22

✏️ Editing: 3.2.html

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

<html xmlns="http://www.w3.org/1999/xhtml">
 <head>
 <meta http-equiv="X-UA-Compatible" content="IE=Edge" />
 <meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
 <title>What’s New In Python 3.2 &#8212; Python 3.6.7 documentation</title>
 <link rel="stylesheet" href="../_static/pydoctheme.css" type="text/css" />
 <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
 <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
 <script type="text/javascript" src="../_static/jquery.js"></script>
 <script type="text/javascript" src="../_static/underscore.js"></script>
 <script type="text/javascript" src="../_static/doctools.js"></script>
 <script type="text/javascript" src="../_static/sidebar.js"></script>
 <link rel="search" type="application/opensearchdescription+xml"
 title="Search within Python 3.6.7 documentation"
 href="../_static/opensearch.xml"/>
 <link rel="author" title="About these documents" href="../about.html" />
 <link rel="index" title="Index" href="../genindex.html" />
 <link rel="search" title="Search" href="../search.html" />
 <link rel="copyright" title="Copyright" href="../copyright.html" />
 <link rel="next" title="What’s New In Python 3.1" href="3.1.html" />
 <link rel="prev" title="What’s New In Python 3.3" href="3.3.html" />
 <link rel="shortcut icon" type="image/png" href="../_static/py.png" />
 <link rel="canonical" href="https://docs.python.org/3/whatsnew/3.2.html" />
 
 <script type="text/javascript" src="../_static/copybutton.js"></script>

</head><body> 
 <div class="related" role="navigation" aria-label="related navigation">
 <h3>Navigation</h3>
 <ul>
 <li class="right" style="margin-right: 10px">
 <a href="../genindex.html" title="General Index"
 accesskey="I">index</a></li>
 <li class="right" >
 <a href="../py-modindex.html" title="Python Module Index"
 >modules</a> |</li>
 <li class="right" >
 <a href="3.1.html" title="What’s New In Python 3.1"
 accesskey="N">next</a> |</li>
 <li class="right" >
 <a href="3.3.html" title="What’s New In Python 3.3"
 accesskey="P">previous</a> |</li>
 <li><img src="../_static/py.png" alt=""
 style="vertical-align: middle; margin-top: -1px"/></li>
 <li><a href="https://www.python.org/">Python</a> &#187;</li>
 <li>
 <a href="../index.html">3.6.7 Documentation</a> &#187;
 </li>

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

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

</ul>
 </div>

<div class="document">
 <div class="documentwrapper">
 <div class="bodywrapper">
 <div class="body" role="main">
 
 <div class="section" id="what-s-new-in-python-3-2">
<h1>What’s New In Python 3.2<a class="headerlink" href="#what-s-new-in-python-3-2" title="Permalink to this headline">¶</a></h1>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">Author:</th><td class="field-body">Raymond Hettinger</td>
</tr>
</tbody>
</table>
This article explains the new features in Python 3.2 as compared to 3.1. It
focuses on a few highlights and gives a few examples. For full details, see the
<a class="reference external" href="https://hg.python.org/cpython/file/3.2/Misc/NEWS">Misc/NEWS</a> file.
<div class="admonition seealso">
See also
<a class="pep reference external" href="https://www.python.org/dev/peps/pep-0392">PEP 392</a> - Python 3.2 Release Schedule
</div>
<div class="section" id="pep-384-defining-a-stable-abi">
<h2>PEP 384: Defining a Stable ABI<a class="headerlink" href="#pep-384-defining-a-stable-abi" title="Permalink to this headline">¶</a></h2>
In the past, extension modules built for one Python version were often
not usable with other Python versions. Particularly on Windows, every
feature release of Python required rebuilding all extension modules that
one wanted to use. This requirement was the result of the free access to
Python interpreter internals that extension modules could use.
With Python 3.2, an alternative approach becomes available: extension
modules which restrict themselves to a limited API (by defining
Py_LIMITED_API) cannot use many of the internals, but are constrained
to a set of API functions that are promised to be stable for several
releases. As a consequence, extension modules built for 3.2 in that
mode will also work with 3.3, 3.4, and so on. Extension modules that
make use of details of memory structures can still be built, but will
need to be recompiled for every feature release.
<div class="admonition seealso">
See also
<dl class="last docutils">
<dt><a class="pep reference external" href="https://www.python.org/dev/peps/pep-0384">PEP 384</a> - Defining a Stable ABI</dt>
<dd>PEP written by Martin von Löwis.</dd>
</dl>
</div>
</div>
<div class="section" id="pep-389-argparse-command-line-parsing-module">
<h2>PEP 389: Argparse Command Line Parsing Module<a class="headerlink" href="#pep-389-argparse-command-line-parsing-module" title="Permalink to this headline">¶</a></h2>
A new module for command line parsing, <a class="reference internal" href="../library/argparse.html#module-argparse" title="argparse: Command-line option and argument parsing library."><code class="xref py py-mod docutils literal notranslate">argparse</code></a>, was introduced to
overcome the limitations of <a class="reference internal" href="../library/optparse.html#module-optparse" title="optparse: Command-line option parsing library. (deprecated)"><code class="xref py py-mod docutils literal notranslate">optparse</code></a> which did not provide support for
positional arguments (not just options), subcommands, required options and other
common patterns of specifying and validating options.
This module has already had widespread success in the community as a
third-party module. Being more fully featured than its predecessor, the
<a class="reference internal" href="../library/argparse.html#module-argparse" title="argparse: Command-line option and argument parsing library."><code class="xref py py-mod docutils literal notranslate">argparse</code></a> module is now the preferred module for command-line processing.
The older module is still being kept available because of the substantial amount
of legacy code that depends on it.
Here’s an annotated example parser showing features like limiting results to a
set of choices, specifying a metavar in the help screen, validating that one
or more positional arguments is present, and making a required option:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>import argparse
parser = argparse.ArgumentParser(
 description = &#39;Manage servers&#39;, # main description for help
 epilog = &#39;Tested on Solaris and Linux&#39;) # displayed after help
parser.add_argument(&#39;action&#39;, # argument name
 choices = [&#39;deploy&#39;, &#39;start&#39;, &#39;stop&#39;], # three allowed values
 help = &#39;action on each target&#39;) # help msg
parser.add_argument(&#39;targets&#39;,
 metavar = &#39;HOSTNAME&#39;, # var name used in help msg
 nargs = &#39;+&#39;, # require one or more targets
 help = &#39;url for target machines&#39;) # help msg explanation
parser.add_argument(&#39;-u&#39;, &#39;--user&#39;, # -u or --user option
 required = True, # make it a required argument
 help = &#39;login as user&#39;)
</pre></div>
</div>
Example of calling the parser on a command string:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; cmd = &#39;deploy sneezy.example.com sleepy.example.com -u skycaptain&#39;
&gt;&gt;&gt; result = parser.parse_args(cmd.split())
&gt;&gt;&gt; result.action
&#39;deploy&#39;
&gt;&gt;&gt; result.targets
[&#39;sneezy.example.com&#39;, &#39;sleepy.example.com&#39;]
&gt;&gt;&gt; result.user
&#39;skycaptain&#39;
</pre></div>
</div>
Example of the parser’s automatically generated help:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; parser.parse_args(&#39;-h&#39;.split())

usage: manage_cloud.py [-h] -u USER
 {deploy,start,stop} HOSTNAME [HOSTNAME ...]

Manage servers

positional arguments:
 {deploy,start,stop} action on each target
 HOSTNAME url for target machines

optional arguments:
 -h, --help show this help message and exit
 -u USER, --user USER login as user

Tested on Solaris and Linux
</pre></div>
</div>
An especially nice <a class="reference internal" href="../library/argparse.html#module-argparse" title="argparse: Command-line option and argument parsing library."><code class="xref py py-mod docutils literal notranslate">argparse</code></a> feature is the ability to define subparsers,
each with their own argument patterns and help displays:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>import argparse
parser = argparse.ArgumentParser(prog=&#39;HELM&#39;)
subparsers = parser.add_subparsers()

parser_l = subparsers.add_parser(&#39;launch&#39;, help=&#39;Launch Control&#39;) # first subgroup
parser_l.add_argument(&#39;-m&#39;, &#39;--missiles&#39;, action=&#39;store_true&#39;)
parser_l.add_argument(&#39;-t&#39;, &#39;--torpedos&#39;, action=&#39;store_true&#39;)

parser_m = subparsers.add_parser(&#39;move&#39;, help=&#39;Move Vessel&#39;, # second subgroup
 aliases=(&#39;steer&#39;, &#39;turn&#39;)) # equivalent names
parser_m.add_argument(&#39;-c&#39;, &#39;--course&#39;, type=int, required=True)
parser_m.add_argument(&#39;-s&#39;, &#39;--speed&#39;, type=int, default=0)
</pre></div>
</div>
<div class="highlight-shell-session notranslate"><div class="highlight"><pre>$ ./helm.py --help # top level help (launch and move)
$ ./helm.py launch --help # help for launch options
$ ./helm.py launch --missiles # set missiles=True and torpedos=False
$ ./helm.py steer --course 180 --speed 5 # set movement parameters
</pre></div>
</div>
<div class="admonition seealso">
See also
<dl class="docutils">
<dt><a class="pep reference external" href="https://www.python.org/dev/peps/pep-0389">PEP 389</a> - New Command Line Parsing Module</dt>
<dd>PEP written by Steven Bethard.</dd>
</dl>
<a class="reference internal" href="../library/argparse.html#upgrading-optparse-code">Upgrading optparse code</a> for details on the differences from <a class="reference internal" href="../library/optparse.html#module-optparse" title="optparse: Command-line option parsing library. (deprecated)"><code class="xref py py-mod docutils literal notranslate">optparse</code></a>.
</div>
</div>
<div class="section" id="pep-391-dictionary-based-configuration-for-logging">
<h2>PEP 391: Dictionary Based Configuration for Logging<a class="headerlink" href="#pep-391-dictionary-based-configuration-for-logging" title="Permalink to this headline">¶</a></h2>
The <a class="reference internal" href="../library/logging.html#module-logging" title="logging: Flexible event logging system for applications."><code class="xref py py-mod docutils literal notranslate">logging</code></a> module provided two kinds of configuration, one style with
function calls for each option or another style driven by an external file saved
in a <code class="xref py py-mod docutils literal notranslate">ConfigParser</code> format. Those options did not provide the flexibility
to create configurations from JSON or YAML files, nor did they support
incremental configuration, which is needed for specifying logger options from a
command line.
To support a more flexible style, the module now offers
<a class="reference internal" href="../library/logging.config.html#logging.config.dictConfig" title="logging.config.dictConfig"><code class="xref py py-func docutils literal notranslate">logging.config.dictConfig()</code></a> for specifying logging configuration with
plain Python dictionaries. The configuration options include formatters,
handlers, filters, and loggers. Here’s a working example of a configuration
dictionary:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>{&quot;version&quot;: 1,
 &quot;formatters&quot;: {&quot;brief&quot;: {&quot;format&quot;: &quot;%(levelname)-8s: %(name)-15s: %(message)s&quot;},
 &quot;full&quot;: {&quot;format&quot;: &quot;%(asctime)s %(name)-15s %(levelname)-8s %(message)s&quot;}
 },
 &quot;handlers&quot;: {&quot;console&quot;: {
 &quot;class&quot;: &quot;logging.StreamHandler&quot;,
 &quot;formatter&quot;: &quot;brief&quot;,
 &quot;level&quot;: &quot;INFO&quot;,
 &quot;stream&quot;: &quot;ext://sys.stdout&quot;},
 &quot;console_priority&quot;: {
 &quot;class&quot;: &quot;logging.StreamHandler&quot;,
 &quot;formatter&quot;: &quot;full&quot;,
 &quot;level&quot;: &quot;ERROR&quot;,
 &quot;stream&quot;: &quot;ext://sys.stderr&quot;}
 },
 &quot;root&quot;: {&quot;level&quot;: &quot;DEBUG&quot;, &quot;handlers&quot;: [&quot;console&quot;, &quot;console_priority&quot;]}}
</pre></div>
</div>
If that dictionary is stored in a file called <code class="file docutils literal notranslate">conf.json</code>, it can be
loaded and called with code like this:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; import json, logging.config
&gt;&gt;&gt; with open(&#39;conf.json&#39;) as f:
... conf = json.load(f)
...
&gt;&gt;&gt; logging.config.dictConfig(conf)
&gt;&gt;&gt; logging.info(&quot;Transaction completed normally&quot;)
INFO : root : Transaction completed normally
&gt;&gt;&gt; logging.critical(&quot;Abnormal termination&quot;)
2011-02-17 11:14:36,694 root CRITICAL Abnormal termination
</pre></div>
</div>
<div class="admonition seealso">
See also
<dl class="last docutils">
<dt><a class="pep reference external" href="https://www.python.org/dev/peps/pep-0391">PEP 391</a> - Dictionary Based Configuration for Logging</dt>
<dd>PEP written by Vinay Sajip.</dd>
</dl>
</div>
</div>
<div class="section" id="pep-3148-the-concurrent-futures-module">
<h2>PEP 3148: The <code class="docutils literal notranslate">concurrent.futures</code> module<a class="headerlink" href="#pep-3148-the-concurrent-futures-module" title="Permalink to this headline">¶</a></h2>
Code for creating and managing concurrency is being collected in a new top-level
namespace, concurrent. Its first member is a futures package which provides
a uniform high-level interface for managing threads and processes.
The design for <a class="reference internal" href="../library/concurrent.futures.html#module-concurrent.futures" title="concurrent.futures: Execute computations concurrently using threads or processes."><code class="xref py py-mod docutils literal notranslate">concurrent.futures</code></a> was inspired by the
java.util.concurrent package. In that model, a running call and its result
are represented by a <a class="reference internal" href="../library/concurrent.futures.html#concurrent.futures.Future" title="concurrent.futures.Future"><code class="xref py py-class docutils literal notranslate">Future</code></a> object that abstracts
features common to threads, processes, and remote procedure calls. That object
supports status checks (running or done), timeouts, cancellations, adding
callbacks, and access to results or exceptions.
The primary offering of the new module is a pair of executor classes for
launching and managing calls. The goal of the executors is to make it easier to
use existing tools for making parallel calls. They save the effort needed to
setup a pool of resources, launch the calls, create a results queue, add
time-out handling, and limit the total number of threads, processes, or remote
procedure calls.
Ideally, each application should share a single executor across multiple
components so that process and thread limits can be centrally managed. This
solves the design challenge that arises when each component has its own
competing strategy for resource management.
Both classes share a common interface with three methods:
<a class="reference internal" href="../library/concurrent.futures.html#concurrent.futures.Executor.submit" title="concurrent.futures.Executor.submit"><code class="xref py py-meth docutils literal notranslate">submit()</code></a> for scheduling a callable and
returning a <a class="reference internal" href="../library/concurrent.futures.html#concurrent.futures.Future" title="concurrent.futures.Future"><code class="xref py py-class docutils literal notranslate">Future</code></a> object;
<a class="reference internal" href="../library/concurrent.futures.html#concurrent.futures.Executor.map" title="concurrent.futures.Executor.map"><code class="xref py py-meth docutils literal notranslate">map()</code></a> for scheduling many asynchronous calls
at a time, and <a class="reference internal" href="../library/concurrent.futures.html#concurrent.futures.Executor.shutdown" title="concurrent.futures.Executor.shutdown"><code class="xref py py-meth docutils literal notranslate">shutdown()</code></a> for freeing
resources. The class is a <a class="reference internal" href="../glossary.html#term-context-manager">context manager</a> and can be used in a
<a class="reference internal" href="../reference/compound_stmts.html#with"><code class="xref std std-keyword docutils literal notranslate">with</code></a> statement to assure that resources are automatically released
when currently pending futures are done executing.
A simple of example of <a class="reference internal" href="../library/concurrent.futures.html#concurrent.futures.ThreadPoolExecutor" title="concurrent.futures.ThreadPoolExecutor"><code class="xref py py-class docutils literal notranslate">ThreadPoolExecutor</code></a> is a
launch of four parallel threads for copying files:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>import concurrent.futures, shutil
with concurrent.futures.ThreadPoolExecutor(max_workers=4) as e:
 e.submit(shutil.copy, &#39;src1.txt&#39;, &#39;dest1.txt&#39;)
 e.submit(shutil.copy, &#39;src2.txt&#39;, &#39;dest2.txt&#39;)
 e.submit(shutil.copy, &#39;src3.txt&#39;, &#39;dest3.txt&#39;)
 e.submit(shutil.copy, &#39;src3.txt&#39;, &#39;dest4.txt&#39;)
</pre></div>
</div>
<div class="admonition seealso">
See also
<dl class="docutils">
<dt><a class="pep reference external" href="https://www.python.org/dev/peps/pep-3148">PEP 3148</a> - Futures – Execute Computations Asynchronously</dt>
<dd>PEP written by Brian Quinlan.</dd>
</dl>
<a class="reference internal" href="../library/concurrent.futures.html#threadpoolexecutor-example">Code for Threaded Parallel URL reads</a>, an
example using threads to fetch multiple web pages in parallel.
<a class="reference internal" href="../library/concurrent.futures.html#processpoolexecutor-example">Code for computing prime numbers in
parallel</a>, an example demonstrating
<a class="reference internal" href="../library/concurrent.futures.html#concurrent.futures.ProcessPoolExecutor" title="concurrent.futures.ProcessPoolExecutor"><code class="xref py py-class docutils literal notranslate">ProcessPoolExecutor</code></a>.
</div>
</div>
<div class="section" id="pep-3147-pyc-repository-directories">
<h2>PEP 3147: PYC Repository Directories<a class="headerlink" href="#pep-3147-pyc-repository-directories" title="Permalink to this headline">¶</a></h2>
Python’s scheme for caching bytecode in .pyc files did not work well in
environments with multiple Python interpreters. If one interpreter encountered
a cached file created by another interpreter, it would recompile the source and
overwrite the cached file, thus losing the benefits of caching.
The issue of “pyc fights” has become more pronounced as it has become
commonplace for Linux distributions to ship with multiple versions of Python.
These conflicts also arise with CPython alternatives such as Unladen Swallow.
To solve this problem, Python’s import machinery has been extended to use
distinct filenames for each interpreter. Instead of Python 3.2 and Python 3.3 and
Unladen Swallow each competing for a file called “mymodule.pyc”, they will now
look for “mymodule.cpython-32.pyc”, “mymodule.cpython-33.pyc”, and
“mymodule.unladen10.pyc”. And to prevent all of these new files from
cluttering source directories, the pyc files are now collected in a
“__pycache__” directory stored under the package directory.
Aside from the filenames and target directories, the new scheme has a few
aspects that are visible to the programmer:
<ul>
<li>Imported modules now have a <a class="reference internal" href="../reference/import.html#__cached__" title="__cached__"><code class="xref py py-attr docutils literal notranslate">__cached__</code></a> attribute which stores the name
of the actual file that was imported:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; import collections
&gt;&gt;&gt; collections.__cached__ 
&#39;c:/py32/lib/__pycache__/collections.cpython-32.pyc&#39;
</pre></div>
</div>
</li>
<li>The tag that is unique to each interpreter is accessible from the <a class="reference internal" href="../library/imp.html#module-imp" title="imp: Access the implementation of the import statement. (deprecated)"><code class="xref py py-mod docutils literal notranslate">imp</code></a>
module:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; import imp
&gt;&gt;&gt; imp.get_tag() 
&#39;cpython-32&#39;
</pre></div>
</div>
</li>
<li>Scripts that try to deduce source filename from the imported file now need to
be smarter. It is no longer sufficient to simply strip the “c” from a “.pyc”
filename. Instead, use the new functions in the <a class="reference internal" href="../library/imp.html#module-imp" title="imp: Access the implementation of the import statement. (deprecated)"><code class="xref py py-mod docutils literal notranslate">imp</code></a> module:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; imp.source_from_cache(&#39;c:/py32/lib/__pycache__/collections.cpython-32.pyc&#39;)
&#39;c:/py32/lib/collections.py&#39;
&gt;&gt;&gt; imp.cache_from_source(&#39;c:/py32/lib/collections.py&#39;) 
&#39;c:/py32/lib/__pycache__/collections.cpython-32.pyc&#39;
</pre></div>
</div>
</li>
<li>The <a class="reference internal" href="../library/py_compile.html#module-py_compile" title="py_compile: Generate byte-code files from Python source files."><code class="xref py py-mod docutils literal notranslate">py_compile</code></a> and <a class="reference internal" href="../library/compileall.html#module-compileall" title="compileall: Tools for byte-compiling all Python source files in a directory tree."><code class="xref py py-mod docutils literal notranslate">compileall</code></a> modules have been updated to
reflect the new naming convention and target directory. The command-line
invocation of compileall has new options: <code class="docutils literal notranslate">-i</code> for
specifying a list of files and directories to compile and <code class="docutils literal notranslate">-b</code> which causes
bytecode files to be written to their legacy location rather than
__pycache__.
</li>
<li>The <a class="reference internal" href="../library/importlib.html#module-importlib.abc" title="importlib.abc: Abstract base classes related to import"><code class="xref py py-mod docutils literal notranslate">importlib.abc</code></a> module has been updated with new <a class="reference internal" href="../glossary.html#term-abstract-base-class">abstract base
classes</a> for loading bytecode files. The obsolete
ABCs, <code class="xref py py-class docutils literal notranslate">PyLoader</code> and
<code class="xref py py-class docutils literal notranslate">PyPycLoader</code>, have been deprecated (instructions on how
to stay Python 3.1 compatible are included with the documentation).
</li>
</ul>
<div class="admonition seealso">
See also
<dl class="last docutils">
<dt><a class="pep reference external" href="https://www.python.org/dev/peps/pep-3147">PEP 3147</a> - PYC Repository Directories</dt>
<dd>PEP written by Barry Warsaw.</dd>
</dl>
</div>
</div>
<div class="section" id="pep-3149-abi-version-tagged-so-files">
<h2>PEP 3149: ABI Version Tagged .so Files<a class="headerlink" href="#pep-3149-abi-version-tagged-so-files" title="Permalink to this headline">¶</a></h2>
The PYC repository directory allows multiple bytecode cache files to be
co-located. This PEP implements a similar mechanism for shared object files by
giving them a common directory and distinct names for each version.
The common directory is “pyshared” and the file names are made distinct by
identifying the Python implementation (such as CPython, PyPy, Jython, etc.), the
major and minor version numbers, and optional build flags (such as “d” for
debug, “m” for pymalloc, “u” for wide-unicode). For an arbitrary package “foo”,
you may see these files when the distribution package is installed:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>/usr/share/pyshared/foo.cpython-32m.so
/usr/share/pyshared/foo.cpython-33md.so
</pre></div>
</div>
In Python itself, the tags are accessible from functions in the <a class="reference internal" href="../library/sysconfig.html#module-sysconfig" title="sysconfig: Python's configuration information"><code class="xref py py-mod docutils literal notranslate">sysconfig</code></a>
module:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; import sysconfig
&gt;&gt;&gt; sysconfig.get_config_var(&#39;SOABI&#39;) # find the version tag
&#39;cpython-32mu&#39;
&gt;&gt;&gt; sysconfig.get_config_var(&#39;EXT_SUFFIX&#39;) # find the full filename extension
&#39;.cpython-32mu.so&#39;
</pre></div>
</div>
<div class="admonition seealso">
See also
<dl class="last docutils">
<dt><a class="pep reference external" href="https://www.python.org/dev/peps/pep-3149">PEP 3149</a> - ABI Version Tagged .so Files</dt>
<dd>PEP written by Barry Warsaw.</dd>
</dl>
</div>
</div>
<div class="section" id="pep-3333-python-web-server-gateway-interface-v1-0-1">
<h2>PEP 3333: Python Web Server Gateway Interface v1.0.1<a class="headerlink" href="#pep-3333-python-web-server-gateway-interface-v1-0-1" title="Permalink to this headline">¶</a></h2>
This informational PEP clarifies how bytes/text issues are to be handled by the
WSGI protocol. The challenge is that string handling in Python 3 is most
conveniently handled with the <a class="reference internal" href="../library/stdtypes.html#str" title="str"><code class="xref py py-class docutils literal notranslate">str</code></a> type even though the HTTP protocol
is itself bytes oriented.
The PEP differentiates so-called native strings that are used for
request/response headers and metadata versus byte strings which are used for
the bodies of requests and responses.
The native strings are always of type <a class="reference internal" href="../library/stdtypes.html#str" title="str"><code class="xref py py-class docutils literal notranslate">str</code></a> but are restricted to code
points between U+0000 through U+00FF which are translatable to bytes using
Latin-1 encoding. These strings are used for the keys and values in the
environment dictionary and for response headers and statuses in the
<code class="xref py py-func docutils literal notranslate">start_response()</code> function. They must follow <a class="rfc reference external" href="https://tools.ietf.org/html/rfc2616.html">RFC 2616</a> with respect to
encoding. That is, they must either be ISO-8859-1 characters or use
<a class="rfc reference external" href="https://tools.ietf.org/html/rfc2047.html">RFC 2047</a> MIME encoding.
For developers porting WSGI applications from Python 2, here are the salient
points:
<ul class="simple">
<li>If the app already used strings for headers in Python 2, no change is needed.</li>
<li>If instead, the app encoded output headers or decoded input headers, then the
headers will need to be re-encoded to Latin-1. For example, an output header
encoded in utf-8 was using <code class="docutils literal notranslate">h.encode('utf-8')</code> now needs to convert from
bytes to native strings using <code class="docutils literal notranslate">h.encode('utf-8').decode('latin-1')</code>.</li>
<li>Values yielded by an application or sent using the <code class="xref py py-meth docutils literal notranslate">write()</code> method
must be byte strings. The <code class="xref py py-func docutils literal notranslate">start_response()</code> function and environ
must use native strings. The two cannot be mixed.</li>
</ul>
For server implementers writing CGI-to-WSGI pathways or other CGI-style
protocols, the users must to be able access the environment using native strings
even though the underlying platform may have a different convention. To bridge
this gap, the <a class="reference internal" href="../library/wsgiref.html#module-wsgiref" title="wsgiref: WSGI Utilities and Reference Implementation."><code class="xref py py-mod docutils literal notranslate">wsgiref</code></a> module has a new function,
<a class="reference internal" href="../library/wsgiref.html#wsgiref.handlers.read_environ" title="wsgiref.handlers.read_environ"><code class="xref py py-func docutils literal notranslate">wsgiref.handlers.read_environ()</code></a> for transcoding CGI variables from
<a class="reference internal" href="../library/os.html#os.environ" title="os.environ"><code class="xref py py-attr docutils literal notranslate">os.environ</code></a> into native strings and returning a new dictionary.
<div class="admonition seealso">
See also
<dl class="last docutils">
<dt><a class="pep reference external" href="https://www.python.org/dev/peps/pep-3333">PEP 3333</a> - Python Web Server Gateway Interface v1.0.1</dt>
<dd>PEP written by Phillip Eby.</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>
Some smaller changes made to the core Python language are:
<ul>
<li>String formatting for <a class="reference internal" href="../library/functions.html#format" title="format"><code class="xref py py-func docutils literal notranslate">format()</code></a> and <a class="reference internal" href="../library/stdtypes.html#str.format" title="str.format"><code class="xref py py-meth docutils literal notranslate">str.format()</code></a> gained new
capabilities for the format character #. Previously, for integers in
binary, octal, or hexadecimal, it caused the output to be prefixed with ‘0b’,
‘0o’, or ‘0x’ respectively. Now it can also handle floats, complex, and
Decimal, causing the output to always have a decimal point even when no digits
follow it.
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; format(20, &#39;#o&#39;)
&#39;0o24&#39;
&gt;&gt;&gt; format(12.34, &#39;#5.0f&#39;)
&#39; 12.&#39;
</pre></div>
</div>
(Suggested by Mark Dickinson and implemented by Eric Smith in <a class="reference external" href="https://bugs.python.org/issue7094">bpo-7094</a>.)
</li>
<li>There is also a new <a class="reference internal" href="../library/stdtypes.html#str.format_map" title="str.format_map"><code class="xref py py-meth docutils literal notranslate">str.format_map()</code></a> method that extends the
capabilities of the existing <a class="reference internal" href="../library/stdtypes.html#str.format" title="str.format"><code class="xref py py-meth docutils literal notranslate">str.format()</code></a> method by accepting arbitrary
<a class="reference internal" href="../glossary.html#term-mapping">mapping</a> objects. This new method makes it possible to use string
formatting with any of Python’s many dictionary-like objects such as
<a class="reference internal" href="../library/collections.html#collections.defaultdict" title="collections.defaultdict"><code class="xref py py-class docutils literal notranslate">defaultdict</code></a>, <a class="reference internal" href="../library/shelve.html#shelve.Shelf" title="shelve.Shelf"><code class="xref py py-class docutils literal notranslate">Shelf</code></a>,
<a class="reference internal" href="../library/configparser.html#configparser.ConfigParser" title="configparser.ConfigParser"><code class="xref py py-class docutils literal notranslate">ConfigParser</code></a>, or <a class="reference internal" href="../library/dbm.html#module-dbm" title="dbm: Interfaces to various Unix &quot;database&quot; formats."><code class="xref py py-mod docutils literal notranslate">dbm</code></a>. It is also useful with
custom <a class="reference internal" href="../library/stdtypes.html#dict" title="dict"><code class="xref py py-class docutils literal notranslate">dict</code></a> subclasses that normalize keys before look-up or that
supply a <a class="reference internal" href="../reference/datamodel.html#object.__missing__" title="object.__missing__"><code class="xref py py-meth docutils literal notranslate">__missing__()</code></a> method for unknown keys:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; import shelve
&gt;&gt;&gt; d = shelve.open(&#39;tmp.shl&#39;)
&gt;&gt;&gt; &#39;The {project_name} status is {status} as of {date}&#39;.format_map(d)
&#39;The testing project status is green as of February 15, 2011&#39;

&gt;&gt;&gt; class LowerCasedDict(dict):
... def __getitem__(self, key):
... return dict.__getitem__(self, key.lower())
&gt;&gt;&gt; lcd = LowerCasedDict(part=&#39;widgets&#39;, quantity=10)
&gt;&gt;&gt; &#39;There are {QUANTITY} {Part} in stock&#39;.format_map(lcd)
&#39;There are 10 widgets in stock&#39;

&gt;&gt;&gt; class PlaceholderDict(dict):
... def __missing__(self, key):
... return &#39;&lt;{}&gt;&#39;.format(key)
&gt;&gt;&gt; &#39;Hello {name}, welcome to {location}&#39;.format_map(PlaceholderDict())
&#39;Hello &lt;name&gt;, welcome to &lt;location&gt;&#39;
</pre></div>
</div>
</li>
</ul>
<blockquote>
<div>(Suggested by Raymond Hettinger and implemented by Eric Smith in
<a class="reference external" href="https://bugs.python.org/issue6081">bpo-6081</a>.)</div></blockquote>
<ul>
<li>The interpreter can now be started with a quiet option, <code class="docutils literal notranslate">-q</code>, to prevent
the copyright and version information from being displayed in the interactive
mode. The option can be introspected using the <a class="reference internal" href="../library/sys.html#sys.flags" title="sys.flags"><code class="xref py py-attr docutils literal notranslate">sys.flags</code></a> attribute:
<div class="highlight-shell-session notranslate"><div class="highlight"><pre>$ python -q
&gt;&gt;&gt; sys.flags
sys.flags(debug=0, division_warning=0, inspect=0, interactive=0,
optimize=0, dont_write_bytecode=0, no_user_site=0, no_site=0,
ignore_environment=0, verbose=0, bytes_warning=0, quiet=1)
</pre></div>
</div>
(Contributed by Marcin Wojdyr in <a class="reference external" href="https://bugs.python.org/issue1772833">bpo-1772833</a>).
</li>
<li>The <a class="reference internal" href="../library/functions.html#hasattr" title="hasattr"><code class="xref py py-func docutils literal notranslate">hasattr()</code></a> function works by calling <a class="reference internal" href="../library/functions.html#getattr" title="getattr"><code class="xref py py-func docutils literal notranslate">getattr()</code></a> and detecting
whether an exception is raised. This technique allows it to detect methods
created dynamically by <a class="reference internal" href="../reference/datamodel.html#object.__getattr__" title="object.__getattr__"><code class="xref py py-meth docutils literal notranslate">__getattr__()</code></a> or <a class="reference internal" href="../reference/datamodel.html#object.__getattribute__" title="object.__getattribute__"><code class="xref py py-meth docutils literal notranslate">__getattribute__()</code></a> which
would otherwise be absent from the class dictionary. Formerly, hasattr
would catch any exception, possibly masking genuine errors. Now, hasattr
has been tightened to only catch <a class="reference internal" href="../library/exceptions.html#AttributeError" title="AttributeError"><code class="xref py py-exc docutils literal notranslate">AttributeError</code></a> and let other
exceptions pass through:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; class A:
... @property
... def f(self):
... return 1 // 0
...
&gt;&gt;&gt; a = A()
&gt;&gt;&gt; hasattr(a, &#39;f&#39;)
Traceback (most recent call last):
 ...
ZeroDivisionError: integer division or modulo by zero
</pre></div>
</div>
(Discovered by Yury Selivanov and fixed by Benjamin Peterson; <a class="reference external" href="https://bugs.python.org/issue9666">bpo-9666</a>.)
</li>
<li>The <a class="reference internal" href="../library/stdtypes.html#str" title="str"><code class="xref py py-func docutils literal notranslate">str()</code></a> of a float or complex number is now the same as its
<a class="reference internal" href="../library/functions.html#repr" title="repr"><code class="xref py py-func docutils literal notranslate">repr()</code></a>. Previously, the <a class="reference internal" href="../library/stdtypes.html#str" title="str"><code class="xref py py-func docutils literal notranslate">str()</code></a> form was shorter but that just
caused confusion and is no longer needed now that the shortest possible
<a class="reference internal" href="../library/functions.html#repr" title="repr"><code class="xref py py-func docutils literal notranslate">repr()</code></a> is displayed by default:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; import math
&gt;&gt;&gt; repr(math.pi)
&#39;3.141592653589793&#39;
&gt;&gt;&gt; str(math.pi)
&#39;3.141592653589793&#39;
</pre></div>
</div>
(Proposed and implemented by Mark Dickinson; <a class="reference external" href="https://bugs.python.org/issue9337">bpo-9337</a>.)
</li>
<li><a class="reference internal" href="../library/stdtypes.html#memoryview" title="memoryview"><code class="xref py py-class docutils literal notranslate">memoryview</code></a> objects now have a <a class="reference internal" href="../library/stdtypes.html#memoryview.release" title="memoryview.release"><code class="xref py py-meth docutils literal notranslate">release()</code></a> method
and they also now support the context management protocol. This allows timely
release of any resources that were acquired when requesting a buffer from the
original object.
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; with memoryview(b&#39;abcdefgh&#39;) as v:
... print(v.tolist())
[97, 98, 99, 100, 101, 102, 103, 104]
</pre></div>
</div>
(Added by Antoine Pitrou; <a class="reference external" href="https://bugs.python.org/issue9757">bpo-9757</a>.)
</li>
<li>Previously it was illegal to delete a name from the local namespace if it
occurs as a free variable in a nested block:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>def outer(x):
 def inner():
 return x
 inner()
 del x
</pre></div>
</div>
This is now allowed. Remember that the target of an <a class="reference internal" href="../reference/compound_stmts.html#except"><code class="xref std std-keyword docutils literal notranslate">except</code></a> clause
is cleared, so this code which used to work with Python 2.6, raised a
<a class="reference internal" href="../library/exceptions.html#SyntaxError" title="SyntaxError"><code class="xref py py-exc docutils literal notranslate">SyntaxError</code></a> with Python 3.1 and now works again:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>def f():
 def print_error():
 print(e)
 try:
 something
 except Exception as e:
 print_error()
 # implicit &quot;del e&quot; here
</pre></div>
</div>
(See <a class="reference external" href="https://bugs.python.org/issue4617">bpo-4617</a>.)
</li>
<li>The internal <code class="xref c c-type docutils literal notranslate">structsequence</code> tool now creates subclasses of tuple.
This means that C structures like those returned by <a class="reference internal" href="../library/os.html#os.stat" title="os.stat"><code class="xref py py-func docutils literal notranslate">os.stat()</code></a>,
<a class="reference internal" href="../library/time.html#time.gmtime" title="time.gmtime"><code class="xref py py-func docutils literal notranslate">time.gmtime()</code></a>, and <a class="reference internal" href="../library/sys.html#sys.version_info" title="sys.version_info"><code class="xref py py-attr docutils literal notranslate">sys.version_info</code></a> now work like a
<a class="reference internal" href="../glossary.html#term-named-tuple">named tuple</a> and now work with functions and methods that
expect a tuple as an argument. This is a big step forward in making the C
structures as flexible as their pure Python counterparts:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; import sys
&gt;&gt;&gt; isinstance(sys.version_info, tuple)
True
&gt;&gt;&gt; &#39;Version %d.%d.%d %s(%d)&#39; % sys.version_info 
&#39;Version 3.2.0 final(0)&#39;
</pre></div>
</div>
(Suggested by Arfrever Frehtes Taifersar Arahesis and implemented
by Benjamin Peterson in <a class="reference external" href="https://bugs.python.org/issue8413">bpo-8413</a>.)
</li>
<li>Warnings are now easier to control using the <a class="reference internal" href="../using/cmdline.html#envvar-PYTHONWARNINGS"><code class="xref std std-envvar docutils literal notranslate">PYTHONWARNINGS</code></a>
environment variable as an alternative to using <code class="docutils literal notranslate">-W</code> at the command line:
<div class="highlight-shell-session notranslate"><div class="highlight"><pre>$ export PYTHONWARNINGS=&#39;ignore::RuntimeWarning::,once::UnicodeWarning::&#39;
</pre></div>
</div>
(Suggested by Barry Warsaw and implemented by Philip Jenvey in <a class="reference external" href="https://bugs.python.org/issue7301">bpo-7301</a>.)
</li>
<li>A new warning category, <a class="reference internal" href="../library/exceptions.html#ResourceWarning" title="ResourceWarning"><code class="xref py py-exc docutils literal notranslate">ResourceWarning</code></a>, has been added. It is
emitted when potential issues with resource consumption or cleanup
are detected. It is silenced by default in normal release builds but
can be enabled through the means provided by the <a class="reference internal" href="../library/warnings.html#module-warnings" title="warnings: Issue warning messages and control their disposition."><code class="xref py py-mod docutils literal notranslate">warnings</code></a>
module, or on the command line.
A <a class="reference internal" href="../library/exceptions.html#ResourceWarning" title="ResourceWarning"><code class="xref py py-exc docutils literal notranslate">ResourceWarning</code></a> is issued at interpreter shutdown if the
<a class="reference internal" href="../library/gc.html#gc.garbage" title="gc.garbage"><code class="xref py py-data docutils literal notranslate">gc.garbage</code></a> list isn’t empty, and if <a class="reference internal" href="../library/gc.html#gc.DEBUG_UNCOLLECTABLE" title="gc.DEBUG_UNCOLLECTABLE"><code class="xref py py-attr docutils literal notranslate">gc.DEBUG_UNCOLLECTABLE</code></a> is
set, all uncollectable objects are printed. This is meant to make the
programmer aware that their code contains object finalization issues.
A <a class="reference internal" href="../library/exceptions.html#ResourceWarning" title="ResourceWarning"><code class="xref py py-exc docutils literal notranslate">ResourceWarning</code></a> is also issued when a <a class="reference internal" href="../glossary.html#term-file-object">file object</a> is destroyed
without having been explicitly closed. While the deallocator for such
object ensures it closes the underlying operating system resource
(usually, a file descriptor), the delay in deallocating the object could
produce various issues, especially under Windows. Here is an example
of enabling the warning from the command line:
<div class="highlight-shell-session notranslate"><div class="highlight"><pre>$ python -q -Wdefault
&gt;&gt;&gt; f = open(&quot;foo&quot;, &quot;wb&quot;)
&gt;&gt;&gt; del f
__main__:1: ResourceWarning: unclosed file &lt;_io.BufferedWriter name=&#39;foo&#39;&gt;
</pre></div>
</div>
(Added by Antoine Pitrou and Georg Brandl in <a class="reference external" href="https://bugs.python.org/issue10093">bpo-10093</a> and <a class="reference external" href="https://bugs.python.org/issue477863">bpo-477863</a>.)
</li>
<li><a class="reference internal" href="../library/stdtypes.html#range" title="range"><code class="xref py py-class docutils literal notranslate">range</code></a> objects now support index and count methods. This is part
of an effort to make more objects fully implement the
<code class="xref py py-class docutils literal notranslate">collections.Sequence</code> <a class="reference internal" href="../glossary.html#term-abstract-base-class">abstract base class</a>. As a result, the
language will have a more uniform API. In addition, <a class="reference internal" href="../library/stdtypes.html#range" title="range"><code class="xref py py-class docutils literal notranslate">range</code></a> objects
now support slicing and negative indices, even with values larger than
<a class="reference internal" href="../library/sys.html#sys.maxsize" title="sys.maxsize"><code class="xref py py-attr docutils literal notranslate">sys.maxsize</code></a>. This makes range more interoperable with lists:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; range(0, 100, 2).count(10)
1
&gt;&gt;&gt; range(0, 100, 2).index(10)
5
&gt;&gt;&gt; range(0, 100, 2)[5]
10
&gt;&gt;&gt; range(0, 100, 2)[0:5]
range(0, 10, 2)
</pre></div>
</div>
(Contributed by Daniel Stutzbach in <a class="reference external" href="https://bugs.python.org/issue9213">bpo-9213</a>, by Alexander Belopolsky
in <a class="reference external" href="https://bugs.python.org/issue2690">bpo-2690</a>, and by Nick Coghlan in <a class="reference external" href="https://bugs.python.org/issue10889">bpo-10889</a>.)
</li>
<li>The <a class="reference internal" href="../library/functions.html#callable" title="callable"><code class="xref py py-func docutils literal notranslate">callable()</code></a> builtin function from Py2.x was resurrected. It provides
a concise, readable alternative to using an <a class="reference internal" href="../glossary.html#term-abstract-base-class">abstract base class</a> in an
expression like <code class="docutils literal notranslate">isinstance(x, collections.Callable)</code>:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; callable(max)
True
&gt;&gt;&gt; callable(20)
False
</pre></div>
</div>
(See <a class="reference external" href="https://bugs.python.org/issue10518">bpo-10518</a>.)
</li>
<li>Python’s import mechanism can now load modules installed in directories with
non-ASCII characters in the path name. This solved an aggravating problem
with home directories for users with non-ASCII characters in their usernames.
</li>
</ul>
<blockquote>
<div>(Required extensive work by Victor Stinner in <a class="reference external" href="https://bugs.python.org/issue9425">bpo-9425</a>.)</div></blockquote>
</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>
Python’s standard library has undergone significant maintenance efforts and
quality improvements.
The biggest news for Python 3.2 is that the <a class="reference internal" href="../library/email.html#module-email" title="email: Package supporting the parsing, manipulating, and generating email messages."><code class="xref py py-mod docutils literal notranslate">email</code></a> package, <a class="reference internal" href="../library/mailbox.html#module-mailbox" title="mailbox: Manipulate mailboxes in various formats"><code class="xref py py-mod docutils literal notranslate">mailbox</code></a>
module, and <a class="reference internal" href="../library/nntplib.html#module-nntplib" title="nntplib: NNTP protocol client (requires sockets)."><code class="xref py py-mod docutils literal notranslate">nntplib</code></a> modules now work correctly with the bytes/text model
in Python 3. For the first time, there is correct handling of messages with
mixed encodings.
Throughout the standard library, there has been more careful attention to
encodings and text versus bytes issues. In particular, interactions with the
operating system are now better able to exchange non-ASCII data using the
Windows MBCS encoding, locale-aware encodings, or UTF-8.
Another significant win is the addition of substantially better support for
SSL connections and security certificates.
In addition, more classes now implement a <a class="reference internal" href="../glossary.html#term-context-manager">context manager</a> to support
convenient and reliable resource clean-up using a <a class="reference internal" href="../reference/compound_stmts.html#with"><code class="xref std std-keyword docutils literal notranslate">with</code></a> statement.
<div class="section" id="email">
<h3>email<a class="headerlink" href="#email" title="Permalink to this headline">¶</a></h3>
The usability of the <a class="reference internal" href="../library/email.html#module-email" title="email: Package supporting the parsing, manipulating, and generating email messages."><code class="xref py py-mod docutils literal notranslate">email</code></a> package in Python 3 has been mostly fixed by
the extensive efforts of R. David Murray. The problem was that emails are
typically read and stored in the form of <a class="reference internal" href="../library/stdtypes.html#bytes" title="bytes"><code class="xref py py-class docutils literal notranslate">bytes</code></a> rather than <a class="reference internal" href="../library/stdtypes.html#str" title="str"><code class="xref py py-class docutils literal notranslate">str</code></a>
text, and they may contain multiple encodings within a single email. So, the
email package had to be extended to parse and generate email messages in bytes
format.
<ul>
<li>New functions <a class="reference internal" href="../library/email.parser.html#email.message_from_bytes" title="email.message_from_bytes"><code class="xref py py-func docutils literal notranslate">message_from_bytes()</code></a> and
<a class="reference internal" href="../library/email.parser.html#email.message_from_binary_file" title="email.message_from_binary_file"><code class="xref py py-func docutils literal notranslate">message_from_binary_file()</code></a>, and new classes
<a class="reference internal" href="../library/email.parser.html#email.parser.BytesFeedParser" title="email.parser.BytesFeedParser"><code class="xref py py-class docutils literal notranslate">BytesFeedParser</code></a> and <a class="reference internal" href="../library/email.parser.html#email.parser.BytesParser" title="email.parser.BytesParser"><code class="xref py py-class docutils literal notranslate">BytesParser</code></a>
allow binary message data to be parsed into model objects.
</li>
<li>Given bytes input to the model, <a class="reference internal" href="../library/email.compat32-message.html#email.message.Message.get_payload" title="email.message.Message.get_payload"><code class="xref py py-meth docutils literal notranslate">get_payload()</code></a>
will by default decode a message body that has a
Content-Transfer-Encoding of 8bit using the charset
specified in the MIME headers and return the resulting string.
</li>
<li>Given bytes input to the model, <a class="reference internal" href="../library/email.generator.html#email.generator.Generator" title="email.generator.Generator"><code class="xref py py-class docutils literal notranslate">Generator</code></a> will
convert message bodies that have a Content-Transfer-Encoding of
8bit to instead have a 7bit Content-Transfer-Encoding.
Headers with unencoded non-ASCII bytes are deemed to be <a class="rfc reference external" href="https://tools.ietf.org/html/rfc2047.html">RFC 2047</a>-encoded
using the unknown-8bit character set.
</li>
<li>A new class <a class="reference internal" href="../library/email.generator.html#email.generator.BytesGenerator" title="email.generator.BytesGenerator"><code class="xref py py-class docutils literal notranslate">BytesGenerator</code></a> produces bytes as output,
preserving any unchanged non-ASCII data that was present in the input used to
build the model, including message bodies with a
Content-Transfer-Encoding of 8bit.
</li>
<li>The <a class="reference internal" href="../library/smtplib.html#module-smtplib" title="smtplib: SMTP protocol client (requires sockets)."><code class="xref py py-mod docutils literal notranslate">smtplib</code></a> <a class="reference internal" href="../library/smtplib.html#smtplib.SMTP" title="smtplib.SMTP"><code class="xref py py-class docutils literal notranslate">SMTP</code></a> class now accepts a byte string
for the msg argument to the <a class="reference internal" href="../library/smtplib.html#smtplib.SMTP.sendmail" title="smtplib.SMTP.sendmail"><code class="xref py py-meth docutils literal notranslate">sendmail()</code></a> method,
and a new method, <a class="reference internal" href="../library/smtplib.html#smtplib.SMTP.send_message" title="smtplib.SMTP.send_message"><code class="xref py py-meth docutils literal notranslate">send_message()</code></a> accepts a
<a class="reference internal" href="../library/email.compat32-message.html#email.message.Message" title="email.message.Message"><code class="xref py py-class docutils literal notranslate">Message</code></a> object and can optionally obtain the
from_addr and to_addrs addresses directly from the object.
</li>
</ul>
(Proposed and implemented by R. David Murray, <a class="reference external" href="https://bugs.python.org/issue4661">bpo-4661</a> and <a class="reference external" href="https://bugs.python.org/issue10321">bpo-10321</a>.)
</div>
<div class="section" id="elementtree">
<h3>elementtree<a class="headerlink" href="#elementtree" title="Permalink to this headline">¶</a></h3>
The <a class="reference internal" href="../library/xml.etree.elementtree.html#module-xml.etree.ElementTree" title="xml.etree.ElementTree: Implementation of the ElementTree API."><code class="xref py py-mod docutils literal notranslate">xml.etree.ElementTree</code></a> package and its <code class="xref py py-mod docutils literal notranslate">xml.etree.cElementTree</code>
counterpart have been updated to version 1.3.
Several new and useful functions and methods have been added:
<ul class="simple">
<li><a class="reference internal" href="../library/xml.etree.elementtree.html#xml.etree.ElementTree.fromstringlist" title="xml.etree.ElementTree.fromstringlist"><code class="xref py py-func docutils literal notranslate">xml.etree.ElementTree.fromstringlist()</code></a> which builds an XML document
from a sequence of fragments</li>
<li><a class="reference internal" href="../library/xml.etree.elementtree.html#xml.etree.ElementTree.register_namespace" title="xml.etree.ElementTree.register_namespace"><code class="xref py py-func docutils literal notranslate">xml.etree.ElementTree.register_namespace()</code></a> for registering a global
namespace prefix</li>
<li><a class="reference internal" href="../library/xml.etree.elementtree.html#xml.etree.ElementTree.tostringlist" title="xml.etree.ElementTree.tostringlist"><code class="xref py py-func docutils literal notranslate">xml.etree.ElementTree.tostringlist()</code></a> for string representation
including all sublists</li>
<li><a class="reference internal" href="../library/xml.etree.elementtree.html#xml.etree.ElementTree.Element.extend" title="xml.etree.ElementTree.Element.extend"><code class="xref py py-meth docutils literal notranslate">xml.etree.ElementTree.Element.extend()</code></a> for appending a sequence of zero
or more elements</li>
<li><a class="reference internal" href="../library/xml.etree.elementtree.html#xml.etree.ElementTree.Element.iterfind" title="xml.etree.ElementTree.Element.iterfind"><code class="xref py py-meth docutils literal notranslate">xml.etree.ElementTree.Element.iterfind()</code></a> searches an element and
subelements</li>
<li><a class="reference internal" href="../library/xml.etree.elementtree.html#xml.etree.ElementTree.Element.itertext" title="xml.etree.ElementTree.Element.itertext"><code class="xref py py-meth docutils literal notranslate">xml.etree.ElementTree.Element.itertext()</code></a> creates a text iterator over
an element and its subelements</li>
<li><a class="reference internal" href="../library/xml.etree.elementtree.html#xml.etree.ElementTree.TreeBuilder.end" title="xml.etree.ElementTree.TreeBuilder.end"><code class="xref py py-meth docutils literal notranslate">xml.etree.ElementTree.TreeBuilder.end()</code></a> closes the current element</li>
<li><a class="reference internal" href="../library/xml.etree.elementtree.html#xml.etree.ElementTree.TreeBuilder.doctype" title="xml.etree.ElementTree.TreeBuilder.doctype"><code class="xref py py-meth docutils literal notranslate">xml.etree.ElementTree.TreeBuilder.doctype()</code></a> handles a doctype
declaration</li>
</ul>
Two methods have been deprecated:
<ul class="simple">
<li><code class="xref py py-meth docutils literal notranslate">xml.etree.ElementTree.getchildren()</code> use <code class="docutils literal notranslate">list(elem)</code> instead.</li>
<li><code class="xref py py-meth docutils literal notranslate">xml.etree.ElementTree.getiterator()</code> use <code class="docutils literal notranslate">Element.iter</code> instead.</li>
</ul>
For details of the update, see <a class="reference external" href="http://effbot.org/zone/elementtree-13-intro.htm">Introducing ElementTree</a> on Fredrik Lundh’s website.
(Contributed by Florent Xicluna and Fredrik Lundh, <a class="reference external" href="https://bugs.python.org/issue6472">bpo-6472</a>.)
</div>
<div class="section" id="functools">
<h3>functools<a class="headerlink" href="#functools" title="Permalink to this headline">¶</a></h3>
<ul>
<li>The <a class="reference internal" href="../library/functools.html#module-functools" title="functools: Higher-order functions and operations on callable objects."><code class="xref py py-mod docutils literal notranslate">functools</code></a> module includes a new decorator for caching function
calls. <a class="reference internal" href="../library/functools.html#functools.lru_cache" title="functools.lru_cache"><code class="xref py py-func docutils literal notranslate">functools.lru_cache()</code></a> can save repeated queries to an external
resource whenever the results are expected to be the same.
For example, adding a caching decorator to a database query function can save
database accesses for popular searches:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; import functools
&gt;&gt;&gt; @functools.lru_cache(maxsize=300)
... def get_phone_number(name):
... c = conn.cursor()
... c.execute(&#39;SELECT phonenumber FROM phonelist WHERE name=?&#39;, (name,))
... return c.fetchone()[0]
</pre></div>
</div>
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; for name in user_requests: 
... get_phone_number(name) # cached lookup
</pre></div>
</div>
To help with choosing an effective cache size, the wrapped function is
instrumented for tracking cache statistics:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; get_phone_number.cache_info() 
CacheInfo(hits=4805, misses=980, maxsize=300, currsize=300)
</pre></div>
</div>
If the phonelist table gets updated, the outdated contents of the cache can be
cleared with:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; get_phone_number.cache_clear()
</pre></div>
</div>
(Contributed by Raymond Hettinger and incorporating design ideas from Jim
Baker, Miki Tebeka, and Nick Coghlan; see <a class="reference external" href="https://code.activestate.com/recipes/498245">recipe 498245</a>, <a class="reference external" href="https://code.activestate.com/recipes/577479">recipe 577479</a>, <a class="reference external" href="https://bugs.python.org/issue10586">bpo-10586</a>, and
<a class="reference external" href="https://bugs.python.org/issue10593">bpo-10593</a>.)
</li>
<li>The <a class="reference internal" href="../library/functools.html#functools.wraps" title="functools.wraps"><code class="xref py py-func docutils literal notranslate">functools.wraps()</code></a> decorator now adds a <code class="xref py py-attr docutils literal notranslate">__wrapped__</code> attribute
pointing to the original callable function. This allows wrapped functions to
be introspected. It also copies <code class="xref py py-attr docutils literal notranslate">__annotations__</code> if defined. And now
it also gracefully skips over missing attributes such as <code class="xref py py-attr docutils literal notranslate">__doc__</code> which
might not be defined for the wrapped callable.
In the above example, the cache can be removed by recovering the original
function:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; get_phone_number = get_phone_number.__wrapped__ # uncached function
</pre></div>
</div>
(By Nick Coghlan and Terrence Cole; <a class="reference external" href="https://bugs.python.org/issue9567">bpo-9567</a>, <a class="reference external" href="https://bugs.python.org/issue3445">bpo-3445</a>, and
<a class="reference external" href="https://bugs.python.org/issue8814">bpo-8814</a>.)
</li>
<li>To help write classes with rich comparison methods, a new decorator
<a class="reference internal" href="../library/functools.html#functools.total_ordering" title="functools.total_ordering"><code class="xref py py-func docutils literal notranslate">functools.total_ordering()</code></a> will use existing equality and inequality
methods to fill in the remaining methods.
For example, supplying __eq__ and __lt__ will enable
<a class="reference internal" href="../library/functools.html#functools.total_ordering" title="functools.total_ordering"><code class="xref py py-func docutils literal notranslate">total_ordering()</code></a> to fill-in __le__, __gt__ and __ge__:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>@total_ordering
class Student:
 def __eq__(self, other):
 return ((self.lastname.lower(), self.firstname.lower()) ==
 (other.lastname.lower(), other.firstname.lower()))

def __lt__(self, other):
 return ((self.lastname.lower(), self.firstname.lower()) &lt;
 (other.lastname.lower(), other.firstname.lower()))
</pre></div>
</div>
With the total_ordering decorator, the remaining comparison methods
are filled in automatically.
(Contributed by Raymond Hettinger.)
</li>
<li>To aid in porting programs from Python 2, the <a class="reference internal" href="../library/functools.html#functools.cmp_to_key" title="functools.cmp_to_key"><code class="xref py py-func docutils literal notranslate">functools.cmp_to_key()</code></a>
function converts an old-style comparison function to
modern <a class="reference internal" href="../glossary.html#term-key-function">key function</a>:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; # locale-aware sort order
&gt;&gt;&gt; sorted(iterable, key=cmp_to_key(locale.strcoll)) 
</pre></div>
</div>
For sorting examples and a brief sorting tutorial, see the <a class="reference external" href="https://wiki.python.org/moin/HowTo/Sorting/">Sorting HowTo</a> tutorial.
(Contributed by Raymond Hettinger.)
</li>
</ul>
</div>
<div class="section" id="itertools">
<h3>itertools<a class="headerlink" href="#itertools" title="Permalink to this headline">¶</a></h3>
<ul>
<li>The <a class="reference internal" href="../library/itertools.html#module-itertools" title="itertools: Functions creating iterators for efficient looping."><code class="xref py py-mod docutils literal notranslate">itertools</code></a> module has a new <a class="reference internal" href="../library/itertools.html#itertools.accumulate" title="itertools.accumulate"><code class="xref py py-func docutils literal notranslate">accumulate()</code></a> function
modeled on APL’s scan operator and Numpy’s accumulate function:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; from itertools import accumulate
&gt;&gt;&gt; list(accumulate([8, 2, 50]))
[8, 10, 60]
</pre></div>
</div>
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; prob_dist = [0.1, 0.4, 0.2, 0.3]
&gt;&gt;&gt; list(accumulate(prob_dist)) # cumulative probability distribution
[0.1, 0.5, 0.7, 1.0]
</pre></div>
</div>
For an example using <a class="reference internal" href="../library/itertools.html#itertools.accumulate" title="itertools.accumulate"><code class="xref py py-func docutils literal notranslate">accumulate()</code></a>, see the <a class="reference internal" href="../library/random.html#random-examples">examples for
the random module</a>.
(Contributed by Raymond Hettinger and incorporating design suggestions
from Mark Dickinson.)
</li>
</ul>
</div>
<div class="section" id="collections">
<h3>collections<a class="headerlink" href="#collections" title="Permalink to this headline">¶</a></h3>
<ul>
<li>The <a class="reference internal" href="../library/collections.html#collections.Counter" title="collections.Counter"><code class="xref py py-class docutils literal notranslate">collections.Counter</code></a> class now has two forms of in-place
subtraction, the existing -= operator for <a class="reference external" href="https://en.wikipedia.org/wiki/Saturation_arithmetic">saturating subtraction</a> and the new
<a class="reference internal" href="../library/collections.html#collections.Counter.subtract" title="collections.Counter.subtract"><code class="xref py py-meth docutils literal notranslate">subtract()</code></a> method for regular subtraction. The
former is suitable for <a class="reference external" href="https://en.wikipedia.org/wiki/Multiset">multisets</a>
which only have positive counts, and the latter is more suitable for use cases
that allow negative counts:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; from collections import Counter
&gt;&gt;&gt; tally = Counter(dogs=5, cats=3)
&gt;&gt;&gt; tally -= Counter(dogs=2, cats=8) # saturating subtraction
&gt;&gt;&gt; tally
Counter({&#39;dogs&#39;: 3})
</pre></div>
</div>
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; tally = Counter(dogs=5, cats=3)
&gt;&gt;&gt; tally.subtract(dogs=2, cats=8) # regular subtraction
&gt;&gt;&gt; tally
Counter({&#39;dogs&#39;: 3, &#39;cats&#39;: -5})
</pre></div>
</div>
(Contributed by Raymond Hettinger.)
</li>
<li>The <a class="reference internal" href="../library/collections.html#collections.OrderedDict" title="collections.OrderedDict"><code class="xref py py-class docutils literal notranslate">collections.OrderedDict</code></a> class has a new method
<a class="reference internal" href="../library/collections.html#collections.OrderedDict.move_to_end" title="collections.OrderedDict.move_to_end"><code class="xref py py-meth docutils literal notranslate">move_to_end()</code></a> which takes an existing key and
moves it to either the first or last position in the ordered sequence.
The default is to move an item to the last position. This is equivalent of
renewing an entry with <code class="docutils literal notranslate">od[k] = od.pop(k)</code>.
A fast move-to-end operation is useful for resequencing entries. For example,
an ordered dictionary can be used to track order of access by aging entries
from the oldest to the most recently accessed.
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; from collections import OrderedDict
&gt;&gt;&gt; d = OrderedDict.fromkeys([&#39;a&#39;, &#39;b&#39;, &#39;X&#39;, &#39;d&#39;, &#39;e&#39;])
&gt;&gt;&gt; list(d)
[&#39;a&#39;, &#39;b&#39;, &#39;X&#39;, &#39;d&#39;, &#39;e&#39;]
&gt;&gt;&gt; d.move_to_end(&#39;X&#39;)
&gt;&gt;&gt; list(d)
[&#39;a&#39;, &#39;b&#39;, &#39;d&#39;, &#39;e&#39;, &#39;X&#39;]
</pre></div>
</div>
(Contributed by Raymond Hettinger.)
</li>
<li>The <a class="reference internal" href="../library/collections.html#collections.deque" title="collections.deque"><code class="xref py py-class docutils literal notranslate">collections.deque</code></a> class grew two new methods
<a class="reference internal" href="../library/collections.html#collections.deque.count" title="collections.deque.count"><code class="xref py py-meth docutils literal notranslate">count()</code></a> and <a class="reference internal" href="../library/collections.html#collections.deque.reverse" title="collections.deque.reverse"><code class="xref py py-meth docutils literal notranslate">reverse()</code></a> that
make them more substitutable for <a class="reference internal" href="../library/stdtypes.html#list" title="list"><code class="xref py py-class docutils literal notranslate">list</code></a> objects:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; from collections import deque
&gt;&gt;&gt; d = deque(&#39;simsalabim&#39;)
&gt;&gt;&gt; d.count(&#39;s&#39;)
2
&gt;&gt;&gt; d.reverse()
&gt;&gt;&gt; d
deque([&#39;m&#39;, &#39;i&#39;, &#39;b&#39;, &#39;a&#39;, &#39;l&#39;, &#39;a&#39;, &#39;s&#39;, &#39;m&#39;, &#39;i&#39;, &#39;s&#39;])
</pre></div>
</div>
(Contributed by Raymond Hettinger.)
</li>
</ul>
</div>
<div class="section" id="threading">
<h3>threading<a class="headerlink" href="#threading" title="Permalink to this headline">¶</a></h3>
The <a class="reference internal" href="../library/threading.html#module-threading" title="threading: Thread-based parallelism."><code class="xref py py-mod docutils literal notranslate">threading</code></a> module has a new <a class="reference internal" href="../library/threading.html#threading.Barrier" title="threading.Barrier"><code class="xref py py-class docutils literal notranslate">Barrier</code></a>
synchronization class for making multiple threads wait until all of them have
reached a common barrier point. Barriers are useful for making sure that a task
with multiple preconditions does not run until all of the predecessor tasks are
complete.
Barriers can work with an arbitrary number of threads. This is a generalization
of a <a class="reference external" href="https://en.wikipedia.org/wiki/Synchronous_rendezvous">Rendezvous</a> which
is defined for only two threads.
Implemented as a two-phase cyclic barrier, <a class="reference internal" href="../library/threading.html#threading.Barrier" title="threading.Barrier"><code class="xref py py-class docutils literal notranslate">Barrier</code></a> objects
are suitable for use in loops. The separate filling and draining phases
assure that all threads get released (drained) before any one of them can loop
back and re-enter the barrier. The barrier fully resets after each cycle.
Example of using barriers:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>from threading import Barrier, Thread

def get_votes(site):
 ballots = conduct_election(site)
 all_polls_closed.wait() # do not count until all polls are closed
 totals = summarize(ballots)
 publish(site, totals)

all_polls_closed = Barrier(len(sites))
for site in sites:
 Thread(target=get_votes, args=(site,)).start()
</pre></div>
</div>
In this example, the barrier enforces a rule that votes cannot be counted at any
polling site until all polls are closed. Notice how a solution with a barrier
is similar to one with <a class="reference internal" href="../library/threading.html#threading.Thread.join" title="threading.Thread.join"><code class="xref py py-meth docutils literal notranslate">threading.Thread.join()</code></a>, but the threads stay alive
and continue to do work (summarizing ballots) after the barrier point is
crossed.
If any of the predecessor tasks can hang or be delayed, a barrier can be created
with an optional timeout parameter. Then if the timeout period elapses before
all the predecessor tasks reach the barrier point, all waiting threads are
released and a <a class="reference internal" href="../library/threading.html#threading.BrokenBarrierError" title="threading.BrokenBarrierError"><code class="xref py py-exc docutils literal notranslate">BrokenBarrierError</code></a> exception is raised:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>def get_votes(site):
 ballots = conduct_election(site)
 try:
 all_polls_closed.wait(timeout=midnight - time.now())
 except BrokenBarrierError:
 lockbox = seal_ballots(ballots)
 queue.put(lockbox)
 else:
 totals = summarize(ballots)
 publish(site, totals)
</pre></div>
</div>
In this example, the barrier enforces a more robust rule. If some election
sites do not finish before midnight, the barrier times-out and the ballots are
sealed and deposited in a queue for later handling.
See <a class="reference external" href="https://parlab.eecs.berkeley.edu/wiki/_media/patterns/paraplop_g1_3.pdf">Barrier Synchronization Patterns</a> for
more examples of how barriers can be used in parallel computing. Also, there is
a simple but thorough explanation of barriers in <a class="reference external" href="http://greenteapress.com/semaphores/downey08semaphores.pdf">The Little Book of Semaphores</a>, section 3.6.
(Contributed by Kristján Valur Jónsson with an API review by Jeffrey Yasskin in
<a class="reference external" href="https://bugs.python.org/issue8777">bpo-8777</a>.)
</div>
<div class="section" id="datetime-and-time">
<h3>datetime and time<a class="headerlink" href="#datetime-and-time" title="Permalink to this headline">¶</a></h3>
<ul>
<li>The <a class="reference internal" href="../library/datetime.html#module-datetime" title="datetime: Basic date and time types."><code class="xref py py-mod docutils literal notranslate">datetime</code></a> module has a new type <a class="reference internal" href="../library/datetime.html#datetime.timezone" title="datetime.timezone"><code class="xref py py-class docutils literal notranslate">timezone</code></a> that
implements the <a class="reference internal" href="../library/datetime.html#datetime.tzinfo" title="datetime.tzinfo"><code class="xref py py-class docutils literal notranslate">tzinfo</code></a> interface by returning a fixed UTC
offset and timezone name. This makes it easier to create timezone-aware
datetime objects:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; from datetime import datetime, timezone

&gt;&gt;&gt; datetime.now(timezone.utc)
datetime.datetime(2010, 12, 8, 21, 4, 2, 923754, tzinfo=datetime.timezone.utc)

&gt;&gt;&gt; datetime.strptime(&quot;01/01/2000 12:00 +0000&quot;, &quot;%m/%d/%Y %H:%M %z&quot;)
datetime.datetime(2000, 1, 1, 12, 0, tzinfo=datetime.timezone.utc)
</pre></div>
</div>
</li>
<li>Also, <a class="reference internal" href="../library/datetime.html#datetime.timedelta" title="datetime.timedelta"><code class="xref py py-class docutils literal notranslate">timedelta</code></a> objects can now be multiplied by
<a class="reference internal" href="../library/functions.html#float" title="float"><code class="xref py py-class docutils literal notranslate">float</code></a> and divided by <a class="reference internal" href="../library/functions.html#float" title="float"><code class="xref py py-class docutils literal notranslate">float</code></a> and <a class="reference internal" href="../library/functions.html#int" title="int"><code class="xref py py-class docutils literal notranslate">int</code></a> objects.
And <a class="reference internal" href="../library/datetime.html#datetime.timedelta" title="datetime.timedelta"><code class="xref py py-class docutils literal notranslate">timedelta</code></a> objects can now divide one another.
</li>
<li>The <a class="reference internal" href="../library/datetime.html#datetime.date.strftime" title="datetime.date.strftime"><code class="xref py py-meth docutils literal notranslate">datetime.date.strftime()</code></a> method is no longer restricted to years
after 1900. The new supported year range is from 1000 to 9999 inclusive.
</li>
<li>Whenever a two-digit year is used in a time tuple, the interpretation has been
governed by <code class="xref py py-attr docutils literal notranslate">time.accept2dyear</code>. The default is <code class="docutils literal notranslate">True</code> which means that
for a two-digit year, the century is guessed according to the POSIX rules
governing the <code class="docutils literal notranslate">%y</code> strptime format.
Starting with Py3.2, use of the century guessing heuristic will emit a
<a class="reference internal" href="../library/exceptions.html#DeprecationWarning" title="DeprecationWarning"><code class="xref py py-exc docutils literal notranslate">DeprecationWarning</code></a>. Instead, it is recommended that
<code class="xref py py-attr docutils literal notranslate">time.accept2dyear</code> be set to <code class="docutils literal notranslate">False</code> so that large date ranges
can be used without guesswork:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; import time, warnings
&gt;&gt;&gt; warnings.resetwarnings() # remove the default warning filters

&gt;&gt;&gt; time.accept2dyear = False # use the full range of allowable dates
&gt;&gt;&gt; time.asctime((11, 1, 1, 12, 34, 56, 4, 1, 0))
&#39;Fri Jan 1 12:34:56 11&#39;
</pre></div>
</div>
Several functions now have significantly expanded date ranges. When
<code class="xref py py-attr docutils literal notranslate">time.accept2dyear</code> is false, the <a class="reference internal" href="../library/time.html#time.asctime" title="time.asctime"><code class="xref py py-func docutils literal notranslate">time.asctime()</code></a> function will
accept any year that fits in a C int, while the <a class="reference internal" href="../library/time.html#time.mktime" title="time.mktime"><code class="xref py py-func docutils literal notranslate">time.mktime()</code></a> and
<a class="reference internal" href="../library/time.html#time.strftime" title="time.strftime"><code class="xref py py-func docutils literal notranslate">time.strftime()</code></a> functions will accept the full range supported by the
corresponding operating system functions.
</li>
</ul>
(Contributed by Alexander Belopolsky and Victor Stinner in <a class="reference external" href="https://bugs.python.org/issue1289118">bpo-1289118</a>,
<a class="reference external" href="https://bugs.python.org/issue5094">bpo-5094</a>, <a class="reference external" href="https://bugs.python.org/issue6641">bpo-6641</a>, <a class="reference external" href="https://bugs.python.org/issue2706">bpo-2706</a>, <a class="reference external" href="https://bugs.python.org/issue1777412">bpo-1777412</a>, <a class="reference external" href="https://bugs.python.org/issue8013">bpo-8013</a>,
and <a class="reference external" href="https://bugs.python.org/issue10827">bpo-10827</a>.)
</div>
<div class="section" id="math">
<h3>math<a class="headerlink" href="#math" title="Permalink to this headline">¶</a></h3>
The <a class="reference internal" href="../library/math.html#module-math" title="math: Mathematical functions (sin() etc.)."><code class="xref py py-mod docutils literal notranslate">math</code></a> module has been updated with six new functions inspired by the
C99 standard.
The <a class="reference internal" href="../library/math.html#math.isfinite" title="math.isfinite"><code class="xref py py-func docutils literal notranslate">isfinite()</code></a> function provides a reliable and fast way to detect
special values. It returns <code class="docutils literal notranslate">True</code> for regular numbers and <code class="docutils literal notranslate">False</code> for Nan or
Infinity:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; from math import isfinite
&gt;&gt;&gt; [isfinite(x) for x in (123, 4.56, float(&#39;Nan&#39;), float(&#39;Inf&#39;))]
[True, True, False, False]
</pre></div>
</div>
The <a class="reference internal" href="../library/math.html#math.expm1" title="math.expm1"><code class="xref py py-func docutils literal notranslate">expm1()</code></a> function computes <code class="docutils literal notranslate">e**x-1</code> for small values of x
without incurring the loss of precision that usually accompanies the subtraction
of nearly equal quantities:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; from math import expm1
&gt;&gt;&gt; expm1(0.013671875) # more accurate way to compute e**x-1 for a small x
0.013765762467652909
</pre></div>
</div>
The <a class="reference internal" href="../library/math.html#math.erf" title="math.erf"><code class="xref py py-func docutils literal notranslate">erf()</code></a> function computes a probability integral or <a class="reference external" href="https://en.wikipedia.org/wiki/Error_function">Gaussian
error function</a>. The
complementary error function, <a class="reference internal" href="../library/math.html#math.erfc" title="math.erfc"><code class="xref py py-func docutils literal notranslate">erfc()</code></a>, is <code class="docutils literal notranslate">1 - erf(x)</code>:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; from math import erf, erfc, sqrt
&gt;&gt;&gt; erf(1.0/sqrt(2.0)) # portion of normal distribution within 1 standard deviation
0.682689492137086
&gt;&gt;&gt; erfc(1.0/sqrt(2.0)) # portion of normal distribution outside 1 standard deviation
0.31731050786291404
&gt;&gt;&gt; erf(1.0/sqrt(2.0)) + erfc(1.0/sqrt(2.0))
1.0
</pre></div>
</div>
The <a class="reference internal" href="../library/math.html#math.gamma" title="math.gamma"><code class="xref py py-func docutils literal notranslate">gamma()</code></a> function is a continuous extension of the factorial
function. See <a class="reference external" href="https://en.wikipedia.org/wiki/Gamma_function">https://en.wikipedia.org/wiki/Gamma_function</a> for details. Because
the function is related to factorials, it grows large even for small values of
x, so there is also a <a class="reference internal" href="../library/math.html#math.lgamma" title="math.lgamma"><code class="xref py py-func docutils literal notranslate">lgamma()</code></a> function for computing the natural
logarithm of the gamma function:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; from math import gamma, lgamma
&gt;&gt;&gt; gamma(7.0) # six factorial
720.0
&gt;&gt;&gt; lgamma(801.0) # log(800 factorial)
4551.950730698041
</pre></div>
</div>
(Contributed by Mark Dickinson.)
</div>
<div class="section" id="abc">
<h3>abc<a class="headerlink" href="#abc" title="Permalink to this headline">¶</a></h3>
The <a class="reference internal" href="../library/abc.html#module-abc" title="abc: Abstract base classes according to PEP 3119."><code class="xref py py-mod docutils literal notranslate">abc</code></a> module now supports <a class="reference internal" href="../library/abc.html#abc.abstractclassmethod" title="abc.abstractclassmethod"><code class="xref py py-func docutils literal notranslate">abstractclassmethod()</code></a> and
<a class="reference internal" href="../library/abc.html#abc.abstractstaticmethod" title="abc.abstractstaticmethod"><code class="xref py py-func docutils literal notranslate">abstractstaticmethod()</code></a>.
These tools make it possible to define an <a class="reference internal" href="../glossary.html#term-abstract-base-class">abstract base class</a> that
requires a particular <a class="reference internal" href="../library/functions.html#classmethod" title="classmethod"><code class="xref py py-func docutils literal notranslate">classmethod()</code></a> or <a class="reference internal" href="../library/functions.html#staticmethod" title="staticmethod"><code class="xref py py-func docutils literal notranslate">staticmethod()</code></a> to be
implemented:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>class Temperature(metaclass=abc.ABCMeta):
 @abc.abstractclassmethod
 def from_fahrenheit(cls, t):
 ...
 @abc.abstractclassmethod
 def from_celsius(cls, t):
 ...
</pre></div>
</div>
(Patch submitted by Daniel Urban; <a class="reference external" href="https://bugs.python.org/issue5867">bpo-5867</a>.)
</div>
<div class="section" id="io">
<h3>io<a class="headerlink" href="#io" title="Permalink to this headline">¶</a></h3>
The <a class="reference internal" href="../library/io.html#io.BytesIO" title="io.BytesIO"><code class="xref py py-class docutils literal notranslate">io.BytesIO</code></a> has a new method, <a class="reference internal" href="../library/io.html#io.BytesIO.getbuffer" title="io.BytesIO.getbuffer"><code class="xref py py-meth docutils literal notranslate">getbuffer()</code></a>, which
provides functionality similar to <a class="reference internal" href="../library/stdtypes.html#memoryview" title="memoryview"><code class="xref py py-func docutils literal notranslate">memoryview()</code></a>. It creates an editable
view of the data without making a copy. The buffer’s random access and support
for slice notation are well-suited to in-place editing:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; REC_LEN, LOC_START, LOC_LEN = 34, 7, 11

&gt;&gt;&gt; def change_location(buffer, record_number, location):
... start = record_number * REC_LEN + LOC_START
... buffer[start: start+LOC_LEN] = location

&gt;&gt;&gt; import io

&gt;&gt;&gt; byte_stream = io.BytesIO(
... b&#39;G3805 storeroom Main chassis &#39;
... b&#39;X7899 shipping Reserve cog &#39;
... b&#39;L6988 receiving Primary sprocket&#39;
... )
&gt;&gt;&gt; buffer = byte_stream.getbuffer()
&gt;&gt;&gt; change_location(buffer, 1, b&#39;warehouse &#39;)
&gt;&gt;&gt; change_location(buffer, 0, b&#39;showroom &#39;)
&gt;&gt;&gt; print(byte_stream.getvalue())
b&#39;G3805 showroom Main chassis &#39;
b&#39;X7899 warehouse Reserve cog &#39;
b&#39;L6988 receiving Primary sprocket&#39;
</pre></div>
</div>
(Contributed by Antoine Pitrou in <a class="reference external" href="https://bugs.python.org/issue5506">bpo-5506</a>.)
</div>
<div class="section" id="reprlib">
<h3>reprlib<a class="headerlink" href="#reprlib" title="Permalink to this headline">¶</a></h3>
When writing a <a class="reference internal" href="../reference/datamodel.html#object.__repr__" title="object.__repr__"><code class="xref py py-meth docutils literal notranslate">__repr__()</code></a> method for a custom container, it is easy to
forget to handle the case where a member refers back to the container itself.
Python’s builtin objects such as <a class="reference internal" href="../library/stdtypes.html#list" title="list"><code class="xref py py-class docutils literal notranslate">list</code></a> and <a class="reference internal" href="../library/stdtypes.html#set" title="set"><code class="xref py py-class docutils literal notranslate">set</code></a> handle
self-reference by displaying “…” in the recursive part of the representation
string.
To help write such <a class="reference internal" href="../reference/datamodel.html#object.__repr__" title="object.__repr__"><code class="xref py py-meth docutils literal notranslate">__repr__()</code></a> methods, the <a class="reference internal" href="../library/reprlib.html#module-reprlib" title="reprlib: Alternate repr() implementation with size limits."><code class="xref py py-mod docutils literal notranslate">reprlib</code></a> module has a new
decorator, <a class="reference internal" href="../library/reprlib.html#reprlib.recursive_repr" title="reprlib.recursive_repr"><code class="xref py py-func docutils literal notranslate">recursive_repr()</code></a>, for detecting recursive calls to
<a class="reference internal" href="../reference/datamodel.html#object.__repr__" title="object.__repr__"><code class="xref py py-meth docutils literal notranslate">__repr__()</code></a> and substituting a placeholder string instead:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; class MyList(list):
... @recursive_repr()
... def __repr__(self):
... return &#39;&lt;&#39; + &#39;|&#39;.join(map(repr, self)) + &#39;&gt;&#39;
...
&gt;&gt;&gt; m = MyList(&#39;abc&#39;)
&gt;&gt;&gt; m.append(m)
&gt;&gt;&gt; m.append(&#39;x&#39;)
&gt;&gt;&gt; print(m)
&lt;&#39;a&#39;|&#39;b&#39;|&#39;c&#39;|...|&#39;x&#39;&gt;
</pre></div>
</div>
(Contributed by Raymond Hettinger in <a class="reference external" href="https://bugs.python.org/issue9826">bpo-9826</a> and <a class="reference external" href="https://bugs.python.org/issue9840">bpo-9840</a>.)
</div>
<div class="section" id="logging">
<h3>logging<a class="headerlink" href="#logging" title="Permalink to this headline">¶</a></h3>
In addition to dictionary-based configuration described above, the
<a class="reference internal" href="../library/logging.html#module-logging" title="logging: Flexible event logging system for applications."><code class="xref py py-mod docutils literal notranslate">logging</code></a> package has many other improvements.
The logging documentation has been augmented by a <a class="reference internal" href="../howto/logging.html#logging-basic-tutorial">basic tutorial</a>, an <a class="reference internal" href="../howto/logging.html#logging-advanced-tutorial">advanced tutorial</a>, and a <a class="reference internal" href="../howto/logging-cookbook.html#logging-cookbook">cookbook</a> of
logging recipes. These documents are the fastest way to learn about logging.
The <a class="reference internal" href="../library/logging.html#logging.basicConfig" title="logging.basicConfig"><code class="xref py py-func docutils literal notranslate">logging.basicConfig()</code></a> set-up function gained a style argument to
support three different types of string formatting. It defaults to “%” for
traditional %-formatting, can be set to “{” for the new <a class="reference internal" href="../library/stdtypes.html#str.format" title="str.format"><code class="xref py py-meth docutils literal notranslate">str.format()</code></a> style, or
can be set to “$” for the shell-style formatting provided by
<a class="reference internal" href="../library/string.html#string.Template" title="string.Template"><code class="xref py py-class docutils literal notranslate">string.Template</code></a>. The following three configurations are equivalent:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; from logging import basicConfig
&gt;&gt;&gt; basicConfig(style=&#39;%&#39;, format=&quot;%(name)s -&gt; %(levelname)s: %(message)s&quot;)
&gt;&gt;&gt; basicConfig(style=&#39;{&#39;, format=&quot;{name} -&gt; {levelname} {message}&quot;)
&gt;&gt;&gt; basicConfig(style=&#39;$&#39;, format=&quot;$name -&gt; $levelname: $message&quot;)
</pre></div>
</div>
If no configuration is set-up before a logging event occurs, there is now a
default configuration using a <a class="reference internal" href="../library/logging.handlers.html#logging.StreamHandler" title="logging.StreamHandler"><code class="xref py py-class docutils literal notranslate">StreamHandler</code></a> directed to
<a class="reference internal" href="../library/sys.html#sys.stderr" title="sys.stderr"><code class="xref py py-attr docutils literal notranslate">sys.stderr</code></a> for events of <code class="docutils literal notranslate">WARNING</code> level or higher. Formerly, an
event occurring before a configuration was set-up would either raise an
exception or silently drop the event depending on the value of
<code class="xref py py-attr docutils literal notranslate">logging.raiseExceptions</code>. The new default handler is stored in
<a class="reference internal" href="../library/logging.html#logging.lastResort" title="logging.lastResort"><code class="xref py py-attr docutils literal notranslate">logging.lastResort</code></a>.
The use of filters has been simplified. Instead of creating a
<a class="reference internal" href="../library/logging.html#logging.Filter" title="logging.Filter"><code class="xref py py-class docutils literal notranslate">Filter</code></a> object, the predicate can be any Python callable that
returns <code class="docutils literal notranslate">True</code> or <code class="docutils literal notranslate">False</code>.
There were a number of other improvements that add flexibility and simplify
configuration. See the module documentation for a full listing of changes in
Python 3.2.
</div>
<div class="section" id="csv">
<h3>csv<a class="headerlink" href="#csv" title="Permalink to this headline">¶</a></h3>
The <a class="reference internal" href="../library/csv.html#module-csv" title="csv: Write and read tabular data to and from delimited files."><code class="xref py py-mod docutils literal notranslate">csv</code></a> module now supports a new dialect, <a class="reference internal" href="../library/csv.html#csv.unix_dialect" title="csv.unix_dialect"><code class="xref py py-class docutils literal notranslate">unix_dialect</code></a>,
which applies quoting for all fields and a traditional Unix style with <code class="docutils literal notranslate">'\n'</code> as
the line terminator. The registered dialect name is <code class="docutils literal notranslate">unix</code>.
The <a class="reference internal" href="../library/csv.html#csv.DictWriter" title="csv.DictWriter"><code class="xref py py-class docutils literal notranslate">csv.DictWriter</code></a> has a new method,
<a class="reference internal" href="../library/csv.html#csv.DictWriter.writeheader" title="csv.DictWriter.writeheader"><code class="xref py py-meth docutils literal notranslate">writeheader()</code></a> for writing-out an initial row to document
the field names:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; import csv, sys
&gt;&gt;&gt; w = csv.DictWriter(sys.stdout, [&#39;name&#39;, &#39;dept&#39;], dialect=&#39;unix&#39;)
&gt;&gt;&gt; w.writeheader()
&quot;name&quot;,&quot;dept&quot;
&gt;&gt;&gt; w.writerows([
... {&#39;name&#39;: &#39;tom&#39;, &#39;dept&#39;: &#39;accounting&#39;},
... {&#39;name&#39;: &#39;susan&#39;, &#39;dept&#39;: &#39;Salesl&#39;}])
&quot;tom&quot;,&quot;accounting&quot;
&quot;susan&quot;,&quot;sales&quot;
</pre></div>
</div>
(New dialect suggested by Jay Talbot in <a class="reference external" href="https://bugs.python.org/issue5975">bpo-5975</a>, and the new method
suggested by Ed Abraham in <a class="reference external" href="https://bugs.python.org/issue1537721">bpo-1537721</a>.)
</div>
<div class="section" id="contextlib">
<h3>contextlib<a class="headerlink" href="#contextlib" title="Permalink to this headline">¶</a></h3>
There is a new and slightly mind-blowing tool
<a class="reference internal" href="../library/contextlib.html#contextlib.ContextDecorator" title="contextlib.ContextDecorator"><code class="xref py py-class docutils literal notranslate">ContextDecorator</code></a> that is helpful for creating a
<a class="reference internal" href="../glossary.html#term-context-manager">context manager</a> that does double duty as a function decorator.
As a convenience, this new functionality is used by
<a class="reference internal" href="../library/contextlib.html#contextlib.contextmanager" title="contextlib.contextmanager"><code class="xref py py-func docutils literal notranslate">contextmanager()</code></a> so that no extra effort is needed to support
both roles.
The basic idea is that both context managers and function decorators can be used
for pre-action and post-action wrappers. Context managers wrap a group of
statements using a <a class="reference internal" href="../reference/compound_stmts.html#with"><code class="xref std std-keyword docutils literal notranslate">with</code></a> statement, and function decorators wrap a
group of statements enclosed in a function. So, occasionally there is a need to
write a pre-action or post-action wrapper that can be used in either role.
For example, it is sometimes useful to wrap functions or groups of statements
with a logger that can track the time of entry and time of exit. Rather than
writing both a function decorator and a context manager for the task, the
<a class="reference internal" href="../library/contextlib.html#contextlib.contextmanager" title="contextlib.contextmanager"><code class="xref py py-func docutils literal notranslate">contextmanager()</code></a> provides both capabilities in a single
definition:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>from contextlib import contextmanager
import logging

logging.basicConfig(level=logging.INFO)

@contextmanager
def track_entry_and_exit(name):
 logging.info(&#39;Entering: %s&#39;, name)
 yield
 logging.info(&#39;Exiting: %s&#39;, name)
</pre></div>
</div>
Formerly, this would have only been usable as a context manager:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>with track_entry_and_exit(&#39;widget loader&#39;):
 print(&#39;Some time consuming activity goes here&#39;)
 load_widget()
</pre></div>
</div>
Now, it can be used as a decorator as well:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>@track_entry_and_exit(&#39;widget loader&#39;)
def activity():
 print(&#39;Some time consuming activity goes here&#39;)
 load_widget()
</pre></div>
</div>
Trying to fulfill two roles at once places some limitations on the technique.
Context managers normally have the flexibility to return an argument usable by
a <a class="reference internal" href="../reference/compound_stmts.html#with"><code class="xref std std-keyword docutils literal notranslate">with</code></a> statement, but there is no parallel for function decorators.
In the above example, there is not a clean way for the track_entry_and_exit
context manager to return a logging instance for use in the body of enclosed
statements.
(Contributed by Michael Foord in <a class="reference external" href="https://bugs.python.org/issue9110">bpo-9110</a>.)
</div>
<div class="section" id="decimal-and-fractions">
<h3>decimal and fractions<a class="headerlink" href="#decimal-and-fractions" title="Permalink to this headline">¶</a></h3>
Mark Dickinson crafted an elegant and efficient scheme for assuring that
different numeric datatypes will have the same hash value whenever their actual
values are equal (<a class="reference external" href="https://bugs.python.org/issue8188">bpo-8188</a>):
<div class="highlight-python3 notranslate"><div class="highlight"><pre>assert hash(Fraction(3, 2)) == hash(1.5) == \
 hash(Decimal(&quot;1.5&quot;)) == hash(complex(1.5, 0))
</pre></div>
</div>
Some of the hashing details are exposed through a new attribute,
<a class="reference internal" href="../library/sys.html#sys.hash_info" title="sys.hash_info"><code class="xref py py-attr docutils literal notranslate">sys.hash_info</code></a>, which describes the bit width of the hash value, the
prime modulus, the hash values for infinity and nan, and the multiplier
used for the imaginary part of a number:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; sys.hash_info 
sys.hash_info(width=64, modulus=2305843009213693951, inf=314159, nan=0, imag=1000003)
</pre></div>
</div>
An early decision to limit the inter-operability of various numeric types has
been relaxed. It is still unsupported (and ill-advised) to have implicit
mixing in arithmetic expressions such as <code class="docutils literal notranslate">Decimal('1.1') + float('1.1')</code>
because the latter loses information in the process of constructing the binary
float. However, since existing floating point value can be converted losslessly
to either a decimal or rational representation, it makes sense to add them to
the constructor and to support mixed-type comparisons.
<ul class="simple">
<li>The <a class="reference internal" href="../library/decimal.html#decimal.Decimal" title="decimal.Decimal"><code class="xref py py-class docutils literal notranslate">decimal.Decimal</code></a> constructor now accepts <a class="reference internal" href="../library/functions.html#float" title="float"><code class="xref py py-class docutils literal notranslate">float</code></a> objects
directly so there in no longer a need to use the <a class="reference internal" href="../library/decimal.html#decimal.Decimal.from_float" title="decimal.Decimal.from_float"><code class="xref py py-meth docutils literal notranslate">from_float()</code></a>
method (<a class="reference external" href="https://bugs.python.org/issue8257">bpo-8257</a>).</li>
<li>Mixed type comparisons are now fully supported so that
<a class="reference internal" href="../library/decimal.html#decimal.Decimal" title="decimal.Decimal"><code class="xref py py-class docutils literal notranslate">Decimal</code></a> objects can be directly compared with <a class="reference internal" href="../library/functions.html#float" title="float"><code class="xref py py-class docutils literal notranslate">float</code></a>
and <a class="reference internal" href="../library/fractions.html#fractions.Fraction" title="fractions.Fraction"><code class="xref py py-class docutils literal notranslate">fractions.Fraction</code></a> (<a class="reference external" href="https://bugs.python.org/issue2531">bpo-2531</a> and <a class="reference external" href="https://bugs.python.org/issue8188">bpo-8188</a>).</li>
</ul>
Similar changes were made to <a class="reference internal" href="../library/fractions.html#fractions.Fraction" title="fractions.Fraction"><code class="xref py py-class docutils literal notranslate">fractions.Fraction</code></a> so that the
<a class="reference internal" href="../library/fractions.html#fractions.Fraction.from_float" title="fractions.Fraction.from_float"><code class="xref py py-meth docutils literal notranslate">from_float()</code></a> and <a class="reference internal" href="../library/fractions.html#fractions.Fraction.from_decimal" title="fractions.Fraction.from_decimal"><code class="xref py py-meth docutils literal notranslate">from_decimal()</code></a>
methods are no longer needed (<a class="reference external" href="https://bugs.python.org/issue8294">bpo-8294</a>):
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; from decimal import Decimal
&gt;&gt;&gt; from fractions import Fraction
&gt;&gt;&gt; Decimal(1.1)
Decimal(&#39;1.100000000000000088817841970012523233890533447265625&#39;)
&gt;&gt;&gt; Fraction(1.1)
Fraction(2476979795053773, 2251799813685248)
</pre></div>
</div>
Another useful change for the <a class="reference internal" href="../library/decimal.html#module-decimal" title="decimal: Implementation of the General Decimal Arithmetic Specification."><code class="xref py py-mod docutils literal notranslate">decimal</code></a> module is that the
<code class="xref py py-attr docutils literal notranslate">Context.clamp</code> attribute is now public. This is useful in creating
contexts that correspond to the decimal interchange formats specified in IEEE
754 (see <a class="reference external" href="https://bugs.python.org/issue8540">bpo-8540</a>).
(Contributed by Mark Dickinson and Raymond Hettinger.)
</div>
<div class="section" id="ftp">
<h3>ftp<a class="headerlink" href="#ftp" title="Permalink to this headline">¶</a></h3>
The <a class="reference internal" href="../library/ftplib.html#ftplib.FTP" title="ftplib.FTP"><code class="xref py py-class docutils literal notranslate">ftplib.FTP</code></a> class now supports the context management protocol to
unconditionally consume <a class="reference internal" href="../library/socket.html#socket.error" title="socket.error"><code class="xref py py-exc docutils literal notranslate">socket.error</code></a> exceptions and to close the FTP
connection when done:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; from ftplib import FTP
&gt;&gt;&gt; with FTP(&quot;ftp1.at.proftpd.org&quot;) as ftp:
 ftp.login()
 ftp.dir()

&#39;230 Anonymous login ok, restrictions apply.&#39;
dr-xr-xr-x 9 ftp ftp 154 May 6 10:43 .
dr-xr-xr-x 9 ftp ftp 154 May 6 10:43 ..
dr-xr-xr-x 5 ftp ftp 4096 May 6 10:43 CentOS
dr-xr-xr-x 3 ftp ftp 18 Jul 10 2008 Fedora
</pre></div>
</div>
Other file-like objects such as <a class="reference internal" href="../library/mmap.html#mmap.mmap" title="mmap.mmap"><code class="xref py py-class docutils literal notranslate">mmap.mmap</code></a> and <a class="reference internal" href="../library/fileinput.html#fileinput.input" title="fileinput.input"><code class="xref py py-func docutils literal notranslate">fileinput.input()</code></a>
also grew auto-closing context managers:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>with fileinput.input(files=(&#39;log1.txt&#39;, &#39;log2.txt&#39;)) as f:
 for line in f:
 process(line)
</pre></div>
</div>
(Contributed by Tarek Ziadé and Giampaolo Rodolà in <a class="reference external" href="https://bugs.python.org/issue4972">bpo-4972</a>, and
by Georg Brandl in <a class="reference external" href="https://bugs.python.org/issue8046">bpo-8046</a> and <a class="reference external" href="https://bugs.python.org/issue1286">bpo-1286</a>.)
The <a class="reference internal" href="../library/ftplib.html#ftplib.FTP_TLS" title="ftplib.FTP_TLS"><code class="xref py py-class docutils literal notranslate">FTP_TLS</code></a> class now accepts a context parameter, which is a
<a class="reference internal" href="../library/ssl.html#ssl.SSLContext" title="ssl.SSLContext"><code class="xref py py-class docutils literal notranslate">ssl.SSLContext</code></a> object allowing bundling SSL configuration options,
certificates and private keys into a single (potentially long-lived) structure.
(Contributed by Giampaolo Rodolà; <a class="reference external" href="https://bugs.python.org/issue8806">bpo-8806</a>.)
</div>
<div class="section" id="popen">
<h3>popen<a class="headerlink" href="#popen" title="Permalink to this headline">¶</a></h3>
The <a class="reference internal" href="../library/os.html#os.popen" title="os.popen"><code class="xref py py-func docutils literal notranslate">os.popen()</code></a> and <a class="reference internal" href="../library/subprocess.html#subprocess.Popen" title="subprocess.Popen"><code class="xref py py-func docutils literal notranslate">subprocess.Popen()</code></a> functions now support
<a class="reference internal" href="../reference/compound_stmts.html#with"><code class="xref std std-keyword docutils literal notranslate">with</code></a> statements for auto-closing of the file descriptors.
(Contributed by Antoine Pitrou and Brian Curtin in <a class="reference external" href="https://bugs.python.org/issue7461">bpo-7461</a> and
<a class="reference external" href="https://bugs.python.org/issue10554">bpo-10554</a>.)
</div>
<div class="section" id="select">
<h3>select<a class="headerlink" href="#select" title="Permalink to this headline">¶</a></h3>
The <a class="reference internal" href="../library/select.html#module-select" title="select: Wait for I/O completion on multiple streams."><code class="xref py py-mod docutils literal notranslate">select</code></a> module now exposes a new, constant attribute,
<a class="reference internal" href="../library/select.html#select.PIPE_BUF" title="select.PIPE_BUF"><code class="xref py py-attr docutils literal notranslate">PIPE_BUF</code></a>, which gives the minimum number of bytes which are
guaranteed not to block when <a class="reference internal" href="../library/select.html#select.select" title="select.select"><code class="xref py py-func docutils literal notranslate">select.select()</code></a> says a pipe is ready
for writing.
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; import select
&gt;&gt;&gt; select.PIPE_BUF
512
</pre></div>
</div>
(Available on Unix systems. Patch by Sébastien Sablé in <a class="reference external" href="https://bugs.python.org/issue9862">bpo-9862</a>)
</div>
<div class="section" id="gzip-and-zipfile">
<h3>gzip and zipfile<a class="headerlink" href="#gzip-and-zipfile" title="Permalink to this headline">¶</a></h3>
<a class="reference internal" href="../library/gzip.html#gzip.GzipFile" title="gzip.GzipFile"><code class="xref py py-class docutils literal notranslate">gzip.GzipFile</code></a> now implements the <a class="reference internal" href="../library/io.html#io.BufferedIOBase" title="io.BufferedIOBase"><code class="xref py py-class docutils literal notranslate">io.BufferedIOBase</code></a>
<a class="reference internal" href="../glossary.html#term-abstract-base-class">abstract base class</a> (except for <code class="docutils literal notranslate">truncate()</code>). It also has a
<a class="reference internal" href="../library/gzip.html#gzip.GzipFile.peek" title="gzip.GzipFile.peek"><code class="xref py py-meth docutils literal notranslate">peek()</code></a> method and supports unseekable as well as
zero-padded file objects.
The <a class="reference internal" href="../library/gzip.html#module-gzip" title="gzip: Interfaces for gzip compression and decompression using file objects."><code class="xref py py-mod docutils literal notranslate">gzip</code></a> module also gains the <a class="reference internal" href="../library/gzip.html#gzip.compress" title="gzip.compress"><code class="xref py py-func docutils literal notranslate">compress()</code></a> and
<a class="reference internal" href="../library/gzip.html#gzip.decompress" title="gzip.decompress"><code class="xref py py-func docutils literal notranslate">decompress()</code></a> functions for easier in-memory compression and
decompression. Keep in mind that text needs to be encoded as <a class="reference internal" href="../library/stdtypes.html#bytes" title="bytes"><code class="xref py py-class docutils literal notranslate">bytes</code></a>
before compressing and decompressing:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; import gzip
&gt;&gt;&gt; s = &#39;Three shall be the number thou shalt count, &#39;
&gt;&gt;&gt; s += &#39;and the number of the counting shall be three&#39;
&gt;&gt;&gt; b = s.encode() # convert to utf-8
&gt;&gt;&gt; len(b)
89
&gt;&gt;&gt; c = gzip.compress(b)
&gt;&gt;&gt; len(c)
77
&gt;&gt;&gt; gzip.decompress(c).decode()[:42] # decompress and convert to text
&#39;Three shall be the number thou shalt count&#39;
</pre></div>
</div>
(Contributed by Anand B. Pillai in <a class="reference external" href="https://bugs.python.org/issue3488">bpo-3488</a>; and by Antoine Pitrou, Nir
Aides and Brian Curtin in <a class="reference external" href="https://bugs.python.org/issue9962">bpo-9962</a>, <a class="reference external" href="https://bugs.python.org/issue1675951">bpo-1675951</a>, <a class="reference external" href="https://bugs.python.org/issue7471">bpo-7471</a> and
<a class="reference external" href="https://bugs.python.org/issue2846">bpo-2846</a>.)
Also, the <code class="xref py py-class docutils literal notranslate">zipfile.ZipExtFile</code> class was reworked internally to represent
files stored inside an archive. The new implementation is significantly faster
and can be wrapped in an <a class="reference internal" href="../library/io.html#io.BufferedReader" title="io.BufferedReader"><code class="xref py py-class docutils literal notranslate">io.BufferedReader</code></a> object for more speedups. It
also solves an issue where interleaved calls to read and readline gave the
wrong results.
(Patch submitted by Nir Aides in <a class="reference external" href="https://bugs.python.org/issue7610">bpo-7610</a>.)
</div>
<div class="section" id="tarfile">
<h3>tarfile<a class="headerlink" href="#tarfile" title="Permalink to this headline">¶</a></h3>
The <a class="reference internal" href="../library/tarfile.html#tarfile.TarFile" title="tarfile.TarFile"><code class="xref py py-class docutils literal notranslate">TarFile</code></a> class can now be used as a context manager. In
addition, its <a class="reference internal" href="../library/tarfile.html#tarfile.TarFile.add" title="tarfile.TarFile.add"><code class="xref py py-meth docutils literal notranslate">add()</code></a> method has a new option, filter,
that controls which files are added to the archive and allows the file metadata
to be edited.
The new filter option replaces the older, less flexible exclude parameter
which is now deprecated. If specified, the optional filter parameter needs to
be a <a class="reference internal" href="../glossary.html#term-keyword-argument">keyword argument</a>. The user-supplied filter function accepts a
<a class="reference internal" href="../library/tarfile.html#tarfile.TarInfo" title="tarfile.TarInfo"><code class="xref py py-class docutils literal notranslate">TarInfo</code></a> object and returns an updated
<a class="reference internal" href="../library/tarfile.html#tarfile.TarInfo" title="tarfile.TarInfo"><code class="xref py py-class docutils literal notranslate">TarInfo</code></a> object, or if it wants the file to be excluded, the
function can return <code class="docutils literal notranslate">None</code>:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; import tarfile, glob

&gt;&gt;&gt; def myfilter(tarinfo):
... if tarinfo.isfile(): # only save real files
... tarinfo.uname = &#39;monty&#39; # redact the user name
... return tarinfo

&gt;&gt;&gt; with tarfile.open(name=&#39;myarchive.tar.gz&#39;, mode=&#39;w:gz&#39;) as tf:
... for filename in glob.glob(&#39;*.txt&#39;):
... tf.add(filename, filter=myfilter)
... tf.list()
-rw-r--r-- monty/501 902 2011-01-26 17:59:11 annotations.txt
-rw-r--r-- monty/501 123 2011-01-26 17:59:11 general_questions.txt
-rw-r--r-- monty/501 3514 2011-01-26 17:59:11 prion.txt
-rw-r--r-- monty/501 124 2011-01-26 17:59:11 py_todo.txt
-rw-r--r-- monty/501 1399 2011-01-26 17:59:11 semaphore_notes.txt
</pre></div>
</div>
(Proposed by Tarek Ziadé and implemented by Lars Gustäbel in <a class="reference external" href="https://bugs.python.org/issue6856">bpo-6856</a>.)
</div>
<div class="section" id="hashlib">
<h3>hashlib<a class="headerlink" href="#hashlib" title="Permalink to this headline">¶</a></h3>
The <a class="reference internal" href="../library/hashlib.html#module-hashlib" title="hashlib: Secure hash and message digest algorithms."><code class="xref py py-mod docutils literal notranslate">hashlib</code></a> module has two new constant attributes listing the hashing
algorithms guaranteed to be present in all implementations and those available
on the current implementation:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; import hashlib

&gt;&gt;&gt; hashlib.algorithms_guaranteed
{&#39;sha1&#39;, &#39;sha224&#39;, &#39;sha384&#39;, &#39;sha256&#39;, &#39;sha512&#39;, &#39;md5&#39;}

&gt;&gt;&gt; hashlib.algorithms_available
{&#39;md2&#39;, &#39;SHA256&#39;, &#39;SHA512&#39;, &#39;dsaWithSHA&#39;, &#39;mdc2&#39;, &#39;SHA224&#39;, &#39;MD4&#39;, &#39;sha256&#39;,
&#39;sha512&#39;, &#39;ripemd160&#39;, &#39;SHA1&#39;, &#39;MDC2&#39;, &#39;SHA&#39;, &#39;SHA384&#39;, &#39;MD2&#39;,
&#39;ecdsa-with-SHA1&#39;,&#39;md4&#39;, &#39;md5&#39;, &#39;sha1&#39;, &#39;DSA-SHA&#39;, &#39;sha224&#39;,
&#39;dsaEncryption&#39;, &#39;DSA&#39;, &#39;RIPEMD160&#39;, &#39;sha&#39;, &#39;MD5&#39;, &#39;sha384&#39;}
</pre></div>
</div>
(Suggested by Carl Chenet in <a class="reference external" href="https://bugs.python.org/issue7418">bpo-7418</a>.)
</div>
<div class="section" id="ast">
<h3>ast<a class="headerlink" href="#ast" title="Permalink to this headline">¶</a></h3>
The <a class="reference internal" href="../library/ast.html#module-ast" title="ast: Abstract Syntax Tree classes and manipulation."><code class="xref py py-mod docutils literal notranslate">ast</code></a> module has a wonderful a general-purpose tool for safely
evaluating expression strings using the Python literal
syntax. The <a class="reference internal" href="../library/ast.html#ast.literal_eval" title="ast.literal_eval"><code class="xref py py-func docutils literal notranslate">ast.literal_eval()</code></a> function serves as a secure alternative to
the builtin <a class="reference internal" href="../library/functions.html#eval" title="eval"><code class="xref py py-func docutils literal notranslate">eval()</code></a> function which is easily abused. Python 3.2 adds
<a class="reference internal" href="../library/stdtypes.html#bytes" title="bytes"><code class="xref py py-class docutils literal notranslate">bytes</code></a> and <a class="reference internal" href="../library/stdtypes.html#set" title="set"><code class="xref py py-class docutils literal notranslate">set</code></a> literals to the list of supported types:
strings, bytes, numbers, tuples, lists, dicts, sets, booleans, and <code class="docutils literal notranslate">None</code>.
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; from ast import literal_eval

&gt;&gt;&gt; request = &quot;{&#39;req&#39;: 3, &#39;func&#39;: &#39;pow&#39;, &#39;args&#39;: (2, 0.5)}&quot;
&gt;&gt;&gt; literal_eval(request)
{&#39;args&#39;: (2, 0.5), &#39;req&#39;: 3, &#39;func&#39;: &#39;pow&#39;}

&gt;&gt;&gt; request = &quot;os.system(&#39;do something harmful&#39;)&quot;
&gt;&gt;&gt; literal_eval(request)
Traceback (most recent call last):
 ...
ValueError: malformed node or string: &lt;_ast.Call object at 0x101739a10&gt;
</pre></div>
</div>
(Implemented by Benjamin Peterson and Georg Brandl.)
</div>
<div class="section" id="os">
<h3>os<a class="headerlink" href="#os" title="Permalink to this headline">¶</a></h3>
Different operating systems use various encodings for filenames and environment
variables. The <a class="reference internal" href="../library/os.html#module-os" title="os: Miscellaneous operating system interfaces."><code class="xref py py-mod docutils literal notranslate">os</code></a> module provides two new functions,
<a class="reference internal" href="../library/os.html#os.fsencode" title="os.fsencode"><code class="xref py py-func docutils literal notranslate">fsencode()</code></a> and <a class="reference internal" href="../library/os.html#os.fsdecode" title="os.fsdecode"><code class="xref py py-func docutils literal notranslate">fsdecode()</code></a>, for encoding and decoding
filenames:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; import os
&gt;&gt;&gt; filename = &#39;Sehenswürdigkeiten&#39;
&gt;&gt;&gt; os.fsencode(filename)
b&#39;Sehensw\xc3\xbcrdigkeiten&#39;
</pre></div>
</div>
Some operating systems allow direct access to encoded bytes in the
environment. If so, the <a class="reference internal" href="../library/os.html#os.supports_bytes_environ" title="os.supports_bytes_environ"><code class="xref py py-attr docutils literal notranslate">os.supports_bytes_environ</code></a> constant will be
true.
For direct access to encoded environment variables (if available),
use the new <a class="reference internal" href="../library/os.html#os.getenvb" title="os.getenvb"><code class="xref py py-func docutils literal notranslate">os.getenvb()</code></a> function or use <a class="reference internal" href="../library/os.html#os.environb" title="os.environb"><code class="xref py py-data docutils literal notranslate">os.environb</code></a>
which is a bytes version of <a class="reference internal" href="../library/os.html#os.environ" title="os.environ"><code class="xref py py-data docutils literal notranslate">os.environ</code></a>.
(Contributed by Victor Stinner.)
</div>
<div class="section" id="shutil">
<h3>shutil<a class="headerlink" href="#shutil" title="Permalink to this headline">¶</a></h3>
The <a class="reference internal" href="../library/shutil.html#shutil.copytree" title="shutil.copytree"><code class="xref py py-func docutils literal notranslate">shutil.copytree()</code></a> function has two new options:
<ul class="simple">
<li>ignore_dangling_symlinks: when <code class="docutils literal notranslate">symlinks=False</code> so that the function
copies a file pointed to by a symlink, not the symlink itself. This option
will silence the error raised if the file doesn’t exist.</li>
<li>copy_function: is a callable that will be used to copy files.
<a class="reference internal" href="../library/shutil.html#shutil.copy2" title="shutil.copy2"><code class="xref py py-func docutils literal notranslate">shutil.copy2()</code></a> is used by default.</li>
</ul>
(Contributed by Tarek Ziadé.)
In addition, the <a class="reference internal" href="../library/shutil.html#module-shutil" title="shutil: High-level file operations, including copying."><code class="xref py py-mod docutils literal notranslate">shutil</code></a> module now supports <a class="reference internal" href="../library/shutil.html#archiving-operations">archiving operations</a> for zipfiles, uncompressed tarfiles, gzipped tarfiles,
and bzipped tarfiles. And there are functions for registering additional
archiving file formats (such as xz compressed tarfiles or custom formats).
The principal functions are <a class="reference internal" href="../library/shutil.html#shutil.make_archive" title="shutil.make_archive"><code class="xref py py-func docutils literal notranslate">make_archive()</code></a> and
<a class="reference internal" href="../library/shutil.html#shutil.unpack_archive" title="shutil.unpack_archive"><code class="xref py py-func docutils literal notranslate">unpack_archive()</code></a>. By default, both operate on the current
directory (which can be set by <a class="reference internal" href="../library/os.html#os.chdir" title="os.chdir"><code class="xref py py-func docutils literal notranslate">os.chdir()</code></a>) and on any sub-directories.
The archive filename needs to be specified with a full pathname. The archiving
step is non-destructive (the original files are left unchanged).
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; import shutil, pprint

&gt;&gt;&gt; os.chdir(&#39;mydata&#39;) # change to the source directory
&gt;&gt;&gt; f = shutil.make_archive(&#39;/var/backup/mydata&#39;,
... &#39;zip&#39;) # archive the current directory
&gt;&gt;&gt; f # show the name of archive
&#39;/var/backup/mydata.zip&#39;
&gt;&gt;&gt; os.chdir(&#39;tmp&#39;) # change to an unpacking
&gt;&gt;&gt; shutil.unpack_archive(&#39;/var/backup/mydata.zip&#39;) # recover the data

&gt;&gt;&gt; pprint.pprint(shutil.get_archive_formats()) # display known formats
[(&#39;bztar&#39;, &quot;bzip2&#39;ed tar-file&quot;),
 (&#39;gztar&#39;, &quot;gzip&#39;ed tar-file&quot;),
 (&#39;tar&#39;, &#39;uncompressed tar file&#39;),
 (&#39;zip&#39;, &#39;ZIP file&#39;)]

&gt;&gt;&gt; shutil.register_archive_format( # register a new archive format
... name=&#39;xz&#39;,
... function=xz.compress, # callable archiving function
... extra_args=[(&#39;level&#39;, 8)], # arguments to the function
... description=&#39;xz compression&#39;
... )
</pre></div>
</div>
(Contributed by Tarek Ziadé.)
</div>
<div class="section" id="sqlite3">
<h3>sqlite3<a class="headerlink" href="#sqlite3" title="Permalink to this headline">¶</a></h3>
The <a class="reference internal" href="../library/sqlite3.html#module-sqlite3" title="sqlite3: A DB-API 2.0 implementation using SQLite 3.x."><code class="xref py py-mod docutils literal notranslate">sqlite3</code></a> module was updated to pysqlite version 2.6.0. It has two new capabilities.
<ul class="simple">
<li>The <code class="xref py py-attr docutils literal notranslate">sqlite3.Connection.in_transit</code> attribute is true if there is an
active transaction for uncommitted changes.</li>
<li>The <a class="reference internal" href="../library/sqlite3.html#sqlite3.Connection.enable_load_extension" title="sqlite3.Connection.enable_load_extension"><code class="xref py py-meth docutils literal notranslate">sqlite3.Connection.enable_load_extension()</code></a> and
<a class="reference internal" href="../library/sqlite3.html#sqlite3.Connection.load_extension" title="sqlite3.Connection.load_extension"><code class="xref py py-meth docutils literal notranslate">sqlite3.Connection.load_extension()</code></a> methods allows you to load SQLite
extensions from “.so” files. One well-known extension is the fulltext-search
extension distributed with SQLite.</li>
</ul>
(Contributed by R. David Murray and Shashwat Anand; <a class="reference external" href="https://bugs.python.org/issue8845">bpo-8845</a>.)
</div>
<div class="section" id="html">
<h3>html<a class="headerlink" href="#html" title="Permalink to this headline">¶</a></h3>
A new <a class="reference internal" href="../library/html.html#module-html" title="html: Helpers for manipulating HTML."><code class="xref py py-mod docutils literal notranslate">html</code></a> module was introduced with only a single function,
<a class="reference internal" href="../library/html.html#html.escape" title="html.escape"><code class="xref py py-func docutils literal notranslate">escape()</code></a>, which is used for escaping reserved characters from HTML
markup:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; import html
&gt;&gt;&gt; html.escape(&#39;x &gt; 2 &amp;&amp; x &lt; 7&#39;)
&#39;x &amp;gt; 2 &amp;amp;&amp;amp; x &amp;lt; 7&#39;
</pre></div>
</div>
</div>
<div class="section" id="socket">
<h3>socket<a class="headerlink" href="#socket" title="Permalink to this headline">¶</a></h3>
The <a class="reference internal" href="../library/socket.html#module-socket" title="socket: Low-level networking interface."><code class="xref py py-mod docutils literal notranslate">socket</code></a> module has two new improvements.
<ul class="simple">
<li>Socket objects now have a <a class="reference internal" href="../library/socket.html#socket.socket.detach" title="socket.socket.detach"><code class="xref py py-meth docutils literal notranslate">detach()</code></a> method which puts
the socket into closed state without actually closing the underlying file
descriptor. The latter can then be reused for other purposes.
(Added by Antoine Pitrou; <a class="reference external" href="https://bugs.python.org/issue8524">bpo-8524</a>.)</li>
<li><a class="reference internal" href="../library/socket.html#socket.create_connection" title="socket.create_connection"><code class="xref py py-func docutils literal notranslate">socket.create_connection()</code></a> now supports the context management protocol
to unconditionally consume <a class="reference internal" href="../library/socket.html#socket.error" title="socket.error"><code class="xref py py-exc docutils literal notranslate">socket.error</code></a> exceptions and to close the
socket when done.
(Contributed by Giampaolo Rodolà; <a class="reference external" href="https://bugs.python.org/issue9794">bpo-9794</a>.)</li>
</ul>
</div>
<div class="section" id="ssl">
<h3>ssl<a class="headerlink" href="#ssl" title="Permalink to this headline">¶</a></h3>
The <a class="reference internal" href="../library/ssl.html#module-ssl" title="ssl: TLS/SSL wrapper for socket objects"><code class="xref py py-mod docutils literal notranslate">ssl</code></a> module added a number of features to satisfy common requirements
for secure (encrypted, authenticated) internet connections:
<ul class="simple">
<li>A new class, <a class="reference internal" href="../library/ssl.html#ssl.SSLContext" title="ssl.SSLContext"><code class="xref py py-class docutils literal notranslate">SSLContext</code></a>, serves as a container for persistent
SSL data, such as protocol settings, certificates, private keys, and various
other options. It includes a <a class="reference internal" href="../library/ssl.html#ssl.SSLContext.wrap_socket" title="ssl.SSLContext.wrap_socket"><code class="xref py py-meth docutils literal notranslate">wrap_socket()</code></a> for creating
an SSL socket from an SSL context.</li>
<li>A new function, <a class="reference internal" href="../library/ssl.html#ssl.match_hostname" title="ssl.match_hostname"><code class="xref py py-func docutils literal notranslate">ssl.match_hostname()</code></a>, supports server identity
verification for higher-level protocols by implementing the rules of HTTPS
(from <a class="rfc reference external" href="https://tools.ietf.org/html/rfc2818.html">RFC 2818</a>) which are also suitable for other protocols.</li>
<li>The <a class="reference internal" href="../library/ssl.html#ssl.wrap_socket" title="ssl.wrap_socket"><code class="xref py py-func docutils literal notranslate">ssl.wrap_socket()</code></a> constructor function now takes a ciphers
argument. The ciphers string lists the allowed encryption algorithms using
the format described in the <a class="reference external" href="https://www.openssl.org/docs/apps/ciphers.html#CIPHER-LIST-FORMAT">OpenSSL documentation</a>.</li>
<li>When linked against recent versions of OpenSSL, the <a class="reference internal" href="../library/ssl.html#module-ssl" title="ssl: TLS/SSL wrapper for socket objects"><code class="xref py py-mod docutils literal notranslate">ssl</code></a> module now
supports the Server Name Indication extension to the TLS protocol, allowing
multiple “virtual hosts” using different certificates on a single IP port.
This extension is only supported in client mode, and is activated by passing
the server_hostname argument to <a class="reference internal" href="../library/ssl.html#ssl.SSLContext.wrap_socket" title="ssl.SSLContext.wrap_socket"><code class="xref py py-meth docutils literal notranslate">ssl.SSLContext.wrap_socket()</code></a>.</li>
<li>Various options have been added to the <a class="reference internal" href="../library/ssl.html#module-ssl" title="ssl: TLS/SSL wrapper for socket objects"><code class="xref py py-mod docutils literal notranslate">ssl</code></a> module, such as
<a class="reference internal" href="../library/ssl.html#ssl.OP_NO_SSLv2" title="ssl.OP_NO_SSLv2"><code class="xref py py-data docutils literal notranslate">OP_NO_SSLv2</code></a> which disables the insecure and obsolete SSLv2
protocol.</li>
<li>The extension now loads all the OpenSSL ciphers and digest algorithms. If
some SSL certificates cannot be verified, they are reported as an “unknown
algorithm” error.</li>
<li>The version of OpenSSL being used is now accessible using the module
attributes <a class="reference internal" href="../library/ssl.html#ssl.OPENSSL_VERSION" title="ssl.OPENSSL_VERSION"><code class="xref py py-data docutils literal notranslate">ssl.OPENSSL_VERSION</code></a> (a string),
<a class="reference internal" href="../library/ssl.html#ssl.OPENSSL_VERSION_INFO" title="ssl.OPENSSL_VERSION_INFO"><code class="xref py py-data docutils literal notranslate">ssl.OPENSSL_VERSION_INFO</code></a> (a 5-tuple), and
<a class="reference internal" href="../library/ssl.html#ssl.OPENSSL_VERSION_NUMBER" title="ssl.OPENSSL_VERSION_NUMBER"><code class="xref py py-data docutils literal notranslate">ssl.OPENSSL_VERSION_NUMBER</code></a> (an integer).</li>
</ul>
(Contributed by Antoine Pitrou in <a class="reference external" href="https://bugs.python.org/issue8850">bpo-8850</a>, <a class="reference external" href="https://bugs.python.org/issue1589">bpo-1589</a>, <a class="reference external" href="https://bugs.python.org/issue8322">bpo-8322</a>,
<a class="reference external" href="https://bugs.python.org/issue5639">bpo-5639</a>, <a class="reference external" href="https://bugs.python.org/issue4870">bpo-4870</a>, <a class="reference external" href="https://bugs.python.org/issue8484">bpo-8484</a>, and <a class="reference external" href="https://bugs.python.org/issue8321">bpo-8321</a>.)
</div>
<div class="section" id="nntp">
<h3>nntp<a class="headerlink" href="#nntp" title="Permalink to this headline">¶</a></h3>
The <a class="reference internal" href="../library/nntplib.html#module-nntplib" title="nntplib: NNTP protocol client (requires sockets)."><code class="xref py py-mod docutils literal notranslate">nntplib</code></a> module has a revamped implementation with better bytes and
text semantics as well as more practical APIs. These improvements break
compatibility with the nntplib version in Python 3.1, which was partly
dysfunctional in itself.
Support for secure connections through both implicit (using
<a class="reference internal" href="../library/nntplib.html#nntplib.NNTP_SSL" title="nntplib.NNTP_SSL"><code class="xref py py-class docutils literal notranslate">nntplib.NNTP_SSL</code></a>) and explicit (using <a class="reference internal" href="../library/nntplib.html#nntplib.NNTP.starttls" title="nntplib.NNTP.starttls"><code class="xref py py-meth docutils literal notranslate">nntplib.NNTP.starttls()</code></a>)
TLS has also been added.
(Contributed by Antoine Pitrou in <a class="reference external" href="https://bugs.python.org/issue9360">bpo-9360</a> and Andrew Vant in <a class="reference external" href="https://bugs.python.org/issue1926">bpo-1926</a>.)
</div>
<div class="section" id="certificates">
<h3>certificates<a class="headerlink" href="#certificates" title="Permalink to this headline">¶</a></h3>
<a class="reference internal" href="../library/http.client.html#http.client.HTTPSConnection" title="http.client.HTTPSConnection"><code class="xref py py-class docutils literal notranslate">http.client.HTTPSConnection</code></a>, <a class="reference internal" href="../library/urllib.request.html#urllib.request.HTTPSHandler" title="urllib.request.HTTPSHandler"><code class="xref py py-class docutils literal notranslate">urllib.request.HTTPSHandler</code></a>
and <a class="reference internal" href="../library/urllib.request.html#urllib.request.urlopen" title="urllib.request.urlopen"><code class="xref py py-func docutils literal notranslate">urllib.request.urlopen()</code></a> now take optional arguments to allow for
server certificate checking against a set of Certificate Authorities,
as recommended in public uses of HTTPS.
(Added by Antoine Pitrou, <a class="reference external" href="https://bugs.python.org/issue9003">bpo-9003</a>.)
</div>
<div class="section" id="imaplib">
<h3>imaplib<a class="headerlink" href="#imaplib" title="Permalink to this headline">¶</a></h3>
Support for explicit TLS on standard IMAP4 connections has been added through
the new <a class="reference internal" href="../library/imaplib.html#imaplib.IMAP4.starttls" title="imaplib.IMAP4.starttls"><code class="xref py py-mod docutils literal notranslate">imaplib.IMAP4.starttls</code></a> method.
(Contributed by Lorenzo M. Catucci and Antoine Pitrou, <a class="reference external" href="https://bugs.python.org/issue4471">bpo-4471</a>.)
</div>
<div class="section" id="http-client">
<h3>http.client<a class="headerlink" href="#http-client" title="Permalink to this headline">¶</a></h3>
There were a number of small API improvements in the <a class="reference internal" href="../library/http.client.html#module-http.client" title="http.client: HTTP and HTTPS protocol client (requires sockets)."><code class="xref py py-mod docutils literal notranslate">http.client</code></a> module.
The old-style HTTP 0.9 simple responses are no longer supported and the strict
parameter is deprecated in all classes.
The <a class="reference internal" href="../library/http.client.html#http.client.HTTPConnection" title="http.client.HTTPConnection"><code class="xref py py-class docutils literal notranslate">HTTPConnection</code></a> and
<a class="reference internal" href="../library/http.client.html#http.client.HTTPSConnection" title="http.client.HTTPSConnection"><code class="xref py py-class docutils literal notranslate">HTTPSConnection</code></a> classes now have a source_address
parameter for a (host, port) tuple indicating where the HTTP connection is made
from.
Support for certificate checking and HTTPS virtual hosts were added to
<a class="reference internal" href="../library/http.client.html#http.client.HTTPSConnection" title="http.client.HTTPSConnection"><code class="xref py py-class docutils literal notranslate">HTTPSConnection</code></a>.
The <a class="reference internal" href="../library/http.client.html#http.client.HTTPConnection.request" title="http.client.HTTPConnection.request"><code class="xref py py-meth docutils literal notranslate">request()</code></a> method on connection objects
allowed an optional body argument so that a <a class="reference internal" href="../glossary.html#term-file-object">file object</a> could be used
to supply the content of the request. Conveniently, the body argument now
also accepts an <a class="reference internal" href="../glossary.html#term-iterable">iterable</a> object so long as it includes an explicit
<code class="docutils literal notranslate">Content-Length</code> header. This extended interface is much more flexible than
before.
To establish an HTTPS connection through a proxy server, there is a new
<a class="reference internal" href="../library/http.client.html#http.client.HTTPConnection.set_tunnel" title="http.client.HTTPConnection.set_tunnel"><code class="xref py py-meth docutils literal notranslate">set_tunnel()</code></a> method that sets the host and
port for HTTP Connect tunneling.
To match the behavior of <a class="reference internal" href="../library/http.server.html#module-http.server" title="http.server: HTTP server and request handlers."><code class="xref py py-mod docutils literal notranslate">http.server</code></a>, the HTTP client library now also
encodes headers with ISO-8859-1 (Latin-1) encoding. It was already doing that
for incoming headers, so now the behavior is consistent for both incoming and
outgoing traffic. (See work by Armin Ronacher in <a class="reference external" href="https://bugs.python.org/issue10980">bpo-10980</a>.)
</div>
<div class="section" id="unittest">
<h3>unittest<a class="headerlink" href="#unittest" title="Permalink to this headline">¶</a></h3>
The unittest module has a number of improvements supporting test discovery for
packages, easier experimentation at the interactive prompt, new testcase
methods, improved diagnostic messages for test failures, and better method
names.
<ul>
<li>The command-line call <code class="docutils literal notranslate">python -m unittest</code> can now accept file paths
instead of module names for running specific tests (<a class="reference external" href="https://bugs.python.org/issue10620">bpo-10620</a>). The new
test discovery can find tests within packages, locating any test importable
from the top-level directory. The top-level directory can be specified with
the <cite>-t</cite> option, a pattern for matching files with <code class="docutils literal notranslate">-p</code>, and a directory to
start discovery with <code class="docutils literal notranslate">-s</code>:
<div class="highlight-shell-session notranslate"><div class="highlight"><pre>$ python -m unittest discover -s my_proj_dir -p _test.py
</pre></div>
</div>
(Contributed by Michael Foord.)
</li>
<li>Experimentation at the interactive prompt is now easier because the
<code class="xref py py-class docutils literal notranslate">unittest.case.TestCase</code> class can now be instantiated without
arguments:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; from unittest import TestCase
&gt;&gt;&gt; TestCase().assertEqual(pow(2, 3), 8)
</pre></div>
</div>
(Contributed by Michael Foord.)
</li>
<li>The <a class="reference internal" href="../library/unittest.html#module-unittest" title="unittest: Unit testing framework for Python."><code class="xref py py-mod docutils literal notranslate">unittest</code></a> module has two new methods,
<a class="reference internal" href="../library/unittest.html#unittest.TestCase.assertWarns" title="unittest.TestCase.assertWarns"><code class="xref py py-meth docutils literal notranslate">assertWarns()</code></a> and
<a class="reference internal" href="../library/unittest.html#unittest.TestCase.assertWarnsRegex" title="unittest.TestCase.assertWarnsRegex"><code class="xref py py-meth docutils literal notranslate">assertWarnsRegex()</code></a> to verify that a given warning type
is triggered by the code under test:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>with self.assertWarns(DeprecationWarning):
 legacy_function(&#39;XYZ&#39;)
</pre></div>
</div>
(Contributed by Antoine Pitrou, <a class="reference external" href="https://bugs.python.org/issue9754">bpo-9754</a>.)
Another new method, <a class="reference internal" href="../library/unittest.html#unittest.TestCase.assertCountEqual" title="unittest.TestCase.assertCountEqual"><code class="xref py py-meth docutils literal notranslate">assertCountEqual()</code></a> is used to
compare two iterables to determine if their element counts are equal (whether
the same elements are present with the same number of occurrences regardless
of order):
<div class="highlight-python3 notranslate"><div class="highlight"><pre>def test_anagram(self):
 self.assertCountEqual(&#39;algorithm&#39;, &#39;logarithm&#39;)
</pre></div>
</div>
(Contributed by Raymond Hettinger.)
</li>
<li>A principal feature of the unittest module is an effort to produce meaningful
diagnostics when a test fails. When possible, the failure is recorded along
with a diff of the output. This is especially helpful for analyzing log files
of failed test runs. However, since diffs can sometime be voluminous, there is
a new <a class="reference internal" href="../library/unittest.html#unittest.TestCase.maxDiff" title="unittest.TestCase.maxDiff"><code class="xref py py-attr docutils literal notranslate">maxDiff</code></a> attribute that sets maximum length of
diffs displayed.
</li>
<li>In addition, the method names in the module have undergone a number of clean-ups.
For example, <a class="reference internal" href="../library/unittest.html#unittest.TestCase.assertRegex" title="unittest.TestCase.assertRegex"><code class="xref py py-meth docutils literal notranslate">assertRegex()</code></a> is the new name for
<code class="xref py py-meth docutils literal notranslate">assertRegexpMatches()</code> which was misnamed because the
test uses <a class="reference internal" href="../library/re.html#re.search" title="re.search"><code class="xref py py-func docutils literal notranslate">re.search()</code></a>, not <a class="reference internal" href="../library/re.html#re.match" title="re.match"><code class="xref py py-func docutils literal notranslate">re.match()</code></a>. Other methods using
regular expressions are now named using short form “Regex” in preference to
“Regexp” – this matches the names used in other unittest implementations,
matches Python’s old name for the <a class="reference internal" href="../library/re.html#module-re" title="re: Regular expression operations."><code class="xref py py-mod docutils literal notranslate">re</code></a> module, and it has unambiguous
camel-casing.
(Contributed by Raymond Hettinger and implemented by Ezio Melotti.)
</li>
<li>To improve consistency, some long-standing method aliases are being
deprecated in favor of the preferred names:
<blockquote>
<div><table border="1" class="docutils">
<colgroup>
<col width="51%" />
<col width="49%" />
</colgroup>
<thead valign="bottom">
<tr class="row-odd"><th class="head">Old Name</th>
<th class="head">Preferred Name</th>
</tr>
</thead>
<tbody valign="top">
<tr class="row-even"><td><code class="xref py py-meth docutils literal notranslate">assert_()</code></td>
<td><a class="reference internal" href="../library/unittest.html#unittest.TestCase.assertTrue" title="unittest.TestCase.assertTrue"><code class="xref py py-meth docutils literal notranslate">assertTrue()</code></a></td>
</tr>
<tr class="row-odd"><td><code class="xref py py-meth docutils literal notranslate">assertEquals()</code></td>
<td><a class="reference internal" href="../library/unittest.html#unittest.TestCase.assertEqual" title="unittest.TestCase.assertEqual"><code class="xref py py-meth docutils literal notranslate">assertEqual()</code></a></td>
</tr>
<tr class="row-even"><td><code class="xref py py-meth docutils literal notranslate">assertNotEquals()</code></td>
<td><a class="reference internal" href="../library/unittest.html#unittest.TestCase.assertNotEqual" title="unittest.TestCase.assertNotEqual"><code class="xref py py-meth docutils literal notranslate">assertNotEqual()</code></a></td>
</tr>
<tr class="row-odd"><td><code class="xref py py-meth docutils literal notranslate">assertAlmostEquals()</code></td>
<td><a class="reference internal" href="../library/unittest.html#unittest.TestCase.assertAlmostEqual" title="unittest.TestCase.assertAlmostEqual"><code class="xref py py-meth docutils literal notranslate">assertAlmostEqual()</code></a></td>
</tr>
<tr class="row-even"><td><code class="xref py py-meth docutils literal notranslate">assertNotAlmostEquals()</code></td>
<td><a class="reference internal" href="../library/unittest.html#unittest.TestCase.assertNotAlmostEqual" title="unittest.TestCase.assertNotAlmostEqual"><code class="xref py py-meth docutils literal notranslate">assertNotAlmostEqual()</code></a></td>
</tr>
</tbody>
</table>
</div></blockquote>
Likewise, the <code class="docutils literal notranslate">TestCase.fail*</code> methods deprecated in Python 3.1 are expected
to be removed in Python 3.3. Also see the <a class="reference internal" href="../library/unittest.html#deprecated-aliases">Deprecated aliases</a> section in
the <a class="reference internal" href="../library/unittest.html#module-unittest" title="unittest: Unit testing framework for Python."><code class="xref py py-mod docutils literal notranslate">unittest</code></a> documentation.
(Contributed by Ezio Melotti; <a class="reference external" href="https://bugs.python.org/issue9424">bpo-9424</a>.)
</li>
<li>The <code class="xref py py-meth docutils literal notranslate">assertDictContainsSubset()</code> method was deprecated
because it was misimplemented with the arguments in the wrong order. This
created hard-to-debug optical illusions where tests like
<code class="docutils literal notranslate">TestCase().assertDictContainsSubset({'a':1, 'b':2}, {'a':1})</code> would fail.
(Contributed by Raymond Hettinger.)
</li>
</ul>
</div>
<div class="section" id="random">
<h3>random<a class="headerlink" href="#random" title="Permalink to this headline">¶</a></h3>
The integer methods in the <a class="reference internal" href="../library/random.html#module-random" title="random: Generate pseudo-random numbers with various common distributions."><code class="xref py py-mod docutils literal notranslate">random</code></a> module now do a better job of producing
uniform distributions. Previously, they computed selections with
<code class="docutils literal notranslate">int(n*random())</code> which had a slight bias whenever n was not a power of two.
Now, multiple selections are made from a range up to the next power of two and a
selection is kept only when it falls within the range <code class="docutils literal notranslate">0 &lt;= x &lt; n</code>. The
functions and methods affected are <a class="reference internal" href="../library/random.html#random.randrange" title="random.randrange"><code class="xref py py-func docutils literal notranslate">randrange()</code></a>,
<a class="reference internal" href="../library/random.html#random.randint" title="random.randint"><code class="xref py py-func docutils literal notranslate">randint()</code></a>, <a class="reference internal" href="../library/random.html#random.choice" title="random.choice"><code class="xref py py-func docutils literal notranslate">choice()</code></a>, <a class="reference internal" href="../library/random.html#random.shuffle" title="random.shuffle"><code class="xref py py-func docutils literal notranslate">shuffle()</code></a> and
<a class="reference internal" href="../library/random.html#random.sample" title="random.sample"><code class="xref py py-func docutils literal notranslate">sample()</code></a>.
(Contributed by Raymond Hettinger; <a class="reference external" href="https://bugs.python.org/issue9025">bpo-9025</a>.)
</div>
<div class="section" id="poplib">
<h3>poplib<a class="headerlink" href="#poplib" title="Permalink to this headline">¶</a></h3>
<a class="reference internal" href="../library/poplib.html#poplib.POP3_SSL" title="poplib.POP3_SSL"><code class="xref py py-class docutils literal notranslate">POP3_SSL</code></a> class now accepts a context parameter, which is a
<a class="reference internal" href="../library/ssl.html#ssl.SSLContext" title="ssl.SSLContext"><code class="xref py py-class docutils literal notranslate">ssl.SSLContext</code></a> object allowing bundling SSL configuration options,
certificates and private keys into a single (potentially long-lived)
structure.
(Contributed by Giampaolo Rodolà; <a class="reference external" href="https://bugs.python.org/issue8807">bpo-8807</a>.)
</div>
<div class="section" id="asyncore">
<h3>asyncore<a class="headerlink" href="#asyncore" title="Permalink to this headline">¶</a></h3>
<a class="reference internal" href="../library/asyncore.html#asyncore.dispatcher" title="asyncore.dispatcher"><code class="xref py py-class docutils literal notranslate">asyncore.dispatcher</code></a> now provides a
<a class="reference internal" href="../library/asyncore.html#asyncore.dispatcher.handle_accepted" title="asyncore.dispatcher.handle_accepted"><code class="xref py py-meth docutils literal notranslate">handle_accepted()</code></a> method
returning a <cite>(sock, addr)</cite> pair which is called when a connection has actually
been established with a new remote endpoint. This is supposed to be used as a
replacement for old <a class="reference internal" href="../library/asyncore.html#asyncore.dispatcher.handle_accept" title="asyncore.dispatcher.handle_accept"><code class="xref py py-meth docutils literal notranslate">handle_accept()</code></a> and avoids
the user to call <a class="reference internal" href="../library/asyncore.html#asyncore.dispatcher.accept" title="asyncore.dispatcher.accept"><code class="xref py py-meth docutils literal notranslate">accept()</code></a> directly.
(Contributed by Giampaolo Rodolà; <a class="reference external" href="https://bugs.python.org/issue6706">bpo-6706</a>.)
</div>
<div class="section" id="tempfile">
<h3>tempfile<a class="headerlink" href="#tempfile" title="Permalink to this headline">¶</a></h3>
The <a class="reference internal" href="../library/tempfile.html#module-tempfile" title="tempfile: Generate temporary files and directories."><code class="xref py py-mod docutils literal notranslate">tempfile</code></a> module has a new context manager,
<a class="reference internal" href="../library/tempfile.html#tempfile.TemporaryDirectory" title="tempfile.TemporaryDirectory"><code class="xref py py-class docutils literal notranslate">TemporaryDirectory</code></a> which provides easy deterministic
cleanup of temporary directories:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>with tempfile.TemporaryDirectory() as tmpdirname:
 print(&#39;created temporary dir:&#39;, tmpdirname)
</pre></div>
</div>
(Contributed by Neil Schemenauer and Nick Coghlan; <a class="reference external" href="https://bugs.python.org/issue5178">bpo-5178</a>.)
</div>
<div class="section" id="inspect">
<h3>inspect<a class="headerlink" href="#inspect" title="Permalink to this headline">¶</a></h3>
<ul>
<li>The <a class="reference internal" href="../library/inspect.html#module-inspect" title="inspect: Extract information and source code from live objects."><code class="xref py py-mod docutils literal notranslate">inspect</code></a> module has a new function
<a class="reference internal" href="../library/inspect.html#inspect.getgeneratorstate" title="inspect.getgeneratorstate"><code class="xref py py-func docutils literal notranslate">getgeneratorstate()</code></a> to easily identify the current state of a
generator-iterator:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; from inspect import getgeneratorstate
&gt;&gt;&gt; def gen():
... yield &#39;demo&#39;
&gt;&gt;&gt; g = gen()
&gt;&gt;&gt; getgeneratorstate(g)
&#39;GEN_CREATED&#39;
&gt;&gt;&gt; next(g)
&#39;demo&#39;
&gt;&gt;&gt; getgeneratorstate(g)
&#39;GEN_SUSPENDED&#39;
&gt;&gt;&gt; next(g, None)
&gt;&gt;&gt; getgeneratorstate(g)
&#39;GEN_CLOSED&#39;
</pre></div>
</div>
(Contributed by Rodolpho Eckhardt and Nick Coghlan, <a class="reference external" href="https://bugs.python.org/issue10220">bpo-10220</a>.)
</li>
<li>To support lookups without the possibility of activating a dynamic attribute,
the <a class="reference internal" href="../library/inspect.html#module-inspect" title="inspect: Extract information and source code from live objects."><code class="xref py py-mod docutils literal notranslate">inspect</code></a> module has a new function, <a class="reference internal" href="../library/inspect.html#inspect.getattr_static" title="inspect.getattr_static"><code class="xref py py-func docutils literal notranslate">getattr_static()</code></a>.
Unlike <a class="reference internal" href="../library/functions.html#hasattr" title="hasattr"><code class="xref py py-func docutils literal notranslate">hasattr()</code></a>, this is a true read-only search, guaranteed not to
change state while it is searching:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; class A:
... @property
... def f(self):
... print(&#39;Running&#39;)
... return 10
...
&gt;&gt;&gt; a = A()
&gt;&gt;&gt; getattr(a, &#39;f&#39;)
Running
10
&gt;&gt;&gt; inspect.getattr_static(a, &#39;f&#39;)
&lt;property object at 0x1022bd788&gt;
</pre></div>
</div>
</li>
</ul>
<blockquote>
<div>(Contributed by Michael Foord.)</div></blockquote>
</div>
<div class="section" id="pydoc">
<h3>pydoc<a class="headerlink" href="#pydoc" title="Permalink to this headline">¶</a></h3>
The <a class="reference internal" href="../library/pydoc.html#module-pydoc" title="pydoc: Documentation generator and online help system."><code class="xref py py-mod docutils literal notranslate">pydoc</code></a> module now provides a much-improved Web server interface, as
well as a new command-line option <code class="docutils literal notranslate">-b</code> to automatically open a browser window
to display that server:
<div class="highlight-shell-session notranslate"><div class="highlight"><pre>$ pydoc3.2 -b
</pre></div>
</div>
(Contributed by Ron Adam; <a class="reference external" href="https://bugs.python.org/issue2001">bpo-2001</a>.)
</div>
<div class="section" id="dis">
<h3>dis<a class="headerlink" href="#dis" title="Permalink to this headline">¶</a></h3>
The <a class="reference internal" href="../library/dis.html#module-dis" title="dis: Disassembler for Python bytecode."><code class="xref py py-mod docutils literal notranslate">dis</code></a> module gained two new functions for inspecting code,
<a class="reference internal" href="../library/dis.html#dis.code_info" title="dis.code_info"><code class="xref py py-func docutils literal notranslate">code_info()</code></a> and <a class="reference internal" href="../library/dis.html#dis.show_code" title="dis.show_code"><code class="xref py py-func docutils literal notranslate">show_code()</code></a>. Both provide detailed code
object information for the supplied function, method, source code string or code
object. The former returns a string and the latter prints it:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; import dis, random
&gt;&gt;&gt; dis.show_code(random.choice)
Name: choice
Filename: /Library/Frameworks/Python.framework/Versions/3.2/lib/python3.2/random.py
Argument count: 2
Kw-only arguments: 0
Number of locals: 3
Stack size: 11
Flags: OPTIMIZED, NEWLOCALS, NOFREE
Constants:
 0: &#39;Choose a random element from a non-empty sequence.&#39;
 1: &#39;Cannot choose from an empty sequence&#39;
Names:
 0: _randbelow
 1: len
 2: ValueError
 3: IndexError
Variable names:
 0: self
 1: seq
 2: i
</pre></div>
</div>
In addition, the <a class="reference internal" href="../library/dis.html#dis.dis" title="dis.dis"><code class="xref py py-func docutils literal notranslate">dis()</code></a> function now accepts string arguments
so that the common idiom <code class="docutils literal notranslate">dis(compile(s, '', 'eval'))</code> can be shortened
to <code class="docutils literal notranslate">dis(s)</code>:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; dis(&#39;3*x+1 if x%2==1 else x//2&#39;)
 1 0 LOAD_NAME 0 (x)
 3 LOAD_CONST 0 (2)
 6 BINARY_MODULO
 7 LOAD_CONST 1 (1)
 10 COMPARE_OP 2 (==)
 13 POP_JUMP_IF_FALSE 28
 16 LOAD_CONST 2 (3)
 19 LOAD_NAME 0 (x)
 22 BINARY_MULTIPLY
 23 LOAD_CONST 1 (1)
 26 BINARY_ADD
 27 RETURN_VALUE
 &gt;&gt; 28 LOAD_NAME 0 (x)
 31 LOAD_CONST 0 (2)
 34 BINARY_FLOOR_DIVIDE
 35 RETURN_VALUE
</pre></div>
</div>
Taken together, these improvements make it easier to explore how CPython is
implemented and to see for yourself what the language syntax does
under-the-hood.
(Contributed by Nick Coghlan in <a class="reference external" href="https://bugs.python.org/issue9147">bpo-9147</a>.)
</div>
<div class="section" id="dbm">
<h3>dbm<a class="headerlink" href="#dbm" title="Permalink to this headline">¶</a></h3>
All database modules now support the <code class="xref py py-meth docutils literal notranslate">get()</code> and <code class="xref py py-meth docutils literal notranslate">setdefault()</code> methods.
(Suggested by Ray Allen in <a class="reference external" href="https://bugs.python.org/issue9523">bpo-9523</a>.)
</div>
<div class="section" id="ctypes">
<h3>ctypes<a class="headerlink" href="#ctypes" title="Permalink to this headline">¶</a></h3>
A new type, <a class="reference internal" href="../library/ctypes.html#ctypes.c_ssize_t" title="ctypes.c_ssize_t"><code class="xref py py-class docutils literal notranslate">ctypes.c_ssize_t</code></a> represents the C <code class="xref c c-type docutils literal notranslate">ssize_t</code> datatype.
</div>
<div class="section" id="site">
<h3>site<a class="headerlink" href="#site" title="Permalink to this headline">¶</a></h3>
The <a class="reference internal" href="../library/site.html#module-site" title="site: Module responsible for site-specific configuration."><code class="xref py py-mod docutils literal notranslate">site</code></a> module has three new functions useful for reporting on the
details of a given Python installation.
<ul class="simple">
<li><a class="reference internal" href="../library/site.html#site.getsitepackages" title="site.getsitepackages"><code class="xref py py-func docutils literal notranslate">getsitepackages()</code></a> lists all global site-packages directories.</li>
<li><a class="reference internal" href="../library/site.html#site.getuserbase" title="site.getuserbase"><code class="xref py py-func docutils literal notranslate">getuserbase()</code></a> reports on the user’s base directory where data can
be stored.</li>
<li><a class="reference internal" href="../library/site.html#site.getusersitepackages" title="site.getusersitepackages"><code class="xref py py-func docutils literal notranslate">getusersitepackages()</code></a> reveals the user-specific site-packages
directory path.</li>
</ul>
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; import site
&gt;&gt;&gt; site.getsitepackages()
[&#39;/Library/Frameworks/Python.framework/Versions/3.2/lib/python3.2/site-packages&#39;,
 &#39;/Library/Frameworks/Python.framework/Versions/3.2/lib/site-python&#39;,
 &#39;/Library/Python/3.2/site-packages&#39;]
&gt;&gt;&gt; site.getuserbase()
&#39;/Users/raymondhettinger/Library/Python/3.2&#39;
&gt;&gt;&gt; site.getusersitepackages()
&#39;/Users/raymondhettinger/Library/Python/3.2/lib/python/site-packages&#39;
</pre></div>
</div>
Conveniently, some of site’s functionality is accessible directly from the
command-line:
<div class="highlight-shell-session notranslate"><div class="highlight"><pre>$ python -m site --user-base
/Users/raymondhettinger/.local
$ python -m site --user-site
/Users/raymondhettinger/.local/lib/python3.2/site-packages
</pre></div>
</div>
(Contributed by Tarek Ziadé in <a class="reference external" href="https://bugs.python.org/issue6693">bpo-6693</a>.)
</div>
<div class="section" id="sysconfig">
<h3>sysconfig<a class="headerlink" href="#sysconfig" title="Permalink to this headline">¶</a></h3>
The new <a class="reference internal" href="../library/sysconfig.html#module-sysconfig" title="sysconfig: Python's configuration information"><code class="xref py py-mod docutils literal notranslate">sysconfig</code></a> module makes it straightforward to discover
installation paths and configuration variables that vary across platforms and
installations.
The module offers access simple access functions for platform and version
information:
<ul class="simple">
<li><a class="reference internal" href="../library/sysconfig.html#sysconfig.get_platform" title="sysconfig.get_platform"><code class="xref py py-func docutils literal notranslate">get_platform()</code></a> returning values like linux-i586 or
macosx-10.6-ppc.</li>
<li><a class="reference internal" href="../library/sysconfig.html#sysconfig.get_python_version" title="sysconfig.get_python_version"><code class="xref py py-func docutils literal notranslate">get_python_version()</code></a> returns a Python version string
such as “3.2”.</li>
</ul>
It also provides access to the paths and variables corresponding to one of
seven named schemes used by <a class="reference internal" href="../library/distutils.html#module-distutils" title="distutils: Support for building and installing Python modules into an existing Python installation."><code class="xref py py-mod docutils literal notranslate">distutils</code></a>. Those include posix_prefix,
posix_home, posix_user, nt, nt_user, os2, os2_home:
<ul class="simple">
<li><a class="reference internal" href="../library/sysconfig.html#sysconfig.get_paths" title="sysconfig.get_paths"><code class="xref py py-func docutils literal notranslate">get_paths()</code></a> makes a dictionary containing installation paths
for the current installation scheme.</li>
<li><a class="reference internal" href="../library/sysconfig.html#sysconfig.get_config_vars" title="sysconfig.get_config_vars"><code class="xref py py-func docutils literal notranslate">get_config_vars()</code></a> returns a dictionary of platform specific
variables.</li>
</ul>
There is also a convenient command-line interface:
<div class="highlight-doscon notranslate"><div class="highlight"><pre>C:\Python32&gt;python -m sysconfig
Platform: &quot;win32&quot;
Python version: &quot;3.2&quot;
Current installation scheme: &quot;nt&quot;

Paths:
 data = &quot;C:\Python32&quot;
 include = &quot;C:\Python32\Include&quot;
 platinclude = &quot;C:\Python32\Include&quot;
 platlib = &quot;C:\Python32\Lib\site-packages&quot;
 platstdlib = &quot;C:\Python32\Lib&quot;
 purelib = &quot;C:\Python32\Lib\site-packages&quot;
 scripts = &quot;C:\Python32\Scripts&quot;
 stdlib = &quot;C:\Python32\Lib&quot;

Variables:
 BINDIR = &quot;C:\Python32&quot;
 BINLIBDEST = &quot;C:\Python32\Lib&quot;
 EXE = &quot;.exe&quot;
 INCLUDEPY = &quot;C:\Python32\Include&quot;
 LIBDEST = &quot;C:\Python32\Lib&quot;
 SO = &quot;.pyd&quot;
 VERSION = &quot;32&quot;
 abiflags = &quot;&quot;
 base = &quot;C:\Python32&quot;
 exec_prefix = &quot;C:\Python32&quot;
 platbase = &quot;C:\Python32&quot;
 prefix = &quot;C:\Python32&quot;
 projectbase = &quot;C:\Python32&quot;
 py_version = &quot;3.2&quot;
 py_version_nodot = &quot;32&quot;
 py_version_short = &quot;3.2&quot;
 srcdir = &quot;C:\Python32&quot;
 userbase = &quot;C:\Documents and Settings\Raymond\Application Data\Python&quot;
</pre></div>
</div>
(Moved out of Distutils by Tarek Ziadé.)
</div>
<div class="section" id="pdb">
<h3>pdb<a class="headerlink" href="#pdb" title="Permalink to this headline">¶</a></h3>
The <a class="reference internal" href="../library/pdb.html#module-pdb" title="pdb: The Python debugger for interactive interpreters."><code class="xref py py-mod docutils literal notranslate">pdb</code></a> debugger module gained a number of usability improvements:
<ul class="simple">
<li><code class="file docutils literal notranslate">pdb.py</code> now has a <code class="docutils literal notranslate">-c</code> option that executes commands as given in a
<code class="file docutils literal notranslate">.pdbrc</code> script file.</li>
<li>A <code class="file docutils literal notranslate">.pdbrc</code> script file can contain <code class="docutils literal notranslate">continue</code> and <code class="docutils literal notranslate">next</code> commands
that continue debugging.</li>
<li>The <code class="xref py py-class docutils literal notranslate">Pdb</code> class constructor now accepts a nosigint argument.</li>
<li>New commands: <code class="docutils literal notranslate">l(list)</code>, <code class="docutils literal notranslate">ll(long list)</code> and <code class="docutils literal notranslate">source</code> for
listing source code.</li>
<li>New commands: <code class="docutils literal notranslate">display</code> and <code class="docutils literal notranslate">undisplay</code> for showing or hiding
the value of an expression if it has changed.</li>
<li>New command: <code class="docutils literal notranslate">interact</code> for starting an interactive interpreter containing
the global and local names found in the current scope.</li>
<li>Breakpoints can be cleared by breakpoint number.</li>
</ul>
(Contributed by Georg Brandl, Antonio Cuni and Ilya Sandler.)
</div>
<div class="section" id="configparser">
<h3>configparser<a class="headerlink" href="#configparser" title="Permalink to this headline">¶</a></h3>
The <a class="reference internal" href="../library/configparser.html#module-configparser" title="configparser: Configuration file parser."><code class="xref py py-mod docutils literal notranslate">configparser</code></a> module was modified to improve usability and
predictability of the default parser and its supported INI syntax. The old
<code class="xref py py-class docutils literal notranslate">ConfigParser</code> class was removed in favor of <code class="xref py py-class docutils literal notranslate">SafeConfigParser</code>
which has in turn been renamed to <a class="reference internal" href="../library/configparser.html#configparser.ConfigParser" title="configparser.ConfigParser"><code class="xref py py-class docutils literal notranslate">ConfigParser</code></a>. Support
for inline comments is now turned off by default and section or option
duplicates are not allowed in a single configuration source.
Config parsers gained a new API based on the mapping protocol:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; parser = ConfigParser()
&gt;&gt;&gt; parser.read_string(&quot;&quot;&quot;
... [DEFAULT]
... location = upper left
... visible = yes
... editable = no
... color = blue
...
... [main]
... title = Main Menu
... color = green
...
... [options]
... title = Options
... &quot;&quot;&quot;)
&gt;&gt;&gt; parser[&#39;main&#39;][&#39;color&#39;]
&#39;green&#39;
&gt;&gt;&gt; parser[&#39;main&#39;][&#39;editable&#39;]
&#39;no&#39;
&gt;&gt;&gt; section = parser[&#39;options&#39;]
&gt;&gt;&gt; section[&#39;title&#39;]
&#39;Options&#39;
&gt;&gt;&gt; section[&#39;title&#39;] = &#39;Options (editable: %(editable)s)&#39;
&gt;&gt;&gt; section[&#39;title&#39;]
&#39;Options (editable: no)&#39;
</pre></div>
</div>
The new API is implemented on top of the classical API, so custom parser
subclasses should be able to use it without modifications.
The INI file structure accepted by config parsers can now be customized. Users
can specify alternative option/value delimiters and comment prefixes, change the
name of the DEFAULT section or switch the interpolation syntax.
There is support for pluggable interpolation including an additional interpolation
handler <a class="reference internal" href="../library/configparser.html#configparser.ExtendedInterpolation" title="configparser.ExtendedInterpolation"><code class="xref py py-class docutils literal notranslate">ExtendedInterpolation</code></a>:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; parser = ConfigParser(interpolation=ExtendedInterpolation())
&gt;&gt;&gt; parser.read_dict({&#39;buildout&#39;: {&#39;directory&#39;: &#39;/home/ambv/zope9&#39;},
... &#39;custom&#39;: {&#39;prefix&#39;: &#39;/usr/local&#39;}})
&gt;&gt;&gt; parser.read_string(&quot;&quot;&quot;
... [buildout]
... parts =
... zope9
... instance
... find-links =
... ${buildout:directory}/downloads/dist
...
... [zope9]
... recipe = plone.recipe.zope9install
... location = /opt/zope
...
... [instance]
... recipe = plone.recipe.zope9instance
... zope9-location = ${zope9:location}
... zope-conf = ${custom:prefix}/etc/zope.conf
... &quot;&quot;&quot;)
&gt;&gt;&gt; parser[&#39;buildout&#39;][&#39;find-links&#39;]
&#39;\n/home/ambv/zope9/downloads/dist&#39;
&gt;&gt;&gt; parser[&#39;instance&#39;][&#39;zope-conf&#39;]
&#39;/usr/local/etc/zope.conf&#39;
&gt;&gt;&gt; instance = parser[&#39;instance&#39;]
&gt;&gt;&gt; instance[&#39;zope-conf&#39;]
&#39;/usr/local/etc/zope.conf&#39;
&gt;&gt;&gt; instance[&#39;zope9-location&#39;]
&#39;/opt/zope&#39;
</pre></div>
</div>
A number of smaller features were also introduced, like support for specifying
encoding in read operations, specifying fallback values for get-functions, or
reading directly from dictionaries and strings.
(All changes contributed by Łukasz Langa.)
</div>
<div class="section" id="urllib-parse">
<h3>urllib.parse<a class="headerlink" href="#urllib-parse" title="Permalink to this headline">¶</a></h3>
A number of usability improvements were made for the <a class="reference internal" href="../library/urllib.parse.html#module-urllib.parse" title="urllib.parse: Parse URLs into or assemble them from components."><code class="xref py py-mod docutils literal notranslate">urllib.parse</code></a> module.
The <a class="reference internal" href="../library/urllib.parse.html#urllib.parse.urlparse" title="urllib.parse.urlparse"><code class="xref py py-func docutils literal notranslate">urlparse()</code></a> function now supports <a class="reference external" href="https://en.wikipedia.org/wiki/IPv6">IPv6</a> addresses as described in <a class="rfc reference external" href="https://tools.ietf.org/html/rfc2732.html">RFC 2732</a>:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; import urllib.parse
&gt;&gt;&gt; urllib.parse.urlparse(&#39;http://[dead:beef:cafe:5417:affe:8FA3:deaf:feed]/foo/&#39;) 
ParseResult(scheme=&#39;http&#39;,
 netloc=&#39;[dead:beef:cafe:5417:affe:8FA3:deaf:feed]&#39;,
 path=&#39;/foo/&#39;,
 params=&#39;&#39;,
 query=&#39;&#39;,
 fragment=&#39;&#39;)
</pre></div>
</div>
The <a class="reference internal" href="../library/urllib.parse.html#urllib.parse.urldefrag" title="urllib.parse.urldefrag"><code class="xref py py-func docutils literal notranslate">urldefrag()</code></a> function now returns a <a class="reference internal" href="../glossary.html#term-named-tuple">named tuple</a>:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; r = urllib.parse.urldefrag(&#39;http://python.org/about/#target&#39;)
&gt;&gt;&gt; r
DefragResult(url=&#39;http://python.org/about/&#39;, fragment=&#39;target&#39;)
&gt;&gt;&gt; r[0]
&#39;http://python.org/about/&#39;
&gt;&gt;&gt; r.fragment
&#39;target&#39;
</pre></div>
</div>
And, the <a class="reference internal" href="../library/urllib.parse.html#urllib.parse.urlencode" title="urllib.parse.urlencode"><code class="xref py py-func docutils literal notranslate">urlencode()</code></a> function is now much more flexible,
accepting either a string or bytes type for the query argument. If it is a
string, then the safe, encoding, and error parameters are sent to
<a class="reference internal" href="../library/urllib.parse.html#urllib.parse.quote_plus" title="urllib.parse.quote_plus"><code class="xref py py-func docutils literal notranslate">quote_plus()</code></a> for encoding:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; urllib.parse.urlencode([
... (&#39;type&#39;, &#39;telenovela&#39;),
... (&#39;name&#39;, &#39;¿Dónde Está Elisa?&#39;)],
... encoding=&#39;latin-1&#39;)
&#39;type=telenovela&amp;name=%BFD%F3nde+Est%E1+Elisa%3F&#39;
</pre></div>
</div>
As detailed in <a class="reference internal" href="../library/urllib.parse.html#parsing-ascii-encoded-bytes">Parsing ASCII Encoded Bytes</a>, all the <a class="reference internal" href="../library/urllib.parse.html#module-urllib.parse" title="urllib.parse: Parse URLs into or assemble them from components."><code class="xref py py-mod docutils literal notranslate">urllib.parse</code></a>
functions now accept ASCII-encoded byte strings as input, so long as they are
not mixed with regular strings. If ASCII-encoded byte strings are given as
parameters, the return types will also be an ASCII-encoded byte strings:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>&gt;&gt;&gt; urllib.parse.urlparse(b&#39;http://www.python.org:80/about/&#39;) 
ParseResultBytes(scheme=b&#39;http&#39;, netloc=b&#39;www.python.org:80&#39;,
 path=b&#39;/about/&#39;, params=b&#39;&#39;, query=b&#39;&#39;, fragment=b&#39;&#39;)
</pre></div>
</div>
(Work by Nick Coghlan, Dan Mahn, and Senthil Kumaran in <a class="reference external" href="https://bugs.python.org/issue2987">bpo-2987</a>,
<a class="reference external" href="https://bugs.python.org/issue5468">bpo-5468</a>, and <a class="reference external" href="https://bugs.python.org/issue9873">bpo-9873</a>.)
</div>
<div class="section" id="mailbox">
<h3>mailbox<a class="headerlink" href="#mailbox" title="Permalink to this headline">¶</a></h3>
Thanks to a concerted effort by R. David Murray, the <a class="reference internal" href="../library/mailbox.html#module-mailbox" title="mailbox: Manipulate mailboxes in various formats"><code class="xref py py-mod docutils literal notranslate">mailbox</code></a> module has
been fixed for Python 3.2. The challenge was that mailbox had been originally
designed with a text interface, but email messages are best represented with
<a class="reference internal" href="../library/stdtypes.html#bytes" title="bytes"><code class="xref py py-class docutils literal notranslate">bytes</code></a> because various parts of a message may have different encodings.
The solution harnessed the <a class="reference internal" href="../library/email.html#module-email" title="email: Package supporting the parsing, manipulating, and generating email messages."><code class="xref py py-mod docutils literal notranslate">email</code></a> package’s binary support for parsing
arbitrary email messages. In addition, the solution required a number of API
changes.
As expected, the <a class="reference internal" href="../library/mailbox.html#mailbox.Mailbox.add" title="mailbox.Mailbox.add"><code class="xref py py-meth docutils literal notranslate">add()</code></a> method for
<a class="reference internal" href="../library/mailbox.html#mailbox.Mailbox" title="mailbox.Mailbox"><code class="xref py py-class docutils literal notranslate">mailbox.Mailbox</code></a> objects now accepts binary input.
<a class="reference internal" href="../library/io.html#io.StringIO" title="io.StringIO"><code class="xref py py-class docutils literal notranslate">StringIO</code></a> and text file input are deprecated. Also, string input
will fail early if non-ASCII characters are used. Previously it would fail when
the email was processed in a later step.
There is also support for binary output. The <a class="reference internal" href="../library/mailbox.html#mailbox.Mailbox.get_file" title="mailbox.Mailbox.get_file"><code class="xref py py-meth docutils literal notranslate">get_file()</code></a>
method now returns a file in the binary mode (where it used to incorrectly set
the file to text-mode). There is also a new <a class="reference internal" href="../library/mailbox.html#mailbox.Mailbox.get_bytes" title="mailbox.Mailbox.get_bytes"><code class="xref py py-meth docutils literal notranslate">get_bytes()</code></a>
method that returns a <a class="reference internal" href="../library/stdtypes.html#bytes" title="bytes"><code class="xref py py-class docutils literal notranslate">bytes</code></a> representation of a message corresponding
to a given key.
It is still possible to get non-binary output using the old API’s
<a class="reference internal" href="../library/mailbox.html#mailbox.Mailbox.get_string" title="mailbox.Mailbox.get_string"><code class="xref py py-meth docutils literal notranslate">get_string()</code></a> method, but that approach
is not very useful. Instead, it is best to extract messages from
a <a class="reference internal" href="../library/mailbox.html#mailbox.Message" title="mailbox.Message"><code class="xref py py-class docutils literal notranslate">Message</code></a> object or to load them from binary input.
(Contributed by R. David Murray, with efforts from Steffen Daode Nurpmeso and an
initial patch by Victor Stinner in <a class="reference external" href="https://bugs.python.org/issue9124">bpo-9124</a>.)
</div>
<div class="section" id="turtledemo">
<h3>turtledemo<a class="headerlink" href="#turtledemo" title="Permalink to this headline">¶</a></h3>
The demonstration code for the <a class="reference internal" href="../library/turtle.html#module-turtle" title="turtle: An educational framework for simple graphics applications"><code class="xref py py-mod docutils literal notranslate">turtle</code></a> module was moved from the Demo
directory to main library. It includes over a dozen sample scripts with
lively displays. Being on <a class="reference internal" href="../library/sys.html#sys.path" title="sys.path"><code class="xref py py-attr docutils literal notranslate">sys.path</code></a>, it can now be run directly
from the command-line:
<div class="highlight-shell-session notranslate"><div class="highlight"><pre>$ python -m turtledemo
</pre></div>
</div>
(Moved from the Demo directory by Alexander Belopolsky in <a class="reference external" href="https://bugs.python.org/issue10199">bpo-10199</a>.)
</div>
</div>
<div class="section" id="multi-threading">
<h2>Multi-threading<a class="headerlink" href="#multi-threading" title="Permalink to this headline">¶</a></h2>
<ul>
<li>The mechanism for serializing execution of concurrently running Python threads
(generally known as the <a class="reference internal" href="../glossary.html#term-gil">GIL</a> or <a class="reference internal" href="../glossary.html#term-global-interpreter-lock">Global Interpreter Lock</a>) has
been rewritten. Among the objectives were more predictable switching
intervals and reduced overhead due to lock contention and the number of
ensuing system calls. The notion of a “check interval” to allow thread
switches has been abandoned and replaced by an absolute duration expressed in
seconds. This parameter is tunable through <a class="reference internal" href="../library/sys.html#sys.setswitchinterval" title="sys.setswitchinterval"><code class="xref py py-func docutils literal notranslate">sys.setswitchinterval()</code></a>.
It currently defaults to 5 milliseconds.
Additional details about the implementation can be read from a <a class="reference external" href="https://mail.python.org/pipermail/python-dev/2009-October/093321.html">python-dev
mailing-list message</a>
(however, “priority requests” as exposed in this message have not been kept
for inclusion).
(Contributed by Antoine Pitrou.)
</li>
<li>Regular and recursive locks now accept an optional timeout argument to their
<a class="reference internal" href="../library/threading.html#threading.Lock.acquire" title="threading.Lock.acquire"><code class="xref py py-meth docutils literal notranslate">acquire()</code></a> method. (Contributed by Antoine Pitrou;
<a class="reference external" href="https://bugs.python.org/issue7316">bpo-7316</a>.)
</li>
<li>Similarly, <a class="reference internal" href="../library/threading.html#threading.Semaphore.acquire" title="threading.Semaphore.acquire"><code class="xref py py-meth docutils literal notranslate">threading.Semaphore.acquire()</code></a> also gained a timeout
argument. (Contributed by Torsten Landschoff; <a class="reference external" href="https://bugs.python.org/issue850728">bpo-850728</a>.)
</li>
<li>Regular and recursive lock acquisitions can now be interrupted by signals on
platforms using Pthreads. This means that Python programs that deadlock while
acquiring locks can be successfully killed by repeatedly sending SIGINT to the
process (by pressing <kbd class="kbd docutils literal notranslate">Ctrl+C</kbd> in most shells).
(Contributed by Reid Kleckner; <a class="reference external" href="https://bugs.python.org/issue8844">bpo-8844</a>.)
</li>
</ul>
</div>
<div class="section" id="optimizations">
<h2>Optimizations<a class="headerlink" href="#optimizations" title="Permalink to this headline">¶</a></h2>
A number of small performance enhancements have been added:
<ul>
<li>Python’s peephole optimizer now recognizes patterns such <code class="docutils literal notranslate">x in {1, 2, 3}</code> as
being a test for membership in a set of constants. The optimizer recasts the
<a class="reference internal" href="../library/stdtypes.html#set" title="set"><code class="xref py py-class docutils literal notranslate">set</code></a> as a <a class="reference internal" href="../library/stdtypes.html#frozenset" title="frozenset"><code class="xref py py-class docutils literal notranslate">frozenset</code></a> and stores the pre-built constant.
Now that the speed penalty is gone, it is practical to start writing
membership tests using set-notation. This style is both semantically clear
and operationally fast:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>extension = name.rpartition(&#39;.&#39;)[2]
if extension in {&#39;xml&#39;, &#39;html&#39;, &#39;xhtml&#39;, &#39;css&#39;}:
 handle(name)
</pre></div>
</div>
(Patch and additional tests contributed by Dave Malcolm; <a class="reference external" href="https://bugs.python.org/issue6690">bpo-6690</a>).
</li>
<li>Serializing and unserializing data using the <a class="reference internal" href="../library/pickle.html#module-pickle" title="pickle: Convert Python objects to streams of bytes and back."><code class="xref py py-mod docutils literal notranslate">pickle</code></a> module is now
several times faster.
(Contributed by Alexandre Vassalotti, Antoine Pitrou
and the Unladen Swallow team in <a class="reference external" href="https://bugs.python.org/issue9410">bpo-9410</a> and <a class="reference external" href="https://bugs.python.org/issue3873">bpo-3873</a>.)
</li>
<li>The <a class="reference external" href="https://en.wikipedia.org/wiki/Timsort">Timsort algorithm</a> used in
<a class="reference internal" href="../library/stdtypes.html#list.sort" title="list.sort"><code class="xref py py-meth docutils literal notranslate">list.sort()</code></a> and <a class="reference internal" href="../library/functions.html#sorted" title="sorted"><code class="xref py py-func docutils literal notranslate">sorted()</code></a> now runs faster and uses less memory
when called with a <a class="reference internal" href="../glossary.html#term-key-function">key function</a>. Previously, every element of
a list was wrapped with a temporary object that remembered the key value
associated with each element. Now, two arrays of keys and values are
sorted in parallel. This saves the memory consumed by the sort wrappers,
and it saves time lost to delegating comparisons.
(Patch by Daniel Stutzbach in <a class="reference external" href="https://bugs.python.org/issue9915">bpo-9915</a>.)
</li>
<li>JSON decoding performance is improved and memory consumption is reduced
whenever the same string is repeated for multiple keys. Also, JSON encoding
now uses the C speedups when the <code class="docutils literal notranslate">sort_keys</code> argument is true.
(Contributed by Antoine Pitrou in <a class="reference external" href="https://bugs.python.org/issue7451">bpo-7451</a> and by Raymond Hettinger and
Antoine Pitrou in <a class="reference external" href="https://bugs.python.org/issue10314">bpo-10314</a>.)
</li>
<li>Recursive locks (created with the <a class="reference internal" href="../library/threading.html#threading.RLock" title="threading.RLock"><code class="xref py py-func docutils literal notranslate">threading.RLock()</code></a> API) now benefit
from a C implementation which makes them as fast as regular locks, and between
10x and 15x faster than their previous pure Python implementation.
(Contributed by Antoine Pitrou; <a class="reference external" href="https://bugs.python.org/issue3001">bpo-3001</a>.)
</li>
<li>The fast-search algorithm in stringlib is now used by the <code class="xref py py-meth docutils literal notranslate">split()</code>,
<code class="xref py py-meth docutils literal notranslate">rsplit()</code>, <code class="xref py py-meth docutils literal notranslate">splitlines()</code> and <code class="xref py py-meth docutils literal notranslate">replace()</code> methods on
<a class="reference internal" href="../library/stdtypes.html#bytes" title="bytes"><code class="xref py py-class docutils literal notranslate">bytes</code></a>, <a class="reference internal" href="../library/stdtypes.html#bytearray" title="bytearray"><code class="xref py py-class docutils literal notranslate">bytearray</code></a> and <a class="reference internal" href="../library/stdtypes.html#str" title="str"><code class="xref py py-class docutils literal notranslate">str</code></a> objects. Likewise, the
algorithm is also used by <code class="xref py py-meth docutils literal notranslate">rfind()</code>, <code class="xref py py-meth docutils literal notranslate">rindex()</code>, <code class="xref py py-meth docutils literal notranslate">rsplit()</code> and
<code class="xref py py-meth docutils literal notranslate">rpartition()</code>.
(Patch by Florent Xicluna in <a class="reference external" href="https://bugs.python.org/issue7622">bpo-7622</a> and <a class="reference external" href="https://bugs.python.org/issue7462">bpo-7462</a>.)
</li>
<li>Integer to string conversions now work two “digits” at a time, reducing the
number of division and modulo operations.
(<a class="reference external" href="https://bugs.python.org/issue6713">bpo-6713</a> by Gawain Bolton, Mark Dickinson, and Victor Stinner.)
</li>
</ul>
There were several other minor optimizations. Set differencing now runs faster
when one operand is much larger than the other (patch by Andress Bennetts in
<a class="reference external" href="https://bugs.python.org/issue8685">bpo-8685</a>). The <code class="xref py py-meth docutils literal notranslate">array.repeat()</code> method has a faster implementation
(<a class="reference external" href="https://bugs.python.org/issue1569291">bpo-1569291</a> by Alexander Belopolsky). The <code class="xref py py-class docutils literal notranslate">BaseHTTPRequestHandler</code>
has more efficient buffering (<a class="reference external" href="https://bugs.python.org/issue3709">bpo-3709</a> by Andrew Schaaf). The
<a class="reference internal" href="../library/operator.html#operator.attrgetter" title="operator.attrgetter"><code class="xref py py-func docutils literal notranslate">operator.attrgetter()</code></a> function has been sped-up (<a class="reference external" href="https://bugs.python.org/issue10160">bpo-10160</a> by
Christos Georgiou). And <code class="xref py py-class docutils literal notranslate">ConfigParser</code> loads multi-line arguments a bit
faster (<a class="reference external" href="https://bugs.python.org/issue7113">bpo-7113</a> by Łukasz Langa).
</div>
<div class="section" id="unicode">
<h2>Unicode<a class="headerlink" href="#unicode" title="Permalink to this headline">¶</a></h2>
Python has been updated to <a class="reference external" href="http://unicode.org/versions/Unicode6.0.0/">Unicode 6.0.0</a>. The update to the standard adds
over 2,000 new characters including <a class="reference external" href="https://en.wikipedia.org/wiki/Emoji">emoji</a>
symbols which are important for mobile phones.
In addition, the updated standard has altered the character properties for two
Kannada characters (U+0CF1, U+0CF2) and one New Tai Lue numeric character
(U+19DA), making the former eligible for use in identifiers while disqualifying
the latter. For more information, see <a class="reference external" href="http://www.unicode.org/versions/Unicode6.0.0/#Database_Changes">Unicode Character Database Changes</a>.
</div>
<div class="section" id="codecs">
<h2>Codecs<a class="headerlink" href="#codecs" title="Permalink to this headline">¶</a></h2>
Support was added for cp720 Arabic DOS encoding (<a class="reference external" href="https://bugs.python.org/issue1616979">bpo-1616979</a>).
MBCS encoding no longer ignores the error handler argument. In the default
strict mode, it raises an <a class="reference internal" href="../library/exceptions.html#UnicodeDecodeError" title="UnicodeDecodeError"><code class="xref py py-exc docutils literal notranslate">UnicodeDecodeError</code></a> when it encounters an
undecodable byte sequence and an <a class="reference internal" href="../library/exceptions.html#UnicodeEncodeError" title="UnicodeEncodeError"><code class="xref py py-exc docutils literal notranslate">UnicodeEncodeError</code></a> for an unencodable
character.
The MBCS codec supports <code class="docutils literal notranslate">'strict'</code> and <code class="docutils literal notranslate">'ignore'</code> error handlers for
decoding, and <code class="docutils literal notranslate">'strict'</code> and <code class="docutils literal notranslate">'replace'</code> for encoding.
To emulate Python3.1 MBCS encoding, select the <code class="docutils literal notranslate">'ignore'</code> handler for decoding
and the <code class="docutils literal notranslate">'replace'</code> handler for encoding.
On Mac OS X, Python decodes command line arguments with <code class="docutils literal notranslate">'utf-8'</code> rather than
the locale encoding.
By default, <a class="reference internal" href="../library/tarfile.html#module-tarfile" title="tarfile: Read and write tar-format archive files."><code class="xref py py-mod docutils literal notranslate">tarfile</code></a> uses <code class="docutils literal notranslate">'utf-8'</code> encoding on Windows (instead of
<code class="docutils literal notranslate">'mbcs'</code>) and the <code class="docutils literal notranslate">'surrogateescape'</code> error handler on all operating
systems.
</div>
<div class="section" id="documentation">
<h2>Documentation<a class="headerlink" href="#documentation" title="Permalink to this headline">¶</a></h2>
The documentation continues to be improved.
<ul>
<li>A table of quick links has been added to the top of lengthy sections such as
<a class="reference internal" href="../library/functions.html#built-in-funcs">Built-in Functions</a>. In the case of <a class="reference internal" href="../library/itertools.html#module-itertools" title="itertools: Functions creating iterators for efficient looping."><code class="xref py py-mod docutils literal notranslate">itertools</code></a>, the links are
accompanied by tables of cheatsheet-style summaries to provide an overview and
memory jog without having to read all of the docs.
</li>
<li>In some cases, the pure Python source code can be a helpful adjunct to the
documentation, so now many modules now feature quick links to the latest
version of the source code. For example, the <a class="reference internal" href="../library/functools.html#module-functools" title="functools: Higher-order functions and operations on callable objects."><code class="xref py py-mod docutils literal notranslate">functools</code></a> module
documentation has a quick link at the top labeled:
<blockquote>
<div>Source code <a class="reference external" href="https://github.com/python/cpython/tree/3.6/Lib/functools.py">Lib/functools.py</a>.
</div></blockquote>
(Contributed by Raymond Hettinger; see
<a class="reference external" href="https://rhettinger.wordpress.com/2011/01/28/open-your-source-more/">rationale</a>.)
</li>
<li>The docs now contain more examples and recipes. In particular, <a class="reference internal" href="../library/re.html#module-re" title="re: Regular expression operations."><code class="xref py py-mod docutils literal notranslate">re</code></a>
module has an extensive section, <a class="reference internal" href="../library/re.html#re-examples">Regular Expression Examples</a>. Likewise, the
<a class="reference internal" href="../library/itertools.html#module-itertools" title="itertools: Functions creating iterators for efficient looping."><code class="xref py py-mod docutils literal notranslate">itertools</code></a> module continues to be updated with new
<a class="reference internal" href="../library/itertools.html#itertools-recipes">Itertools Recipes</a>.
</li>
<li>The <a class="reference internal" href="../library/datetime.html#module-datetime" title="datetime: Basic date and time types."><code class="xref py py-mod docutils literal notranslate">datetime</code></a> module now has an auxiliary implementation in pure Python.
No functionality was changed. This just provides an easier-to-read alternate
implementation.
(Contributed by Alexander Belopolsky in <a class="reference external" href="https://bugs.python.org/issue9528">bpo-9528</a>.)
</li>
<li>The unmaintained <code class="file docutils literal notranslate">Demo</code> directory has been removed. Some demos were
integrated into the documentation, some were moved to the <code class="file docutils literal notranslate">Tools/demo</code>
directory, and others were removed altogether.
(Contributed by Georg Brandl in <a class="reference external" href="https://bugs.python.org/issue7962">bpo-7962</a>.)
</li>
</ul>
</div>
<div class="section" id="idle">
<h2>IDLE<a class="headerlink" href="#idle" title="Permalink to this headline">¶</a></h2>
<ul>
<li>The format menu now has an option to clean source files by stripping
trailing whitespace.
(Contributed by Raymond Hettinger; <a class="reference external" href="https://bugs.python.org/issue5150">bpo-5150</a>.)
</li>
<li>IDLE on Mac OS X now works with both Carbon AquaTk and Cocoa AquaTk.
(Contributed by Kevin Walzer, Ned Deily, and Ronald Oussoren; <a class="reference external" href="https://bugs.python.org/issue6075">bpo-6075</a>.)
</li>
</ul>
</div>
<div class="section" id="code-repository">
<h2>Code Repository<a class="headerlink" href="#code-repository" title="Permalink to this headline">¶</a></h2>
In addition to the existing Subversion code repository at <a class="reference external" href="http://svn.python.org">http://svn.python.org</a>
there is now a <a class="reference external" href="https://www.mercurial-scm.org/">Mercurial</a> repository at
<a class="reference external" href="https://hg.python.org/">https://hg.python.org/</a>.
After the 3.2 release, there are plans to switch to Mercurial as the primary
repository. This distributed version control system should make it easier for
members of the community to create and share external changesets. See
<a class="pep reference external" href="https://www.python.org/dev/peps/pep-0385">PEP 385</a> for details.
To learn to use the new version control system, see the <a class="reference external" href="http://hginit.com">tutorial by Joel
Spolsky</a> or the <a class="reference external" href="https://www.mercurial-scm.org/guide">Guide to Mercurial Workflows</a>.
</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>
Changes to Python’s build process and to the C API include:
<ul>
<li>The idle, pydoc and 2to3 scripts are now installed with a
version-specific suffix on <code class="docutils literal notranslate">make altinstall</code> (<a class="reference external" href="https://bugs.python.org/issue10679">bpo-10679</a>).
</li>
<li>The C functions that access the Unicode Database now accept and return
characters from the full Unicode range, even on narrow unicode builds
(Py_UNICODE_TOLOWER, Py_UNICODE_ISDECIMAL, and others). A visible difference
in Python is that <a class="reference internal" href="../library/unicodedata.html#unicodedata.numeric" title="unicodedata.numeric"><code class="xref py py-func docutils literal notranslate">unicodedata.numeric()</code></a> now returns the correct value
for large code points, and <a class="reference internal" href="../library/functions.html#repr" title="repr"><code class="xref py py-func docutils literal notranslate">repr()</code></a> may consider more characters as
printable.
(Reported by Bupjoe Lee and fixed by Amaury Forgeot D’Arc; <a class="reference external" href="https://bugs.python.org/issue5127">bpo-5127</a>.)
</li>
<li>Computed gotos are now enabled by default on supported compilers (which are
detected by the configure script). They can still be disabled selectively by
specifying <code class="docutils literal notranslate">--without-computed-gotos</code>.
(Contributed by Antoine Pitrou; <a class="reference external" href="https://bugs.python.org/issue9203">bpo-9203</a>.)
</li>
<li>The option <code class="docutils literal notranslate">--with-wctype-functions</code> was removed. The built-in unicode
database is now used for all functions.
(Contributed by Amaury Forgeot D’Arc; <a class="reference external" href="https://bugs.python.org/issue9210">bpo-9210</a>.)
</li>
<li>Hash values are now values of a new type, <code class="xref c c-type docutils literal notranslate">Py_hash_t</code>, which is
defined to be the same size as a pointer. Previously they were of type long,
which on some 64-bit operating systems is still only 32 bits long. As a
result of this fix, <a class="reference internal" href="../library/stdtypes.html#set" title="set"><code class="xref py py-class docutils literal notranslate">set</code></a> and <a class="reference internal" href="../library/stdtypes.html#dict" title="dict"><code class="xref py py-class docutils literal notranslate">dict</code></a> can now hold more than
<code class="docutils literal notranslate">2**32</code> entries on builds with 64-bit pointers (previously, they could grow
to that size but their performance degraded catastrophically).
(Suggested by Raymond Hettinger and implemented by Benjamin Peterson;
<a class="reference external" href="https://bugs.python.org/issue9778">bpo-9778</a>.)
</li>
<li>A new macro <code class="xref c c-macro docutils literal notranslate">Py_VA_COPY</code> copies the state of the variable argument
list. It is equivalent to C99 va_copy but available on all Python platforms
(<a class="reference external" href="https://bugs.python.org/issue2443">bpo-2443</a>).
</li>
<li>A new C API function <a class="reference internal" href="../c-api/init.html#c.PySys_SetArgvEx" title="PySys_SetArgvEx"><code class="xref c c-func docutils literal notranslate">PySys_SetArgvEx()</code></a> allows an embedded interpreter
to set <a class="reference internal" href="../library/sys.html#sys.argv" title="sys.argv"><code class="xref py py-attr docutils literal notranslate">sys.argv</code></a> without also modifying <a class="reference internal" href="../library/sys.html#sys.path" title="sys.path"><code class="xref py py-attr docutils literal notranslate">sys.path</code></a>
(<a class="reference external" href="https://bugs.python.org/issue5753">bpo-5753</a>).
</li>
<li><code class="xref c c-macro docutils literal notranslate">PyEval_CallObject</code> is now only available in macro form. The
function declaration, which was kept for backwards compatibility reasons, is
now removed – the macro was introduced in 1997 (<a class="reference external" href="https://bugs.python.org/issue8276">bpo-8276</a>).
</li>
<li>There is a new function <a class="reference internal" href="../c-api/long.html#c.PyLong_AsLongLongAndOverflow" title="PyLong_AsLongLongAndOverflow"><code class="xref c c-func docutils literal notranslate">PyLong_AsLongLongAndOverflow()</code></a> which
is analogous to <a class="reference internal" href="../c-api/long.html#c.PyLong_AsLongAndOverflow" title="PyLong_AsLongAndOverflow"><code class="xref c c-func docutils literal notranslate">PyLong_AsLongAndOverflow()</code></a>. They both serve to
convert Python <a class="reference internal" href="../library/functions.html#int" title="int"><code class="xref py py-class docutils literal notranslate">int</code></a> into a native fixed-width type while providing
detection of cases where the conversion won’t fit (<a class="reference external" href="https://bugs.python.org/issue7767">bpo-7767</a>).
</li>
<li>The <a class="reference internal" href="../c-api/unicode.html#c.PyUnicode_CompareWithASCIIString" title="PyUnicode_CompareWithASCIIString"><code class="xref c c-func docutils literal notranslate">PyUnicode_CompareWithASCIIString()</code></a> function now returns not
equal if the Python string is NUL terminated.
</li>
<li>There is a new function <a class="reference internal" href="../c-api/exceptions.html#c.PyErr_NewExceptionWithDoc" title="PyErr_NewExceptionWithDoc"><code class="xref c c-func docutils literal notranslate">PyErr_NewExceptionWithDoc()</code></a> that is
like <a class="reference internal" href="../c-api/exceptions.html#c.PyErr_NewException" title="PyErr_NewException"><code class="xref c c-func docutils literal notranslate">PyErr_NewException()</code></a> but allows a docstring to be specified.
This lets C exceptions have the same self-documenting capabilities as
their pure Python counterparts (<a class="reference external" href="https://bugs.python.org/issue7033">bpo-7033</a>).
</li>
<li>When compiled with the <code class="docutils literal notranslate">--with-valgrind</code> option, the pymalloc
allocator will be automatically disabled when running under Valgrind. This
gives improved memory leak detection when running under Valgrind, while taking
advantage of pymalloc at other times (<a class="reference external" href="https://bugs.python.org/issue2422">bpo-2422</a>).
</li>
<li>Removed the <code class="docutils literal notranslate">O?</code> format from the PyArg_Parse functions. The format is no
longer used and it had never been documented (<a class="reference external" href="https://bugs.python.org/issue8837">bpo-8837</a>).
</li>
</ul>
There were a number of other small changes to the C-API. See the
<a class="reference external" href="https://github.com/python/cpython/tree/3.6/Misc/NEWS">Misc/NEWS</a> file for a complete list.
Also, there were a number of updates to the Mac OS X build, see
<a class="reference external" href="https://github.com/python/cpython/tree/3.6/Mac/BuildScript/README.txt">Mac/BuildScript/README.txt</a> for details. For users running a 32/64-bit
build, there is a known problem with the default Tcl/Tk on Mac OS X 10.6.
Accordingly, we recommend installing an updated alternative such as
<a class="reference external" href="https://www.activestate.com/activetcl/downloads">ActiveState Tcl/Tk 8.5.9</a>.
See <a class="reference external" href="https://www.python.org/download/mac/tcltk/">https://www.python.org/download/mac/tcltk/</a> for additional details.
</div>
<div class="section" id="porting-to-python-3-2">
<h2>Porting to Python 3.2<a class="headerlink" href="#porting-to-python-3-2" title="Permalink to this headline">¶</a></h2>
This section lists previously described changes and other bugfixes that may
require changes to your code:
<ul>
<li>The <a class="reference internal" href="../library/configparser.html#module-configparser" title="configparser: Configuration file parser."><code class="xref py py-mod docutils literal notranslate">configparser</code></a> module has a number of clean-ups. The major change is
to replace the old <code class="xref py py-class docutils literal notranslate">ConfigParser</code> class with long-standing preferred
alternative <code class="xref py py-class docutils literal notranslate">SafeConfigParser</code>. In addition there are a number of
smaller incompatibilities:
<ul class="simple">
<li>The interpolation syntax is now validated on
<a class="reference internal" href="../library/configparser.html#configparser.ConfigParser.get" title="configparser.ConfigParser.get"><code class="xref py py-meth docutils literal notranslate">get()</code></a> and
<a class="reference internal" href="../library/configparser.html#configparser.ConfigParser.set" title="configparser.ConfigParser.set"><code class="xref py py-meth docutils literal notranslate">set()</code></a> operations. In the default
interpolation scheme, only two tokens with percent signs are valid: <code class="docutils literal notranslate">%(name)s</code>
and <code class="docutils literal notranslate">%%</code>, the latter being an escaped percent sign.</li>
<li>The <a class="reference internal" href="../library/configparser.html#configparser.ConfigParser.set" title="configparser.ConfigParser.set"><code class="xref py py-meth docutils literal notranslate">set()</code></a> and
<a class="reference internal" href="../library/configparser.html#configparser.ConfigParser.add_section" title="configparser.ConfigParser.add_section"><code class="xref py py-meth docutils literal notranslate">add_section()</code></a> methods now verify that
values are actual strings. Formerly, unsupported types could be introduced
unintentionally.</li>
<li>Duplicate sections or options from a single source now raise either
<a class="reference internal" href="../library/configparser.html#configparser.DuplicateSectionError" title="configparser.DuplicateSectionError"><code class="xref py py-exc docutils literal notranslate">DuplicateSectionError</code></a> or
<a class="reference internal" href="../library/configparser.html#configparser.DuplicateOptionError" title="configparser.DuplicateOptionError"><code class="xref py py-exc docutils literal notranslate">DuplicateOptionError</code></a>. Formerly, duplicates would
silently overwrite a previous entry.</li>
<li>Inline comments are now disabled by default so now the ; character
can be safely used in values.</li>
<li>Comments now can be indented. Consequently, for ; or # to appear at
the start of a line in multiline values, it has to be interpolated. This
keeps comment prefix characters in values from being mistaken as comments.</li>
<li><code class="docutils literal notranslate">&quot;&quot;</code> is now a valid value and is no longer automatically converted to an
empty string. For empty strings, use <code class="docutils literal notranslate">&quot;option =&quot;</code> in a line.</li>
</ul>
</li>
<li>The <a class="reference internal" href="../library/nntplib.html#module-nntplib" title="nntplib: NNTP protocol client (requires sockets)."><code class="xref py py-mod docutils literal notranslate">nntplib</code></a> module was reworked extensively, meaning that its APIs
are often incompatible with the 3.1 APIs.
</li>
<li><a class="reference internal" href="../library/stdtypes.html#bytearray" title="bytearray"><code class="xref py py-class docutils literal notranslate">bytearray</code></a> objects can no longer be used as filenames; instead,
they should be converted to <a class="reference internal" href="../library/stdtypes.html#bytes" title="bytes"><code class="xref py py-class docutils literal notranslate">bytes</code></a>.
</li>
<li>The <code class="xref py py-meth docutils literal notranslate">array.tostring()</code> and <code class="xref py py-meth docutils literal notranslate">array.fromstring()</code> have been renamed to
<code class="xref py py-meth docutils literal notranslate">array.tobytes()</code> and <code class="xref py py-meth docutils literal notranslate">array.frombytes()</code> for clarity. The old names
have been deprecated. (See <a class="reference external" href="https://bugs.python.org/issue8990">bpo-8990</a>.)
</li>
<li><code class="docutils literal notranslate">PyArg_Parse*()</code> functions:
<ul class="simple">
<li>“t#” format has been removed: use “s#” or “s*” instead</li>
<li>“w” and “w#” formats has been removed: use “w*” instead</li>
</ul>
</li>
<li>The <code class="xref c c-type docutils literal notranslate">PyCObject</code> type, deprecated in 3.1, has been removed. To wrap
opaque C pointers in Python objects, the <a class="reference internal" href="../c-api/capsule.html#c.PyCapsule" title="PyCapsule"><code class="xref c c-type docutils literal notranslate">PyCapsule</code></a> API should be used
instead; the new type has a well-defined interface for passing typing safety
information and a less complicated signature for calling a destructor.
</li>
<li>The <code class="xref py py-func docutils literal notranslate">sys.setfilesystemencoding()</code> function was removed because
it had a flawed design.
</li>
<li>The <a class="reference internal" href="../library/random.html#random.seed" title="random.seed"><code class="xref py py-func docutils literal notranslate">random.seed()</code></a> function and method now salt string seeds with an
sha512 hash function. To access the previous version of seed in order to
reproduce Python 3.1 sequences, set the version argument to 1,
<code class="docutils literal notranslate">random.seed(s, version=1)</code>.
</li>
<li>The previously deprecated <code class="xref py py-func docutils literal notranslate">string.maketrans()</code> function has been removed
in favor of the static methods <a class="reference internal" href="../library/stdtypes.html#bytes.maketrans" title="bytes.maketrans"><code class="xref py py-meth docutils literal notranslate">bytes.maketrans()</code></a> and
<a class="reference internal" href="../library/stdtypes.html#bytearray.maketrans" title="bytearray.maketrans"><code class="xref py py-meth docutils literal notranslate">bytearray.maketrans()</code></a>. This change solves the confusion around which
types were supported by the <a class="reference internal" href="../library/string.html#module-string" title="string: Common string operations."><code class="xref py py-mod docutils literal notranslate">string</code></a> module. Now, <a class="reference internal" href="../library/stdtypes.html#str" title="str"><code class="xref py py-class docutils literal notranslate">str</code></a>,
<a class="reference internal" href="../library/stdtypes.html#bytes" title="bytes"><code class="xref py py-class docutils literal notranslate">bytes</code></a>, and <a class="reference internal" href="../library/stdtypes.html#bytearray" title="bytearray"><code class="xref py py-class docutils literal notranslate">bytearray</code></a> each have their own maketrans and
translate methods with intermediate translation tables of the appropriate
type.
(Contributed by Georg Brandl; <a class="reference external" href="https://bugs.python.org/issue5675">bpo-5675</a>.)
</li>
<li>The previously deprecated <code class="xref py py-func docutils literal notranslate">contextlib.nested()</code> function has been removed
in favor of a plain <a class="reference internal" href="../reference/compound_stmts.html#with"><code class="xref std std-keyword docutils literal notranslate">with</code></a> statement which can accept multiple
context managers. The latter technique is faster (because it is built-in),
and it does a better job finalizing multiple context managers when one of them
raises an exception:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>with open(&#39;mylog.txt&#39;) as infile, open(&#39;a.out&#39;, &#39;w&#39;) as outfile:
 for line in infile:
 if &#39;&lt;critical&gt;&#39; in line:
 outfile.write(line)
</pre></div>
</div>
(Contributed by Georg Brandl and Mattias Brändström;
<a class="reference external" href="https://codereview.appspot.com/53094">appspot issue 53094</a>.)
</li>
<li><a class="reference internal" href="../library/struct.html#struct.pack" title="struct.pack"><code class="xref py py-func docutils literal notranslate">struct.pack()</code></a> now only allows bytes for the <code class="docutils literal notranslate">s</code> string pack code.
Formerly, it would accept text arguments and implicitly encode them to bytes
using UTF-8. This was problematic because it made assumptions about the
correct encoding and because a variable-length encoding can fail when writing
to fixed length segment of a structure.
Code such as <code class="docutils literal notranslate">struct.pack('&lt;6sHHBBB', 'GIF87a', x, y)</code> should be rewritten
with to use bytes instead of text, <code class="docutils literal notranslate">struct.pack('&lt;6sHHBBB', b'GIF87a', x, y)</code>.
(Discovered by David Beazley and fixed by Victor Stinner; <a class="reference external" href="https://bugs.python.org/issue10783">bpo-10783</a>.)
</li>
<li>The <a class="reference internal" href="../library/xml.etree.elementtree.html#module-xml.etree.ElementTree" title="xml.etree.ElementTree: Implementation of the ElementTree API."><code class="xref py py-class docutils literal notranslate">xml.etree.ElementTree</code></a> class now raises an
<a class="reference internal" href="../library/xml.etree.elementtree.html#xml.etree.ElementTree.ParseError" title="xml.etree.ElementTree.ParseError"><code class="xref py py-exc docutils literal notranslate">xml.etree.ElementTree.ParseError</code></a> when a parse fails. Previously it
raised an <a class="reference internal" href="../library/pyexpat.html#xml.parsers.expat.ExpatError" title="xml.parsers.expat.ExpatError"><code class="xref py py-exc docutils literal notranslate">xml.parsers.expat.ExpatError</code></a>.
</li>
<li>The new, longer <a class="reference internal" href="../library/stdtypes.html#str" title="str"><code class="xref py py-func docutils literal notranslate">str()</code></a> value on floats may break doctests which rely on
the old output format.
</li>
<li>In <a class="reference internal" href="../library/subprocess.html#subprocess.Popen" title="subprocess.Popen"><code class="xref py py-class docutils literal notranslate">subprocess.Popen</code></a>, the default value for close_fds is now
<code class="docutils literal notranslate">True</code> under Unix; under Windows, it is <code class="docutils literal notranslate">True</code> if the three standard
streams are set to <code class="docutils literal notranslate">None</code>, <code class="docutils literal notranslate">False</code> otherwise. Previously, close_fds
was always <code class="docutils literal notranslate">False</code> by default, which produced difficult to solve bugs
or race conditions when open file descriptors would leak into the child
process.
</li>
<li>Support for legacy HTTP 0.9 has been removed from <a class="reference internal" href="../library/urllib.request.html#module-urllib.request" title="urllib.request: Extensible library for opening URLs."><code class="xref py py-mod docutils literal notranslate">urllib.request</code></a>
and <a class="reference internal" href="../library/http.client.html#module-http.client" title="http.client: HTTP and HTTPS protocol client (requires sockets)."><code class="xref py py-mod docutils literal notranslate">http.client</code></a>. Such support is still present on the server side
(in <a class="reference internal" href="../library/http.server.html#module-http.server" title="http.server: HTTP server and request handlers."><code class="xref py py-mod docutils literal notranslate">http.server</code></a>).
(Contributed by Antoine Pitrou, <a class="reference external" href="https://bugs.python.org/issue10711">bpo-10711</a>.)
</li>
<li>SSL sockets in timeout mode now raise <a class="reference internal" href="../library/socket.html#socket.timeout" title="socket.timeout"><code class="xref py py-exc docutils literal notranslate">socket.timeout</code></a> when a timeout
occurs, rather than a generic <a class="reference internal" href="../library/ssl.html#ssl.SSLError" title="ssl.SSLError"><code class="xref py py-exc docutils literal notranslate">SSLError</code></a>.
(Contributed by Antoine Pitrou, <a class="reference external" href="https://bugs.python.org/issue10272">bpo-10272</a>.)
</li>
<li>The misleading functions <a class="reference internal" href="../c-api/init.html#c.PyEval_AcquireLock" title="PyEval_AcquireLock"><code class="xref c c-func docutils literal notranslate">PyEval_AcquireLock()</code></a> and
<a class="reference internal" href="../c-api/init.html#c.PyEval_ReleaseLock" title="PyEval_ReleaseLock"><code class="xref c c-func docutils literal notranslate">PyEval_ReleaseLock()</code></a> have been officially deprecated. The
thread-state aware APIs (such as <a class="reference internal" href="../c-api/init.html#c.PyEval_SaveThread" title="PyEval_SaveThread"><code class="xref c c-func docutils literal notranslate">PyEval_SaveThread()</code></a>
and <a class="reference internal" href="../c-api/init.html#c.PyEval_RestoreThread" title="PyEval_RestoreThread"><code class="xref c c-func docutils literal notranslate">PyEval_RestoreThread()</code></a>) should be used instead.
</li>
<li>Due to security risks, <code class="xref py py-func docutils literal notranslate">asyncore.handle_accept()</code> has been deprecated, and
a new function, <code class="xref py py-func docutils literal notranslate">asyncore.handle_accepted()</code>, was added to replace it.
(Contributed by Giampaolo Rodola in <a class="reference external" href="https://bugs.python.org/issue6706">bpo-6706</a>.)
</li>
<li>Due to the new <a class="reference internal" href="../glossary.html#term-gil">GIL</a> implementation, <a class="reference internal" href="../c-api/init.html#c.PyEval_InitThreads" title="PyEval_InitThreads"><code class="xref c c-func docutils literal notranslate">PyEval_InitThreads()</code></a>
cannot be called before <a class="reference internal" href="../c-api/init.html#c.Py_Initialize" title="Py_Initialize"><code class="xref c c-func docutils literal notranslate">Py_Initialize()</code></a> anymore.
</li>
</ul>
</div>
</div>

</div>
 </div>
 </div>
 <div class="sphinxsidebar" role="navigation" aria-label="main navigation">
 <div class="sphinxsidebarwrapper">
 <h3><a href="../contents.html">Table Of Contents</a></h3>
 <ul>
<li><a class="reference internal" href="#">What’s New In Python 3.2</a><ul>
<li><a class="reference internal" href="#pep-384-defining-a-stable-abi">PEP 384: Defining a Stable ABI</a></li>
<li><a class="reference internal" href="#pep-389-argparse-command-line-parsing-module">PEP 389: Argparse Command Line Parsing Module</a></li>
<li><a class="reference internal" href="#pep-391-dictionary-based-configuration-for-logging">PEP 391: Dictionary Based Configuration for Logging</a></li>
<li><a class="reference internal" href="#pep-3148-the-concurrent-futures-module">PEP 3148: The <code class="docutils literal notranslate">concurrent.futures</code> module</a></li>
<li><a class="reference internal" href="#pep-3147-pyc-repository-directories">PEP 3147: PYC Repository Directories</a></li>
<li><a class="reference internal" href="#pep-3149-abi-version-tagged-so-files">PEP 3149: ABI Version Tagged .so Files</a></li>
<li><a class="reference internal" href="#pep-3333-python-web-server-gateway-interface-v1-0-1">PEP 3333: Python Web Server Gateway Interface v1.0.1</a></li>
<li><a class="reference internal" href="#other-language-changes">Other Language Changes</a></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="#email">email</a></li>
<li><a class="reference internal" href="#elementtree">elementtree</a></li>
<li><a class="reference internal" href="#functools">functools</a></li>
<li><a class="reference internal" href="#itertools">itertools</a></li>
<li><a class="reference internal" href="#collections">collections</a></li>
<li><a class="reference internal" href="#threading">threading</a></li>
<li><a class="reference internal" href="#datetime-and-time">datetime and time</a></li>
<li><a class="reference internal" href="#math">math</a></li>
<li><a class="reference internal" href="#abc">abc</a></li>
<li><a class="reference internal" href="#io">io</a></li>
<li><a class="reference internal" href="#reprlib">reprlib</a></li>
<li><a class="reference internal" href="#logging">logging</a></li>
<li><a class="reference internal" href="#csv">csv</a></li>
<li><a class="reference internal" href="#contextlib">contextlib</a></li>
<li><a class="reference internal" href="#decimal-and-fractions">decimal and fractions</a></li>
<li><a class="reference internal" href="#ftp">ftp</a></li>
<li><a class="reference internal" href="#popen">popen</a></li>
<li><a class="reference internal" href="#select">select</a></li>
<li><a class="reference internal" href="#gzip-and-zipfile">gzip and zipfile</a></li>
<li><a class="reference internal" href="#tarfile">tarfile</a></li>
<li><a class="reference internal" href="#hashlib">hashlib</a></li>
<li><a class="reference internal" href="#ast">ast</a></li>
<li><a class="reference internal" href="#os">os</a></li>
<li><a class="reference internal" href="#shutil">shutil</a></li>
<li><a class="reference internal" href="#sqlite3">sqlite3</a></li>
<li><a class="reference internal" href="#html">html</a></li>
<li><a class="reference internal" href="#socket">socket</a></li>
<li><a class="reference internal" href="#ssl">ssl</a></li>
<li><a class="reference internal" href="#nntp">nntp</a></li>
<li><a class="reference internal" href="#certificates">certificates</a></li>
<li><a class="reference internal" href="#imaplib">imaplib</a></li>
<li><a class="reference internal" href="#http-client">http.client</a></li>
<li><a class="reference internal" href="#unittest">unittest</a></li>
<li><a class="reference internal" href="#random">random</a></li>
<li><a class="reference internal" href="#poplib">poplib</a></li>
<li><a class="reference internal" href="#asyncore">asyncore</a></li>
<li><a class="reference internal" href="#tempfile">tempfile</a></li>
<li><a class="reference internal" href="#inspect">inspect</a></li>
<li><a class="reference internal" href="#pydoc">pydoc</a></li>
<li><a class="reference internal" href="#dis">dis</a></li>
<li><a class="reference internal" href="#dbm">dbm</a></li>
<li><a class="reference internal" href="#ctypes">ctypes</a></li>
<li><a class="reference internal" href="#site">site</a></li>
<li><a class="reference internal" href="#sysconfig">sysconfig</a></li>
<li><a class="reference internal" href="#pdb">pdb</a></li>
<li><a class="reference internal" href="#configparser">configparser</a></li>
<li><a class="reference internal" href="#urllib-parse">urllib.parse</a></li>
<li><a class="reference internal" href="#mailbox">mailbox</a></li>
<li><a class="reference internal" href="#turtledemo">turtledemo</a></li>
</ul>
</li>
<li><a class="reference internal" href="#multi-threading">Multi-threading</a></li>
<li><a class="reference internal" href="#optimizations">Optimizations</a></li>
<li><a class="reference internal" href="#unicode">Unicode</a></li>
<li><a class="reference internal" href="#codecs">Codecs</a></li>
<li><a class="reference internal" href="#documentation">Documentation</a></li>
<li><a class="reference internal" href="#idle">IDLE</a></li>
<li><a class="reference internal" href="#code-repository">Code Repository</a></li>
<li><a class="reference internal" href="#build-and-c-api-changes">Build and C API Changes</a></li>
<li><a class="reference internal" href="#porting-to-python-3-2">Porting to Python 3.2</a></li>
</ul>
</li>
</ul>

<h4>Previous topic</h4>
 <a href="3.3.html"
 title="previous chapter">What’s New In Python 3.3</a>
 <h4>Next topic</h4>
 <a href="3.1.html"
 title="next chapter">What’s New In Python 3.1</a>
 <div role="note" aria-label="source link">
 <h3>This Page</h3>
 <ul class="this-page-menu">
 <li><a href="../bugs.html">Report a Bug</a></li>
 <li>
 <a href="https://github.com/python/cpython/blob/3.6/Doc/whatsnew/3.2.rst"
 rel="nofollow">Show Source
 </a>
 </li>
 </ul>
 </div>
 </div>
 </div>
 <div class="clearer"></div>
 </div> 
 <div class="related" role="navigation" aria-label="related navigation">
 <h3>Navigation</h3>
 <ul>
 <li class="right" style="margin-right: 10px">
 <a href="../genindex.html" title="General Index"
 >index</a></li>
 <li class="right" >
 <a href="../py-modindex.html" title="Python Module Index"
 >modules</a> |</li>
 <li class="right" >
 <a href="3.1.html" title="What’s New In Python 3.1"
 >next</a> |</li>
 <li class="right" >
 <a href="3.3.html" title="What’s New In Python 3.3"
 >previous</a> |</li>
 <li><img src="../_static/py.png" alt=""
 style="vertical-align: middle; margin-top: -1px"/></li>
 <li><a href="https://www.python.org/">Python</a> &#187;</li>
 <li>
 <a href="../index.html">3.6.7 Documentation</a> &#187;
 </li>

<li class="nav-item nav-item-1"><a href="index.html" >What’s New in Python</a> &#187;</li>
 <li class="right">

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

</body>
</html>

MR SKB