
 return x*y

finder = doctest.DocTestFinder()

# Get list of DocTest instances
tests = finder.find(f)
</pre></div>
</div>
The new <tt class="xref py py-class docutils literal">DocTestRunner</tt> class then runs individual tests and can produce
a summary of the results:
<div class="highlight-python"><div class="highlight"><pre>runner = doctest.DocTestRunner()
for t in tests:
 tried, failed = runner.run(t)

runner.summarize(verbose=1)
</pre></div>
</div>
The above example produces the following output:
<div class="highlight-python"><pre>1 items passed all tests:
 2 tests in f
2 tests in 1 items.
2 passed and 0 failed.
Test passed.</pre>
</div>
<tt class="xref py py-class docutils literal">DocTestRunner</tt> uses an instance of the <tt class="xref py py-class docutils literal">OutputChecker</tt> class to
compare the expected output with the actual output. This class takes a number
of different flags that customize its behaviour; ambitious users can also write
a completely new subclass of <tt class="xref py py-class docutils literal">OutputChecker</tt>.
The default output checker provides a number of handy features. For example,
with the <a class="reference internal" href="../library/doctest.html#doctest.ELLIPSIS" title="doctest.ELLIPSIS"><tt class="xref py py-const docutils literal">doctest.ELLIPSIS</tt></a> option flag, an ellipsis (<tt class="docutils literal">...</tt>) in the
expected output matches any substring, making it easier to accommodate outputs
that vary in minor ways:
<div class="highlight-python"><div class="highlight"><pre>def o (n):
 &quot;&quot;&quot;&gt;&gt;&gt; o(1)
&lt;__main__.C instance at 0x...&gt;
&gt;&gt;&gt;
&quot;&quot;&quot;
</pre></div>
</div>
Another special string, <tt class="docutils literal">&lt;BLANKLINE&gt;</tt>, matches a blank line:
<div class="highlight-python"><div class="highlight"><pre>def p (n):
 &quot;&quot;&quot;&gt;&gt;&gt; p(1)
&lt;BLANKLINE&gt;
&gt;&gt;&gt;
&quot;&quot;&quot;
</pre></div>
</div>
Another new capability is producing a diff-style display of the output by
specifying the <a class="reference internal" href="../library/doctest.html#doctest.REPORT_UDIFF" title="doctest.REPORT_UDIFF"><tt class="xref py py-const docutils literal">doctest.REPORT_UDIFF</tt></a> (unified diffs),
<a class="reference internal" href="../library/doctest.html#doctest.REPORT_CDIFF" title="doctest.REPORT_CDIFF"><tt class="xref py py-const docutils literal">doctest.REPORT_CDIFF</tt></a> (context diffs), or <a class="reference internal" href="../library/doctest.html#doctest.REPORT_NDIFF" title="doctest.REPORT_NDIFF"><tt class="xref py py-const docutils literal">doctest.REPORT_NDIFF</tt></a>
(delta-style) option flags. For example:
<div class="highlight-python"><div class="highlight"><pre>def g (n):
 &quot;&quot;&quot;&gt;&gt;&gt; g(4)
here
is
a
lengthy
&gt;&gt;&gt;&quot;&quot;&quot;
 L = &#39;here is a rather lengthy list of words&#39;.split()
 for word in L[:n]:
 print word
</pre></div>
</div>
Running the above function&#8217;s tests with <a class="reference internal" href="../library/doctest.html#doctest.REPORT_UDIFF" title="doctest.REPORT_UDIFF"><tt class="xref py py-const docutils literal">doctest.REPORT_UDIFF</tt></a> specified,
you get the following output:
<div class="highlight-python"><pre>**********************************************************************
File "t.py", line 15, in g
Failed example:
 g(4)
Differences (unified diff with -expected +actual):
 @@ -2,3 +2,3 @@
 is
 a
 -lengthy
 +rather
**********************************************************************</pre>
</div>
</div>
</div>
<div class="section" id="build-and-c-api-changes">
<h2>Build and C API Changes<a class="headerlink" href="#build-and-c-api-changes" title="Permalink to this headline">¶</a></h2>
Some of the changes to Python&#8217;s build process and to the C API are:
<ul class="simple">
<li>Three new convenience macros were added for common return values from
extension functions: <a class="reference internal" href="../c-api/none.html#Py_RETURN_NONE" title="Py_RETURN_NONE"><tt class="xref c c-macro docutils literal">Py_RETURN_NONE</tt></a>, <a class="reference internal" href="../c-api/bool.html#Py_RETURN_TRUE" title="Py_RETURN_TRUE"><tt class="xref c c-macro docutils literal">Py_RETURN_TRUE</tt></a>, and
<a class="reference internal" href="../c-api/bool.html#Py_RETURN_FALSE" title="Py_RETURN_FALSE"><tt class="xref c c-macro docutils literal">Py_RETURN_FALSE</tt></a>. (Contributed by Brett Cannon.)</li>
<li>Another new macro, <tt class="xref c c-macro docutils literal">Py_CLEAR(obj)</tt>, decreases the reference count of
obj and sets obj to the null pointer. (Contributed by Jim Fulton.)</li>
<li>A new function, <tt class="xref c c-func docutils literal">PyTuple_Pack(N, obj1, obj2, ..., objN)()</tt>, constructs
tuples from a variable length argument list of Python objects. (Contributed by
Raymond Hettinger.)</li>
<li>A new function, <tt class="xref c c-func docutils literal">PyDict_Contains(d, k)()</tt>, implements fast dictionary
lookups without masking exceptions raised during the look-up process.
(Contributed by Raymond Hettinger.)</li>
<li>The <tt class="xref c c-macro docutils literal">Py_IS_NAN(X)</tt> macro returns 1 if its float or double argument
X is a NaN. (Contributed by Tim Peters.)</li>
<li>C code can avoid unnecessary locking by using the new
<a class="reference internal" href="../c-api/init.html#PyEval_ThreadsInitialized" title="PyEval_ThreadsInitialized"><tt class="xref c c-func docutils literal">PyEval_ThreadsInitialized()</tt></a> function to tell if any thread operations
have been performed. If this function returns false, no lock operations are
needed. (Contributed by Nick Coghlan.)</li>
<li>A new function, <a class="reference internal" href="../c-api/arg.html#PyArg_VaParseTupleAndKeywords" title="PyArg_VaParseTupleAndKeywords"><tt class="xref c c-func docutils literal">PyArg_VaParseTupleAndKeywords()</tt></a>, is the same as
<a class="reference internal" href="../c-api/arg.html#PyArg_ParseTupleAndKeywords" title="PyArg_ParseTupleAndKeywords"><tt class="xref c c-func docutils literal">PyArg_ParseTupleAndKeywords()</tt></a> but takes a <tt class="xref c c-type docutils literal">va_list</tt> instead of a
number of arguments. (Contributed by Greg Chapman.)</li>
<li>A new method flag, <tt class="xref py py-const docutils literal">METH_COEXISTS</tt>, allows a function defined in slots
to co-exist with a <a class="reference internal" href="../c-api/structures.html#PyCFunction" title="PyCFunction"><tt class="xref c c-type docutils literal">PyCFunction</tt></a> having the same name. This can halve
the access time for a method such as <tt class="xref py py-meth docutils literal">set.__contains__()</tt>. (Contributed by
Raymond Hettinger.)</li>
<li>Python can now be built with additional profiling for the interpreter itself,
intended as an aid to people developing the Python core. Providing
----enable-profiling to the configure script will let you
profile the interpreter with gprof, and providing the
----with-tsc switch enables profiling using the Pentium&#8217;s Time-Stamp-
Counter register. Note that the ----with-tsc switch is slightly
misnamed, because the profiling feature also works on the PowerPC platform,
though that processor architecture doesn&#8217;t call that register &#8220;the TSC
register&#8221;. (Contributed by Jeremy Hylton.)</li>
<li>The <tt class="xref c c-type docutils literal">tracebackobject</tt> type has been renamed to
<tt class="xref c c-type docutils literal">PyTracebackObject</tt>.</li>
</ul>
<div class="section" id="port-specific-changes">
<h3>Port-Specific Changes<a class="headerlink" href="#port-specific-changes" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li>The Windows port now builds under MSVC++ 7.1 as well as version 6.
(Contributed by Martin von Löwis.)</li>
</ul>
</div>
</div>
<div class="section" id="porting-to-python-2-4">
<h2>Porting to Python 2.4<a class="headerlink" href="#porting-to-python-2-4" title="Permalink to this headline">¶</a></h2>
This section lists previously described changes that may require changes to your
code:
<ul class="simple">
<li>Left shifts and hexadecimal/octal constants that are too large no longer
trigger a <a class="reference internal" href="../library/exceptions.html#exceptions.FutureWarning" title="exceptions.FutureWarning"><tt class="xref py py-exc docutils literal">FutureWarning</tt></a> and return a value limited to 32 or 64 bits;
instead they return a long integer.</li>
<li>Integer operations will no longer trigger an <tt class="xref py py-exc docutils literal">OverflowWarning</tt>. The
<tt class="xref py py-exc docutils literal">OverflowWarning</tt> warning will disappear in Python 2.5.</li>
<li>The <a class="reference internal" href="../library/functions.html#zip" title="zip"><tt class="xref py py-func docutils literal">zip()</tt></a> built-in function and <a class="reference internal" href="../library/itertools.html#itertools.izip" title="itertools.izip"><tt class="xref py py-func docutils literal">itertools.izip()</tt></a> now return an
empty list instead of raising a <a class="reference internal" href="../library/exceptions.html#exceptions.TypeError" title="exceptions.TypeError"><tt class="xref py py-exc docutils literal">TypeError</tt></a> exception if called with no
arguments.</li>
<li>You can no longer compare the <tt class="xref py py-class docutils literal">date</tt> and <a class="reference internal" href="../library/datetime.html#module-datetime" title="datetime: Basic date and time types."><tt class="xref py py-class docutils literal">datetime</tt></a> instances
provided by the <a class="reference internal" href="../library/datetime.html#module-datetime" title="datetime: Basic date and time types."><tt class="xref py py-mod docutils literal">datetime</tt></a> module. Two instances of different classes
will now always be unequal, and relative comparisons (<tt class="docutils literal">&lt;</tt>, <tt class="docutils literal">&gt;</tt>) will raise
a <a class="reference internal" href="../library/exceptions.html#exceptions.TypeError" title="exceptions.TypeError"><tt class="xref py py-exc docutils literal">TypeError</tt></a>.</li>
<li><a class="reference internal" href="../library/dircache.html#dircache.listdir" title="dircache.listdir"><tt class="xref py py-func docutils literal">dircache.listdir()</tt></a> now passes exceptions to the caller instead of
returning empty lists.</li>
<li><tt class="xref py py-func docutils literal">LexicalHandler.startDTD()</tt> used to receive the public and system IDs in
the wrong order. This has been corrected; applications relying on the wrong
order need to be fixed.</li>
<li><a class="reference internal" href="../library/fcntl.html#fcntl.ioctl" title="fcntl.ioctl"><tt class="xref py py-func docutils literal">fcntl.ioctl()</tt></a> now warns if the mutate argument is omitted and
relevant.</li>
<li>The <a class="reference internal" href="../library/tarfile.html#module-tarfile" title="tarfile: Read and write tar-format archive files."><tt class="xref py py-mod docutils literal">tarfile</tt></a> module now generates GNU-format tar files by default.</li>
<li>Encountering a failure while importing a module no longer leaves a partially-
initialized module object in <tt class="docutils literal">sys.modules</tt>.</li>
<li><a class="reference internal" href="../library/constants.html#None" title="None"><tt class="xref py py-const docutils literal">None</tt></a> is now a constant; code that binds a new value to the name
<tt class="docutils literal">None</tt> is now a syntax error.</li>
<li>The <tt class="xref py py-func docutils literal">signals.signal()</tt> function now raises a <a class="reference internal" href="../library/exceptions.html#exceptions.RuntimeError" title="exceptions.RuntimeError"><tt class="xref py py-exc docutils literal">RuntimeError</tt></a> exception
for certain illegal values; previously these errors would pass silently. For
example, you can no longer set a handler on the <tt class="xref py py-const docutils literal">SIGKILL</tt> signal.</li>
</ul>
</div>
<div class="section" id="acknowledgements">
<h2>Acknowledgements<a class="headerlink" href="#acknowledgements" title="Permalink to this headline">¶</a></h2>
The author would like to thank the following people for offering suggestions,
corrections and assistance with various drafts of this article: Koray Can,
Hye-Shik Chang, Michael Dyck, Raymond Hettinger, Brian Hurt, Hamish Lawson,
Fredrik Lundh, Sean Reifschneider, Sadruddin Rejeb.
</div>
</div>

</div>
 </div>
 </div>
 <div class="sphinxsidebar">
 <div class="sphinxsidebarwrapper">
 <h3><a href="../contents.html">Table Of Contents</a></h3>
 <ul>
<li><a class="reference internal" href="#">What&#8217;s New in Python 2.4</a><ul>
<li><a class="reference internal" href="#pep-218-built-in-set-objects">PEP 218: Built-In Set Objects</a></li>
<li><a class="reference internal" href="#pep-237-unifying-long-integers-and-integers">PEP 237: Unifying Long Integers and Integers</a></li>
<li><a class="reference internal" href="#pep-289-generator-expressions">PEP 289: Generator Expressions</a></li>
<li><a class="reference internal" href="#pep-292-simpler-string-substitutions">PEP 292: Simpler String Substitutions</a></li>
<li><a class="reference internal" href="#pep-318-decorators-for-functions-and-methods">PEP 318: Decorators for Functions and Methods</a></li>
<li><a class="reference internal" href="#pep-322-reverse-iteration">PEP 322: Reverse Iteration</a></li>
<li><a class="reference internal" href="#pep-324-new-subprocess-module">PEP 324: New subprocess Module</a></li>
<li><a class="reference internal" href="#pep-327-decimal-data-type">PEP 327: Decimal Data Type</a><ul>
<li><a class="reference internal" href="#why-is-decimal-needed">Why is Decimal needed?</a></li>
<li><a class="reference internal" href="#the-decimal-type">The <tt class="docutils literal">Decimal</tt> type</a></li>
<li><a class="reference internal" href="#the-context-type">The <tt class="docutils literal">Context</tt> type</a></li>
</ul>
</li>
<li><a class="reference internal" href="#pep-328-multi-line-imports">PEP 328: Multi-line Imports</a></li>
<li><a class="reference internal" href="#pep-331-locale-independent-float-string-conversions">PEP 331: Locale-Independent Float/String Conversions</a></li>
<li><a class="reference internal" href="#other-language-changes">Other Language Changes</a><ul>
<li><a class="reference internal" href="#optimizations">Optimizations</a></li>
</ul>
</li>
<li><a class="reference internal" href="#new-improved-and-deprecated-modules">New, Improved, and Deprecated Modules</a><ul>
<li><a class="reference internal" href="#cookielib">cookielib</a></li>
<li><a class="reference internal" href="#doctest">doctest</a></li>
</ul>
</li>
<li><a class="reference internal" href="#build-and-c-api-changes">Build and C API Changes</a><ul>
<li><a class="reference internal" href="#port-specific-changes">Port-Specific Changes</a></li>
</ul>
</li>
<li><a class="reference internal" href="#porting-to-python-2-4">Porting to Python 2.4</a></li>
<li><a class="reference internal" href="#acknowledgements">Acknowledgements</a></li>
</ul>
</li>
</ul>

<h4>Previous topic</h4>
 <a href="2.5.html"
 title="previous chapter">What&#8217;s New in Python 2.5</a>
 <h4>Next topic</h4>
 <a href="2.3.html"
 title="next chapter">What&#8217;s New in Python 2.3</a>
<h3>This Page</h3>
<ul class="this-page-menu">
 <li><a href="../bugs.html">Report a Bug</a></li>
 <li><a href="../_sources/whatsnew/2.4.txt"
 rel="nofollow">Show Source</a></li>
</ul>

<div id="searchbox" style="display: none">
 <h3>Quick search</h3>
 <form class="search" action="../search.html" method="get">
 <input 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>
 
 Enter search terms or a module, class or function name.
 
</div>
<script type="text/javascript">$('#searchbox').show(0);</script>
 </div>
 </div>
 <div class="clearer"></div>
 </div>
 <div class="related">
 <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="2.3.html" title="What’s New in Python 2.3"
 >next</a> |</li>
 <li class="right" >
 <a href="2.5.html" title="What’s New in Python 2.5"
 >previous</a> |</li>
 <li><img src="../_static/py.png" alt=""
 style="vertical-align: middle; margin-top: -1px"/></li>
 <li><a href="http://www.python.org/">Python</a> &raquo;</li>
 <li>
 <a href="../index.html">Python 2.7.5 documentation</a> &raquo;
 </li>

<li><a href="index.html" >What&#8217;s New in Python</a> &raquo;</li> 
 </ul>
 </div>
 <div class="footer">
 &copy; <a href="../copyright.html">Copyright</a> 1990-2020, Python Software Foundation.
 
 The Python Software Foundation is a non-profit corporation.
 <a href="http://www.python.org/psf/donations/">Please donate.</a>
 
 Last updated on Oct 13, 2020.
 <a href="../bugs.html">Found a bug</a>?
 
 Created using <a href="http://sphinx.pocoo.org/">Sphinx</a> 1.1.3.
 </div>

</body>
</html>