CINXE.COM

<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"> <html xmlns="http://www.w3.org/1999/xhtml"> <head> <meta http-equiv="Content-Type" content="text/html; charset=utf-8" /> <title>What鈥檚 New in Python 2.6 — Python v3.1.5 documentation</title> <link rel="stylesheet" href="../_static/default.css" type="text/css" /> <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> <script type="text/javascript"> var DOCUMENTATION_OPTIONS = { URL_ROOT: '../', VERSION: '3.1.5', COLLAPSE_MODINDEX: false, FILE_SUFFIX: '.html', HAS_SOURCE: true }; </script> <script type="text/javascript" src="../_static/jquery.js"></script> <script type="text/javascript" src="../_static/doctools.js"></script> <link rel="search" type="application/opensearchdescription+xml" title="Search within Python v3.1.5 documentation" href="../_static/opensearch.xml"/> <link rel="author" title="About these documents" href="../about.html" /> <link rel="copyright" title="Copyright" href="../copyright.html" /> <link rel="top" title="Python v3.1.5 documentation" href="../index.html" /> <link rel="up" title="What鈥檚 New in Python" href="index.html" /> <link rel="next" title="What鈥檚 New in Python 2.5" href="2.5.html" /> <link rel="prev" title="What鈥檚 New In Python 3.0" href="3.0.html" /> <link rel="shortcut icon" type="image/png" href="../_static/py.png" /> <link rel="canonical" href="https://docs.python.org/3/whatsnew/2.6.html" /> </head> <body> <div class="related"> <h3>Navigation</h3> <ul> <li class="right" style="margin-right: 10px"> <a href="../genindex.html" title="General Index" accesskey="I">index</a></li> <li class="right" > <a href="../modindex.html" title="Global Module Index" accesskey="M">modules</a> |</li> <li class="right" > <a href="2.5.html" title="What鈥檚 New in Python 2.5" accesskey="N">next</a> |</li> <li class="right" > <a href="3.0.html" title="What鈥檚 New In Python 3.0" accesskey="P">previous</a> |</li> <li><img src="../_static/py.png" alt="" style="vertical-align: middle; margin-top: -1px"/></li> <li><a href="../index.html">Python v3.1.5 documentation</a> »</li> <li><a href="index.html" accesskey="U">What’s New in Python</a> »</li> </ul> </div> <div class="document"> <div class="documentwrapper"> <div class="bodywrapper"> <div class="body"> <div class="section" id="what-s-new-in-python-2-6"> <h1>What’s New in Python 2.6<a class="headerlink" href="#what-s-new-in-python-2-6" title="Permalink to this headline">露</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"><th class="field-name">Author:</th><td class="field-body">A.M. Kuchling (amk at amk.ca)</td> </tr> <tr class="field"><th class="field-name">Release:</th><td class="field-body">3.1.5</td> </tr> <tr class="field"><th class="field-name">Date:</th><td class="field-body">April 09, 2012</td> </tr> </tbody> </table> This article explains the new features in Python 2.6, released on October 1 2008. The release schedule is described in <a class="reference external" href="http://www.python.org/dev/peps/pep-0361">PEP 361</a>. The major theme of Python 2.6 is preparing the migration path to Python 3.0, a major redesign of the language. Whenever possible, Python 2.6 incorporates new features and syntax from 3.0 while remaining compatible with existing code by not removing older features or syntax. When it’s not possible to do that, Python 2.6 tries to do what it can, adding compatibility functions in a <tt class="xref docutils literal">future_builtins</tt> module and a -3 switch to warn about usages that will become unsupported in 3.0. Some significant new packages have been added to the standard library, such as the <a title="Process-based parallelism." class="reference external" href="../library/multiprocessing.html#module-multiprocessing"><tt class="xref docutils literal">multiprocessing</tt></a> and <a title="Encode and decode the JSON format." class="reference external" href="../library/json.html#module-json"><tt class="xref docutils literal">json</tt></a> modules, but there aren’t many new features that aren’t related to Python 3.0 in some way. Python 2.6 also sees a number of improvements and bugfixes throughout the source. A search through the change logs finds there were 259 patches applied and 612 bugs fixed between Python 2.5 and 2.6. Both figures are likely to be underestimates. This article doesn’t attempt to provide a complete specification of the new features, but instead provides a convenient overview. For full details, you should refer to the documentation for Python 2.6. If you want to understand the rationale for the design and implementation, refer to the PEP for a particular new feature. Whenever possible, “What’s New in Python” links to the bug/patch item for each change. <div class="section" id="python-3-0"> <h2>Python 3.0<a class="headerlink" href="#python-3-0" title="Permalink to this headline">露</a></h2> The development cycle for Python versions 2.6 and 3.0 was synchronized, with the alpha and beta releases for both versions being made on the same days. The development of 3.0 has influenced many features in 2.6. Python 3.0 is a far-ranging redesign of Python that breaks compatibility with the 2.x series. This means that existing Python code will need some conversion in order to run on Python 3.0. However, not all the changes in 3.0 necessarily break compatibility. In cases where new features won’t cause existing code to break, they’ve been backported to 2.6 and are described in this document in the appropriate place. Some of the 3.0-derived features are: <ul class="simple"> <li>A <a title="object.__complex__" class="reference external" href="../reference/datamodel.html#object.__complex__"><tt class="xref docutils literal">__complex__()</tt></a> method for converting objects to a complex number.</li> <li>Alternate syntax for catching exceptions: <tt class="docutils literal">except TypeError as exc</tt>.</li> <li>The addition of <a title="functools.reduce" class="reference external" href="../library/functools.html#functools.reduce"><tt class="xref docutils literal">functools.reduce()</tt></a> as a synonym for the built-in <tt class="xref docutils literal">reduce()</tt> function.</li> </ul> Python 3.0 adds several new built-in functions and changes the semantics of some existing builtins. Functions that are new in 3.0 such as <a title="bin" class="reference external" href="../library/functions.html#bin"><tt class="xref docutils literal">bin()</tt></a> have simply been added to Python 2.6, but existing builtins haven’t been changed; instead, the <tt class="xref docutils literal">future_builtins</tt> module has versions with the new 3.0 semantics. Code written to be compatible with 3.0 can do <tt class="docutils literal">from future_builtins import hex, map</tt> as necessary. A new command-line switch, -3, enables warnings about features that will be removed in Python 3.0. You can run code with this switch to see how much work will be necessary to port code to 3.0. The value of this switch is available to Python code as the boolean variable <tt class="xref docutils literal">sys.py3kwarning</tt>, and to C extension code as <tt class="xref docutils literal">Py_Py3kWarningFlag</tt>. <div class="admonition-see-also admonition seealso"> See also The 3xxx series of PEPs, which contains proposals for Python 3.0. <a class="reference external" href="http://www.python.org/dev/peps/pep-3000">PEP 3000</a> describes the development process for Python 3.0. Start with <a class="reference external" href="http://www.python.org/dev/peps/pep-3100">PEP 3100</a> that describes the general goals for Python 3.0, and then explore the higher-numbered PEPS that propose specific features. </div> </div> <div class="section" id="changes-to-the-development-process"> <h2>Changes to the Development Process<a class="headerlink" href="#changes-to-the-development-process" title="Permalink to this headline">露</a></h2> While 2.6 was being developed, the Python development process underwent two significant changes: we switched from SourceForge’s issue tracker to a customized Roundup installation, and the documentation was converted from LaTeX to reStructuredText. <div class="section" id="new-issue-tracker-roundup"> <h3>New Issue Tracker: Roundup<a class="headerlink" href="#new-issue-tracker-roundup" title="Permalink to this headline">露</a></h3> For a long time, the Python developers had been growing increasingly annoyed by SourceForge’s bug tracker. SourceForge’s hosted solution doesn’t permit much customization; for example, it wasn’t possible to customize the life cycle of issues. The infrastructure committee of the Python Software Foundation therefore posted a call for issue trackers, asking volunteers to set up different products and import some of the bugs and patches from SourceForge. Four different trackers were examined: <a class="reference external" href="http://www.atlassian.com/software/jira/">Jira</a>, <a class="reference external" href="http://www.launchpad.net">Launchpad</a>, <a class="reference external" href="http://roundup.sourceforge.net/">Roundup</a>, and <a class="reference external" href="http://trac.edgewall.org/">Trac</a>. The committee eventually settled on Jira and Roundup as the two candidates. Jira is a commercial product that offers no-cost hosted instances to free-software projects; Roundup is an open-source project that requires volunteers to administer it and a server to host it. After posting a call for volunteers, a new Roundup installation was set up at <a class="reference external" href="http://bugs.python.org">http://bugs.python.org</a>. One installation of Roundup can host multiple trackers, and this server now also hosts issue trackers for Jython and for the Python web site. It will surely find other uses in the future. Where possible, this edition of “What’s New in Python” links to the bug/patch item for each change. Hosting of the Python bug tracker is kindly provided by <a class="reference external" href="http://www.upfrontsystems.co.za/">Upfront Systems</a> of Stellenbosch, South Africa. Martin von Loewis put a lot of effort into importing existing bugs and patches from SourceForge; his scripts for this import operation are at <a class="reference external" href="http://svn.python.org/view/tracker/importer/">http://svn.python.org/view/tracker/importer/</a> and may be useful to other projects wishing to move from SourceForge to Roundup. <div class="admonition-see-also admonition seealso"> See also <dl class="last docutils"> <dt><a class="reference external" href="http://bugs.python.org">http://bugs.python.org</a></dt> <dd>The Python bug tracker.</dd> <dt><a class="reference external" href="http://bugs.jython.org">http://bugs.jython.org</a>:</dt> <dd>The Jython bug tracker.</dd> <dt><a class="reference external" href="http://roundup.sourceforge.net/">http://roundup.sourceforge.net/</a></dt> <dd>Roundup downloads and documentation.</dd> <dt><a class="reference external" href="http://svn.python.org/view/tracker/importer/">http://svn.python.org/view/tracker/importer/</a></dt> <dd>Martin von Loewis’s conversion scripts.</dd> </dl> </div> </div> <div class="section" id="new-documentation-format-restructuredtext-using-sphinx"> <h3>New Documentation Format: reStructuredText Using Sphinx<a class="headerlink" href="#new-documentation-format-restructuredtext-using-sphinx" title="Permalink to this headline">露</a></h3> The Python documentation was written using LaTeX since the project started around 1989. In the 1980s and early 1990s, most documentation was printed out for later study, not viewed online. LaTeX was widely used because it provided attractive printed output while remaining straightforward to write once the basic rules of the markup were learned. Today LaTeX is still used for writing publications destined for printing, but the landscape for programming tools has shifted. We no longer print out reams of documentation; instead, we browse through it online and HTML has become the most important format to support. Unfortunately, converting LaTeX to HTML is fairly complicated and Fred L. Drake Jr., the long-time Python documentation editor, spent a lot of time maintaining the conversion process. Occasionally people would suggest converting the documentation into SGML and later XML, but performing a good conversion is a major task and no one ever committed the time required to finish the job. During the 2.6 development cycle, Georg Brandl put a lot of effort into building a new toolchain for processing the documentation. The resulting package is called Sphinx, and is available from <a class="reference external" href="http://sphinx.pocoo.org/">http://sphinx.pocoo.org/</a>. Sphinx concentrates on HTML output, producing attractively styled and modern HTML; printed output is still supported through conversion to LaTeX. The input format is reStructuredText, a markup syntax supporting custom extensions and directives that is commonly used in the Python community. Sphinx is a standalone package that can be used for writing, and almost two dozen other projects (<a class="reference external" href="http://sphinx.pocoo.org/examples.html">listed on the Sphinx web site</a>) have adopted Sphinx as their documentation tool. <div class="admonition-see-also admonition seealso"> See also <dl class="last docutils"> <dt><a class="reference external" href="../documenting/index.html#documenting-index">Documenting Python</a></dt> <dd>Describes how to write for Python’s documentation.</dd> <dt><a class="reference external" href="http://sphinx.pocoo.org/">Sphinx</a></dt> <dd>Documentation and code for the Sphinx toolchain.</dd> <dt><a class="reference external" href="http://docutils.sf.net">Docutils</a></dt> <dd>The underlying reStructuredText parser and toolset.</dd> </dl> </div> </div> </div> <div class="section" id="pep-343-the-with-statement"> <h2>PEP 343: The ‘with’ statement<a class="headerlink" href="#pep-343-the-with-statement" title="Permalink to this headline">露</a></h2> The previous version, Python 2.5, added the ‘<a class="reference external" href="../reference/compound_stmts.html#with"><tt class="xref docutils literal">with</tt></a>‘ statement as an optional feature, to be enabled by a <tt class="docutils literal">from __future__ import with_statement</tt> directive. In 2.6 the statement no longer needs to be specially enabled; this means that <a class="reference external" href="../reference/compound_stmts.html#with"><tt class="xref docutils literal">with</tt></a> is now always a keyword. The rest of this section is a copy of the corresponding section from the “What’s New in Python 2.5” document; if you’re familiar with the ‘<a class="reference external" href="../reference/compound_stmts.html#with"><tt class="xref docutils literal">with</tt></a>‘ statement from Python 2.5, you can skip this section. The ‘<a class="reference external" href="../reference/compound_stmts.html#with"><tt class="xref docutils literal">with</tt></a>‘ statement clarifies code that previously would use <tt class="docutils literal">try...finally</tt> blocks to ensure that clean-up code is executed. In this section, I’ll discuss the statement as it will commonly be used. In the next section, I’ll examine the implementation details and show how to write objects for use with this statement. The ‘<a class="reference external" href="../reference/compound_stmts.html#with"><tt class="xref docutils literal">with</tt></a>‘ statement is a control-flow structure whose basic structure is: <div class="highlight-python3"><div class="highlight"><pre>with expression [as variable]: with-block </pre></div> </div> The expression is evaluated, and it should result in an object that supports the context management protocol (that is, has <a title="object.__enter__" class="reference external" href="../reference/datamodel.html#object.__enter__"><tt class="xref docutils literal">__enter__()</tt></a> and <a title="object.__exit__" class="reference external" href="../reference/datamodel.html#object.__exit__"><tt class="xref docutils literal">__exit__()</tt></a> methods). The object’s <a title="object.__enter__" class="reference external" href="../reference/datamodel.html#object.__enter__"><tt class="xref docutils literal">__enter__()</tt></a> is called before with-block is executed and therefore can run set-up code. It also may return a value that is bound to the name variable, if given. (Note carefully that variable is not assigned the result of expression.) After execution of the with-block is finished, the object’s <a title="object.__exit__" class="reference external" href="../reference/datamodel.html#object.__exit__"><tt class="xref docutils literal">__exit__()</tt></a> method is called, even if the block raised an exception, and can therefore run clean-up code. Some standard Python objects now support the context management protocol and can be used with the ‘<a class="reference external" href="../reference/compound_stmts.html#with"><tt class="xref docutils literal">with</tt></a>‘ statement. File objects are one example: <div class="highlight-python3"><div class="highlight"><pre>with open('/etc/passwd', 'r') as f: for line in f: print line ... more processing code ... </pre></div> </div> After this statement has executed, the file object in f will have been automatically closed, even if the <a class="reference external" href="../reference/compound_stmts.html#for"><tt class="xref docutils literal">for</tt></a> loop raised an exception part- way through the block. <div class="admonition note"> Note In this case, f is the same object created by <a title="open" class="reference external" href="../library/functions.html#open"><tt class="xref docutils literal">open()</tt></a>, because <tt class="xref docutils literal">file.__enter__()</tt> returns self. </div> The <a title="Thread-based parallelism." class="reference external" href="../library/threading.html#module-threading"><tt class="xref docutils literal">threading</tt></a> module’s locks and condition variables also support the ‘<a class="reference external" href="../reference/compound_stmts.html#with"><tt class="xref docutils literal">with</tt></a>‘ statement: <div class="highlight-python3"><div class="highlight"><pre>lock = threading.Lock() with lock: # Critical section of code ... </pre></div> </div> The lock is acquired before the block is executed and always released once the block is complete. The <tt class="xref docutils literal">localcontext()</tt> function in the <a title="Implementation of the General Decimal Arithmetic Specification." class="reference external" href="../library/decimal.html#module-decimal"><tt class="xref docutils literal">decimal</tt></a> module makes it easy to save and restore the current decimal context, which encapsulates the desired precision and rounding characteristics for computations: <div class="highlight-python3"><div class="highlight"><pre>from decimal import Decimal, Context, localcontext # Displays with default precision of 28 digits v = Decimal('578') print v.sqrt() with localcontext(Context(prec=16)): # All code in this block uses a precision of 16 digits. # The original context is restored on exiting the block. print v.sqrt() </pre></div> </div> <div class="section" id="writing-context-managers"> <h3>Writing Context Managers<a class="headerlink" href="#writing-context-managers" title="Permalink to this headline">露</a></h3> Under the hood, the ‘<a class="reference external" href="../reference/compound_stmts.html#with"><tt class="xref docutils literal">with</tt></a>‘ statement is fairly complicated. Most people will only use ‘<a class="reference external" href="../reference/compound_stmts.html#with"><tt class="xref docutils literal">with</tt></a>‘ in company with existing objects and don’t need to know these details, so you can skip the rest of this section if you like. Authors of new objects will need to understand the details of the underlying implementation and should keep reading. A high-level explanation of the context management protocol is: <ul class="simple"> <li>The expression is evaluated and should result in an object called a “context manager”. The context manager must have <a title="object.__enter__" class="reference external" href="../reference/datamodel.html#object.__enter__"><tt class="xref docutils literal">__enter__()</tt></a> and <a title="object.__exit__" class="reference external" href="../reference/datamodel.html#object.__exit__"><tt class="xref docutils literal">__exit__()</tt></a> methods.</li> <li>The context manager’s <a title="object.__enter__" class="reference external" href="../reference/datamodel.html#object.__enter__"><tt class="xref docutils literal">__enter__()</tt></a> method is called. The value returned is assigned to VAR. If no <tt class="docutils literal">as VAR</tt> clause is present, the value is simply discarded.</li> <li>The code in BLOCK is executed.</li> <li>If BLOCK raises an exception, the context manager’s <a title="object.__exit__" class="reference external" href="../reference/datamodel.html#object.__exit__"><tt class="xref docutils literal">__exit__()</tt></a> method is called with three arguments, the exception details (<tt class="docutils literal">type, value, traceback</tt>, the same values returned by <a title="sys.exc_info" class="reference external" href="../library/sys.html#sys.exc_info"><tt class="xref docutils literal">sys.exc_info()</tt></a>, which can also be <tt class="xref docutils literal">None</tt> if no exception occurred). The method’s return value controls whether an exception is re-raised: any false value re-raises the exception, and <tt class="xref docutils literal">True</tt> will result in suppressing it. You’ll only rarely want to suppress the exception, because if you do the author of the code containing the ‘<a class="reference external" href="../reference/compound_stmts.html#with"><tt class="xref docutils literal">with</tt></a>‘ statement will never realize anything went wrong.</li> <li>If BLOCK didn’t raise an exception, the <a title="object.__exit__" class="reference external" href="../reference/datamodel.html#object.__exit__"><tt class="xref docutils literal">__exit__()</tt></a> method is still called, but type, value, and traceback are all <tt class="xref docutils literal">None</tt>.</li> </ul> Let’s think through an example. I won’t present detailed code but will only sketch the methods necessary for a database that supports transactions. (For people unfamiliar with database terminology: a set of changes to the database are grouped into a transaction. Transactions can be either committed, meaning that all the changes are written into the database, or rolled back, meaning that the changes are all discarded and the database is unchanged. See any database textbook for more information.) Let’s assume there’s an object representing a database connection. Our goal will be to let the user write code like this: <div class="highlight-python3"><div class="highlight"><pre>db_connection = DatabaseConnection() with db_connection as cursor: cursor.execute('insert into ...') cursor.execute('delete from ...') # ... more operations ... </pre></div> </div> The transaction should be committed if the code in the block runs flawlessly or rolled back if there’s an exception. Here’s the basic interface for <tt class="xref docutils literal">DatabaseConnection</tt> that I’ll assume: <div class="highlight-python3"><div class="highlight"><pre>class DatabaseConnection: # Database interface def cursor(self): "Returns a cursor object and starts a new transaction" def commit(self): "Commits current transaction" def rollback(self): "Rolls back current transaction" </pre></div> </div> The <a title="object.__enter__" class="reference external" href="../reference/datamodel.html#object.__enter__"><tt class="xref docutils literal">__enter__()</tt></a> method is pretty easy, having only to start a new transaction. For this application the resulting cursor object would be a useful result, so the method will return it. The user can then add <tt class="docutils literal">as cursor</tt> to their ‘<a class="reference external" href="../reference/compound_stmts.html#with"><tt class="xref docutils literal">with</tt></a>‘ statement to bind the cursor to a variable name. <div class="highlight-python3"><div class="highlight"><pre>class DatabaseConnection: ... def __enter__(self): # Code to start a new transaction cursor = self.cursor() return cursor </pre></div> </div> The <a title="object.__exit__" class="reference external" href="../reference/datamodel.html#object.__exit__"><tt class="xref docutils literal">__exit__()</tt></a> method is the most complicated because it’s where most of the work has to be done. The method has to check if an exception occurred. If there was no exception, the transaction is committed. The transaction is rolled back if there was an exception. In the code below, execution will just fall off the end of the function, returning the default value of <tt class="xref docutils literal">None</tt>. <tt class="xref docutils literal">None</tt> is false, so the exception will be re-raised automatically. If you wished, you could be more explicit and add a <a class="reference external" href="../reference/simple_stmts.html#return"><tt class="xref docutils literal">return</tt></a> statement at the marked location. <div class="highlight-python3"><div class="highlight"><pre>class DatabaseConnection: ... def __exit__(self, type, value, tb): if tb is None: # No exception, so commit self.commit() else: # Exception occurred, so rollback. self.rollback() # return False </pre></div> </div> </div> <div class="section" id="the-contextlib-module"> <h3>The contextlib module<a class="headerlink" href="#the-contextlib-module" title="Permalink to this headline">露</a></h3> The <a title="Utilities for with-statement contexts." class="reference external" href="../library/contextlib.html#module-contextlib"><tt class="xref docutils literal">contextlib</tt></a> module provides some functions and a decorator that are useful when writing objects for use with the ‘<a class="reference external" href="../reference/compound_stmts.html#with"><tt class="xref docutils literal">with</tt></a>‘ statement. The decorator is called <tt class="xref docutils literal">contextmanager()</tt>, and lets you write a single generator function instead of defining a new class. The generator should yield exactly one value. The code up to the <a class="reference external" href="../reference/simple_stmts.html#yield"><tt class="xref docutils literal">yield</tt></a> will be executed as the <a title="object.__enter__" class="reference external" href="../reference/datamodel.html#object.__enter__"><tt class="xref docutils literal">__enter__()</tt></a> method, and the value yielded will be the method’s return value that will get bound to the variable in the ‘<a class="reference external" href="../reference/compound_stmts.html#with"><tt class="xref docutils literal">with</tt></a>‘ statement’s <a class="reference external" href="../reference/compound_stmts.html#as"><tt class="xref docutils literal">as</tt></a> clause, if any. The code after the <a class="reference external" href="../reference/simple_stmts.html#yield"><tt class="xref docutils literal">yield</tt></a> will be executed in the <a title="object.__exit__" class="reference external" href="../reference/datamodel.html#object.__exit__"><tt class="xref docutils literal">__exit__()</tt></a> method. Any exception raised in the block will be raised by the <a class="reference external" href="../reference/simple_stmts.html#yield"><tt class="xref docutils literal">yield</tt></a> statement. Using this decorator, our database example from the previous section could be written as: <div class="highlight-python3"><div class="highlight"><pre>from contextlib import contextmanager @contextmanager def db_transaction(connection): cursor = connection.cursor() try: yield cursor except: connection.rollback() raise else: connection.commit() db = DatabaseConnection() with db_transaction(db) as cursor: ... </pre></div> </div> The <a title="Utilities for with-statement contexts." class="reference external" href="../library/contextlib.html#module-contextlib"><tt class="xref docutils literal">contextlib</tt></a> module also has a <tt class="docutils literal">nested(mgr1, mgr2, ...)</tt> function that combines a number of context managers so you don’t need to write nested ‘<a class="reference external" href="../reference/compound_stmts.html#with"><tt class="xref docutils literal">with</tt></a>‘ statements. In this example, the single ‘<a class="reference external" href="../reference/compound_stmts.html#with"><tt class="xref docutils literal">with</tt></a>‘ statement both starts a database transaction and acquires a thread lock: <div class="highlight-python3"><div class="highlight"><pre>lock = threading.Lock() with nested (db_transaction(db), lock) as (cursor, locked): ... </pre></div> </div> Finally, the <tt class="xref docutils literal">closing()</tt> function returns its argument so that it can be bound to a variable, and calls the argument’s <tt class="docutils literal">.close()</tt> method at the end of the block. <div class="highlight-python3"><div class="highlight"><pre>import urllib, sys from contextlib import closing with closing(urllib.urlopen('http://www.yahoo.com')) as f: for line in f: sys.stdout.write(line) </pre></div> </div> <div class="admonition-see-also admonition seealso"> See also <dl class="docutils"> <dt><a class="reference external" href="http://www.python.org/dev/peps/pep-0343">PEP 343</a> - The “with” statement</dt> <dd>PEP written by Guido van Rossum and Nick Coghlan; implemented by Mike Bland, Guido van Rossum, and Neal Norwitz. The PEP shows the code generated for a ‘<a class="reference external" href="../reference/compound_stmts.html#with"><tt class="xref docutils literal">with</tt></a>‘ statement, which can be helpful in learning how the statement works.</dd> </dl> The documentation for the <a title="Utilities for with-statement contexts." class="reference external" href="../library/contextlib.html#module-contextlib"><tt class="xref docutils literal">contextlib</tt></a> module. </div> </div> </div> <div class="section" id="pep-366-explicit-relative-imports-from-a-main-module"> <h2>PEP 366: Explicit Relative Imports From a Main Module<a class="headerlink" href="#pep-366-explicit-relative-imports-from-a-main-module" title="Permalink to this headline">露</a></h2> Python’s <a class="reference external" href="../using/cmdline.html#cmdoption-trace-m">-m</a> switch allows running a module as a script. When you ran a module that was located inside a package, relative imports didn’t work correctly. The fix for Python 2.6 adds a <tt class="xref docutils literal">__package__</tt> attribute to modules. When this attribute is present, relative imports will be relative to the value of this attribute instead of the <tt class="xref docutils literal">__name__</tt> attribute. PEP 302-style importers can then set <tt class="xref docutils literal">__package__</tt> as necessary. The <a title="Locate and run Python modules without importing them first." class="reference external" href="../library/runpy.html#module-runpy"><tt class="xref docutils literal">runpy</tt></a> module that implements the <a class="reference external" href="../using/cmdline.html#cmdoption-trace-m">-m</a> switch now does this, so relative imports will now work correctly in scripts running from inside a package. </div> <div class="section" id="pep-370-per-user-site-packages-directory"> <h2>PEP 370: Per-user <tt class="docutils literal">site-packages</tt> Directory<a class="headerlink" href="#pep-370-per-user-site-packages-directory" title="Permalink to this headline">露</a></h2> When you run Python, the module search path <tt class="docutils literal">sys.path</tt> usually includes a directory whose path ends in <tt class="docutils literal">"site-packages"</tt>. This directory is intended to hold locally-installed packages available to all users using a machine or a particular site installation. Python 2.6 introduces a convention for user-specific site directories. The directory varies depending on the platform: <ul class="simple"> <li>Unix and Mac OS X: <tt class="docutils literal">~/.local/</tt></li> <li>Windows: <tt class="docutils literal">%APPDATA%/Python</tt></li> </ul> Within this directory, there will be version-specific subdirectories, such as <tt class="docutils literal">lib/python2.6/site-packages</tt> on Unix/Mac OS and <tt class="docutils literal">Python26/site-packages</tt> on Windows. If you don’t like the default directory, it can be overridden by an environment variable. <a class="reference external" href="../using/cmdline.html#envvar-PYTHONUSERBASE">PYTHONUSERBASE</a> sets the root directory used for all Python versions supporting this feature. On Windows, the directory for application-specific data can be changed by setting the APPDATA environment variable. You can also modify the <tt class="docutils literal">site.py</tt> file for your Python installation. The feature can be disabled entirely by running Python with the <a class="reference external" href="../using/cmdline.html#cmdoption-trace-s">-s</a> option or setting the <a class="reference external" href="../using/cmdline.html#envvar-PYTHONNOUSERSITE">PYTHONNOUSERSITE</a> environment variable. <div class="admonition-see-also admonition seealso"> See also <dl class="last docutils"> <dt><a class="reference external" href="http://www.python.org/dev/peps/pep-0370">PEP 370</a> - Per-user <tt class="docutils literal">site-packages</tt> Directory</dt> <dd>PEP written and implemented by Christian Heimes.</dd> </dl> </div> </div> <div class="section" id="pep-371-the-multiprocessing-package"> <h2>PEP 371: The <tt class="docutils literal">multiprocessing</tt> Package<a class="headerlink" href="#pep-371-the-multiprocessing-package" title="Permalink to this headline">露</a></h2> The new <a title="Process-based parallelism." class="reference external" href="../library/multiprocessing.html#module-multiprocessing"><tt class="xref docutils literal">multiprocessing</tt></a> package lets Python programs create new processes that will perform a computation and return a result to the parent. The parent and child processes can communicate using queues and pipes, synchronize their operations using locks and semaphores, and can share simple arrays of data. The <a title="Process-based parallelism." class="reference external" href="../library/multiprocessing.html#module-multiprocessing"><tt class="xref docutils literal">multiprocessing</tt></a> module started out as an exact emulation of the <a title="Thread-based parallelism." class="reference external" href="../library/threading.html#module-threading"><tt class="xref docutils literal">threading</tt></a> module using processes instead of threads. That goal was discarded along the path to Python 2.6, but the general approach of the module is still similar. The fundamental class is the <tt class="xref docutils literal">Process</tt>, which is passed a callable object and a collection of arguments. The <tt class="xref docutils literal">start()</tt> method sets the callable running in a subprocess, after which you can call the <tt class="xref docutils literal">is_alive()</tt> method to check whether the subprocess is still running and the <tt class="xref docutils literal">join()</tt> method to wait for the process to exit. Here’s a simple example where the subprocess will calculate a factorial. The function doing the calculation is written strangely so that it takes significantly longer when the input argument is a multiple of 4. <div class="highlight-python3"><div class="highlight"><pre>import time from multiprocessing import Process, Queue def factorial(queue, N): "Compute a factorial." # If N is a multiple of 4, this function will take much longer. if (N % 4) == 0: time.sleep(.05 * N/4) # Calculate the result fact = 1L for i in range(1, N+1): fact = fact * i # Put the result on the queue queue.put(fact) if __name__ == '__main__': queue = Queue() N = 5 p = Process(target=factorial, args=(queue, N)) p.start() p.join() result = queue.get() print 'Factorial', N, '=', result </pre></div> </div> A <tt class="xref docutils literal">Queue</tt> is used to communicate the input parameter N and the result. The <tt class="xref docutils literal">Queue</tt> object is stored in a global variable. The child process will use the value of the variable when the child was created; because it’s a <tt class="xref docutils literal">Queue</tt>, parent and child can use the object to communicate. (If the parent were to change the value of the global variable, the child’s value would be unaffected, and vice versa.) Two other classes, <tt class="xref docutils literal">Pool</tt> and <tt class="xref docutils literal">Manager</tt>, provide higher-level interfaces. <tt class="xref docutils literal">Pool</tt> will create a fixed number of worker processes, and requests can then be distributed to the workers by calling <tt class="xref docutils literal">apply()</tt> or <tt class="xref docutils literal">apply_async()</tt> to add a single request, and <a title="map" class="reference external" href="../library/functions.html#map"><tt class="xref docutils literal">map()</tt></a> or <tt class="xref docutils literal">map_async()</tt> to add a number of requests. The following code uses a <tt class="xref docutils literal">Pool</tt> to spread requests across 5 worker processes and retrieve a list of results: <div class="highlight-python3"><div class="highlight"><pre>from multiprocessing import Pool def factorial(N, dictionary): "Compute a factorial." ... p = Pool(5) result = p.map(factorial, range(1, 1000, 10)) for v in result: print v </pre></div> </div> This produces the following output: <div class="highlight-python3"><div class="highlight"><pre>1 39916800 51090942171709440000 8222838654177922817725562880000000 33452526613163807108170062053440751665152000000000 ... </pre></div> </div> The other high-level interface, the <tt class="xref docutils literal">Manager</tt> class, creates a separate server process that can hold master copies of Python data structures. Other processes can then access and modify these data structures using proxy objects. The following example creates a shared dictionary by calling the <a title="dict" class="reference external" href="../library/stdtypes.html#dict"><tt class="xref docutils literal">dict()</tt></a> method; the worker processes then insert values into the dictionary. (Locking is not done for you automatically, which doesn’t matter in this example. <tt class="xref docutils literal">Manager</tt>‘s methods also include <tt class="xref docutils literal">Lock()</tt>, <tt class="xref docutils literal">RLock()</tt>, and <tt class="xref docutils literal">Semaphore()</tt> to create shared locks.) <div class="highlight-python3"><div class="highlight"><pre>import time from multiprocessing import Pool, Manager def factorial(N, dictionary): "Compute a factorial." # Calculate the result fact = 1L for i in range(1, N+1): fact = fact * i # Store result in dictionary dictionary[N] = fact if __name__ == '__main__': p = Pool(5) mgr = Manager() d = mgr.dict() # Create shared dictionary # Run tasks using the pool for N in range(1, 1000, 10): p.apply_async(factorial, (N, d)) # Mark pool as closed -- no more tasks can be added. p.close() # Wait for tasks to exit p.join() # Output results for k, v in sorted(d.items()): print k, v </pre></div> </div> This will produce the output: <div class="highlight-python3"><div class="highlight"><pre>1 1 11 39916800 21 51090942171709440000 31 8222838654177922817725562880000000 41 33452526613163807108170062053440751665152000000000 51 15511187532873822802242430164693032110632597200169861120000... </pre></div> </div> <div class="admonition-see-also admonition seealso"> See also The documentation for the <a title="Process-based parallelism." class="reference external" href="../library/multiprocessing.html#module-multiprocessing"><tt class="xref docutils literal">multiprocessing</tt></a> module. <dl class="last docutils"> <dt><a class="reference external" href="http://www.python.org/dev/peps/pep-0371">PEP 371</a> - Addition of the multiprocessing package</dt> <dd>PEP written by Jesse Noller and Richard Oudkerk; implemented by Richard Oudkerk and Jesse Noller.</dd> </dl> </div> </div> <div class="section" id="pep-3101-advanced-string-formatting"> <h2>PEP 3101: Advanced String Formatting<a class="headerlink" href="#pep-3101-advanced-string-formatting" title="Permalink to this headline">露</a></h2> In Python 3.0, the <cite>%</cite> operator is supplemented by a more powerful string formatting method, <a title="format" class="reference external" href="../library/functions.html#format"><tt class="xref docutils literal">format()</tt></a>. Support for the <a title="str.format" class="reference external" href="../library/stdtypes.html#str.format"><tt class="xref docutils literal">str.format()</tt></a> method has been backported to Python 2.6. In 2.6, both 8-bit and Unicode strings have a <cite>.format()</cite> method that treats the string as a template and takes the arguments to be formatted. The formatting template uses curly brackets (<cite>{</cite>, <cite>}</cite>) as special characters: <div class="highlight-python3"><div class="highlight"><pre>>>> # Substitute positional argument 0 into the string. >>> "User ID: {0}".format("root") 'User ID: root' >>> # Use the named keyword arguments >>> "User ID: {uid} Last seen: {last_login}".format( ... uid="root", ... last_login = "5 Mar 2008 07:20") 'User ID: root Last seen: 5 Mar 2008 07:20' </pre></div> </div> Curly brackets can be escaped by doubling them: <div class="highlight-python3"><div class="highlight"><pre>>>> "Empty dict: {{}}".format() "Empty dict: {}" </pre></div> </div> Field names can be integers indicating positional arguments, such as <tt class="docutils literal">{0}</tt>, <tt class="docutils literal">{1}</tt>, etc. or names of keyword arguments. You can also supply compound field names that read attributes or access dictionary keys: <div class="highlight-python3"><div class="highlight"><pre>>>> import sys >>> print 'Platform: {0.platform}\nPython version: {0.version}'.format(sys) Platform: darwin Python version: 2.6a1+ (trunk:61261M, Mar 5 2008, 20:29:41) [GCC 4.0.1 (Apple Computer, Inc. build 5367)]' >>> import mimetypes >>> 'Content-type: {0[.mp4]}'.format(mimetypes.types_map) 'Content-type: video/mp4' </pre></div> </div> Note that when using dictionary-style notation such as <tt class="docutils literal">[.mp4]</tt>, you don’t need to put any quotation marks around the string; it will look up the value using <tt class="docutils literal">.mp4</tt> as the key. Strings beginning with a number will be converted to an integer. You can’t write more complicated expressions inside a format string. So far we’ve shown how to specify which field to substitute into the resulting string. The precise formatting used is also controllable by adding a colon followed by a format specifier. For example: <div class="highlight-python3"><div class="highlight"><pre>>>> # Field 0: left justify, pad to 15 characters >>> # Field 1: right justify, pad to 6 characters >>> fmt = '{0:15} ${1:>6}' >>> fmt.format('Registration', 35) 'Registration $ 35' >>> fmt.format('Tutorial', 50) 'Tutorial $ 50' >>> fmt.format('Banquet', 125) 'Banquet $ 125' </pre></div> </div> Format specifiers can reference other fields through nesting: <div class="highlight-python3"><div class="highlight"><pre>>>> fmt = '{0:{1}}' >>> width = 15 >>> fmt.format('Invoice #1234', width) 'Invoice #1234 ' >>> width = 35 >>> fmt.format('Invoice #1234', width) 'Invoice #1234 ' </pre></div> </div> The alignment of a field within the desired width can be specified: <table border="1" class="docutils"> <colgroup> <col width="27%" /> <col width="73%" /> </colgroup> <thead valign="bottom"> <tr><th class="head">Character</th> <th class="head">Effect</th> </tr> </thead> <tbody valign="top"> <tr><td>< (default)</td> <td>Left-align</td> </tr> <tr><td>></td> <td>Right-align</td> </tr> <tr><td>^</td> <td>Center</td> </tr> <tr><td>=</td> <td>(For numeric types only) Pad after the sign.</td> </tr> </tbody> </table> Format specifiers can also include a presentation type, which controls how the value is formatted. For example, floating-point numbers can be formatted as a general number or in exponential notation: <div class="highlight-python3"><div class="highlight"><pre>>>> '{0:g}'.format(3.75) '3.75' >>> '{0:e}'.format(3.75) '3.750000e+00' </pre></div> </div> A variety of presentation types are available. Consult the 2.6 documentation for a <a class="reference external" href="../library/string.html#formatstrings">complete list</a>; here’s a sample: <table border="1" class="docutils"> <colgroup> <col width="6%" /> <col width="94%" /> </colgroup> <tbody valign="top"> <tr><td><tt class="docutils literal">b</tt></td> <td>Binary. Outputs the number in base 2.</td> </tr> <tr><td><tt class="docutils literal">c</tt></td> <td>Character. Converts the integer to the corresponding Unicode character before printing.</td> </tr> <tr><td><tt class="docutils literal">d</tt></td> <td>Decimal Integer. Outputs the number in base 10.</td> </tr> <tr><td><tt class="docutils literal">o</tt></td> <td>Octal format. Outputs the number in base 8.</td> </tr> <tr><td><tt class="docutils literal">x</tt></td> <td>Hex format. Outputs the number in base 16, using lower-case letters for the digits above 9.</td> </tr> <tr><td><tt class="docutils literal">e</tt></td> <td>Exponent notation. Prints the number in scientific notation using the letter ‘e’ to indicate the exponent.</td> </tr> <tr><td><tt class="docutils literal">g</tt></td> <td>General format. This prints the number as a fixed-point number, unless the number is too large, in which case it switches to ‘e’ exponent notation.</td> </tr> <tr><td><tt class="docutils literal">n</tt></td> <td>Number. This is the same as ‘g’ (for floats) or ‘d’ (for integers), except that it uses the current locale setting to insert the appropriate number separator characters.</td> </tr> <tr><td><tt class="docutils literal">%</tt></td> <td>Percentage. Multiplies the number by 100 and displays in fixed (‘f’) format, followed by a percent sign.</td> </tr> </tbody> </table> Classes and types can define a <a title="object.__format__" class="reference external" href="../reference/datamodel.html#object.__format__"><tt class="xref docutils literal">__format__()</tt></a> method to control how they’re formatted. It receives a single argument, the format specifier: <div class="highlight-python3"><div class="highlight"><pre>def __format__(self, format_spec): if isinstance(format_spec, unicode): return unicode(str(self)) else: return str(self) </pre></div> </div> There’s also a <a title="format" class="reference external" href="../library/functions.html#format"><tt class="xref docutils literal">format()</tt></a> builtin that will format a single value. It calls the type’s <a title="object.__format__" class="reference external" href="../reference/datamodel.html#object.__format__"><tt class="xref docutils literal">__format__()</tt></a> method with the provided specifier: <div class="highlight-python3"><div class="highlight"><pre>>>> format(75.6564, '.2f') '75.66' </pre></div> </div> <div class="admonition-see-also admonition seealso"> See also <dl class="last docutils"> <dt><a class="reference external" href="../library/string.html#formatstrings">Format String Syntax</a></dt> <dd>The reference documentation for format fields.</dd> <dt><a class="reference external" href="http://www.python.org/dev/peps/pep-3101">PEP 3101</a> - Advanced String Formatting</dt> <dd>PEP written by Talin. Implemented by Eric Smith.</dd> </dl> </div> </div> <div class="section" id="pep-3105-print-as-a-function"> <h2>PEP 3105: <tt class="docutils literal">print</tt> As a Function<a class="headerlink" href="#pep-3105-print-as-a-function" title="Permalink to this headline">露</a></h2> The <tt class="docutils literal">print</tt> statement becomes the <a title="print" class="reference external" href="../library/functions.html#print"><tt class="xref docutils literal">print()</tt></a> function in Python 3.0. Making <a title="print" class="reference external" href="../library/functions.html#print"><tt class="xref docutils literal">print()</tt></a> a function makes it possible to replace the function by doing <tt class="docutils literal">def print(...)</tt> or importing a new function from somewhere else. Python 2.6 has a <tt class="docutils literal">__future__</tt> import that removes <tt class="docutils literal">print</tt> as language syntax, letting you use the functional form instead. For example: <div class="highlight-python3"><div class="highlight"><pre>>>> from __future__ import print_function >>> print('# of entries', len(dictionary), file=sys.stderr) </pre></div> </div> The signature of the new function is: <div class="highlight-python3"><div class="highlight"><pre>def print(*args, sep=' ', end='\n', file=None) </pre></div> </div> The parameters are: <blockquote> <ul class="simple"> <li>args: positional arguments whose values will be printed out.</li> <li>sep: the separator, which will be printed between arguments.</li> <li>end: the ending text, which will be printed after all of the arguments have been output.</li> <li>file: the file object to which the output will be sent.</li> </ul> </blockquote> <div class="admonition-see-also admonition seealso"> See also <dl class="last docutils"> <dt><a class="reference external" href="http://www.python.org/dev/peps/pep-3105">PEP 3105</a> - Make print a function</dt> <dd>PEP written by Georg Brandl.</dd> </dl> </div> </div> <div class="section" id="pep-3110-exception-handling-changes"> <h2>PEP 3110: Exception-Handling Changes<a class="headerlink" href="#pep-3110-exception-handling-changes" title="Permalink to this headline">露</a></h2> One error that Python programmers occasionally make is writing the following code: <div class="highlight-python3"><div class="highlight"><pre>try: ... except TypeError, ValueError: # Wrong! ... </pre></div> </div> The author is probably trying to catch both <a title="TypeError" class="reference external" href="../library/exceptions.html#TypeError"><tt class="xref docutils literal">TypeError</tt></a> and <a title="ValueError" class="reference external" href="../library/exceptions.html#ValueError"><tt class="xref docutils literal">ValueError</tt></a> exceptions, but this code actually does something different: it will catch <a title="TypeError" class="reference external" href="../library/exceptions.html#TypeError"><tt class="xref docutils literal">TypeError</tt></a> and bind the resulting exception object to the local name <tt class="docutils literal">"ValueError"</tt>. The <a title="ValueError" class="reference external" href="../library/exceptions.html#ValueError"><tt class="xref docutils literal">ValueError</tt></a> exception will not be caught at all. The correct code specifies a tuple of exceptions: <div class="highlight-python3"><div class="highlight"><pre>try: ... except (TypeError, ValueError): ... </pre></div> </div> This error happens because the use of the comma here is ambiguous: does it indicate two different nodes in the parse tree, or a single node that’s a tuple? Python 3.0 makes this unambiguous by replacing the comma with the word “as”. To catch an exception and store the exception object in the variable <tt class="docutils literal">exc</tt>, you must write: <div class="highlight-python3"><div class="highlight"><pre>try: ... except TypeError as exc: ... </pre></div> </div> Python 3.0 will only support the use of “as”, and therefore interprets the first example as catching two different exceptions. Python 2.6 supports both the comma and “as”, so existing code will continue to work. We therefore suggest using “as” when writing new Python code that will only be executed with 2.6. <div class="admonition-see-also admonition seealso"> See also <dl class="last docutils"> <dt><a class="reference external" href="http://www.python.org/dev/peps/pep-3110">PEP 3110</a> - Catching Exceptions in Python 3000</dt> <dd>PEP written and implemented by Collin Winter.</dd> </dl> </div> </div> <div class="section" id="pep-3112-byte-literals"> <h2>PEP 3112: Byte Literals<a class="headerlink" href="#pep-3112-byte-literals" title="Permalink to this headline">露</a></h2> Python 3.0 adopts Unicode as the language’s fundamental string type and denotes 8-bit literals differently, either as <tt class="docutils literal">b'string'</tt> or using a <a title="bytes" class="reference external" href="../library/functions.html#bytes"><tt class="xref docutils literal">bytes</tt></a> constructor. For future compatibility, Python 2.6 adds <a title="bytes" class="reference external" href="../library/functions.html#bytes"><tt class="xref docutils literal">bytes</tt></a> as a synonym for the <a title="str" class="reference external" href="../library/functions.html#str"><tt class="xref docutils literal">str</tt></a> type, and it also supports the <tt class="docutils literal">b''</tt> notation. The 2.6 <a title="str" class="reference external" href="../library/functions.html#str"><tt class="xref docutils literal">str</tt></a> differs from 3.0’s <a title="bytes" class="reference external" href="../library/functions.html#bytes"><tt class="xref docutils literal">bytes</tt></a> type in various ways; most notably, the constructor is completely different. In 3.0, <tt class="docutils literal">bytes([65, 66, 67])</tt> is 3 elements long, containing the bytes representing <tt class="docutils literal">ABC</tt>; in 2.6, <tt class="docutils literal">bytes([65, 66, 67])</tt> returns the 12-byte string representing the <a title="str" class="reference external" href="../library/functions.html#str"><tt class="xref docutils literal">str()</tt></a> of the list. The primary use of <a title="bytes" class="reference external" href="../library/functions.html#bytes"><tt class="xref docutils literal">bytes</tt></a> in 2.6 will be to write tests of object type such as <tt class="docutils literal">isinstance(x, bytes)</tt>. This will help the 2to3 converter, which can’t tell whether 2.x code intends strings to contain either characters or 8-bit bytes; you can now use either <a title="bytes" class="reference external" href="../library/functions.html#bytes"><tt class="xref docutils literal">bytes</tt></a> or <a title="str" class="reference external" href="../library/functions.html#str"><tt class="xref docutils literal">str</tt></a> to represent your intention exactly, and the resulting code will also be correct in Python 3.0. There’s also a <tt class="docutils literal">__future__</tt> import that causes all string literals to become Unicode strings. This means that <tt class="docutils literal">\u</tt> escape sequences can be used to include Unicode characters: <div class="highlight-python3"><div class="highlight"><pre>from __future__ import unicode_literals s = ('\u751f\u3080\u304e\u3000\u751f\u3054' '\u3081\u3000\u751f\u305f\u307e\u3054') print len(s) # 12 Unicode characters </pre></div> </div> At the C level, Python 3.0 will rename the existing 8-bit string type, called <tt class="xref docutils literal">PyStringObject</tt> in Python 2.x, to <a title="PyBytesObject" class="reference external" href="../c-api/bytes.html#PyBytesObject"><tt class="xref docutils literal">PyBytesObject</tt></a>. Python 2.6 uses <tt class="docutils literal">#define</tt> to support using the names <a title="PyBytesObject" class="reference external" href="../c-api/bytes.html#PyBytesObject"><tt class="xref docutils literal">PyBytesObject()</tt></a>, <a title="PyBytes_Check" class="reference external" href="../c-api/bytes.html#PyBytes_Check"><tt class="xref docutils literal">PyBytes_Check()</tt></a>, <a title="PyBytes_FromStringAndSize" class="reference external" href="../c-api/bytes.html#PyBytes_FromStringAndSize"><tt class="xref docutils literal">PyBytes_FromStringAndSize()</tt></a>, and all the other functions and macros used with strings. Instances of the <a title="bytes" class="reference external" href="../library/functions.html#bytes"><tt class="xref docutils literal">bytes</tt></a> type are immutable just as strings are. A new <a title="bytearray" class="reference external" href="../library/functions.html#bytearray"><tt class="xref docutils literal">bytearray</tt></a> type stores a mutable sequence of bytes: <div class="highlight-python3"><div class="highlight"><pre>>>> bytearray([65, 66, 67]) bytearray(b'ABC') >>> b = bytearray(u'\u21ef\u3244', 'utf-8') >>> b bytearray(b'\xe2\x87\xaf\xe3\x89\x84') >>> b[0] = '\xe3' >>> b bytearray(b'\xe3\x87\xaf\xe3\x89\x84') >>> unicode(str(b), 'utf-8') u'\u31ef \u3244' </pre></div> </div> Byte arrays support most of the methods of string types, such as <tt class="xref docutils literal">startswith()</tt>/<tt class="xref docutils literal">endswith()</tt>, <tt class="xref docutils literal">find()</tt>/<tt class="xref docutils literal">rfind()</tt>, and some of the methods of lists, such as <tt class="xref docutils literal">append()</tt>, <tt class="xref docutils literal">pop()</tt>, and <tt class="xref docutils literal">reverse()</tt>. <div class="highlight-python3"><div class="highlight"><pre>>>> b = bytearray('ABC') >>> b.append('d') >>> b.append(ord('e')) >>> b bytearray(b'ABCde') </pre></div> </div> There’s also a corresponding C API, with <a title="PyByteArray_FromObject" class="reference external" href="../c-api/bytearray.html#PyByteArray_FromObject"><tt class="xref docutils literal">PyByteArray_FromObject()</tt></a>, <a title="PyByteArray_FromStringAndSize" class="reference external" href="../c-api/bytearray.html#PyByteArray_FromStringAndSize"><tt class="xref docutils literal">PyByteArray_FromStringAndSize()</tt></a>, and various other functions. <div class="admonition-see-also admonition seealso"> See also <dl class="last docutils"> <dt><a class="reference external" href="http://www.python.org/dev/peps/pep-3112">PEP 3112</a> - Bytes literals in Python 3000</dt> <dd>PEP written by Jason Orendorff; backported to 2.6 by Christian Heimes.</dd> </dl> </div> </div> <div class="section" id="pep-3116-new-i-o-library"> <h2>PEP 3116: New I/O Library<a class="headerlink" href="#pep-3116-new-i-o-library" title="Permalink to this headline">露</a></h2> Python’s built-in file objects support a number of methods, but file-like objects don’t necessarily support all of them. Objects that imitate files usually support <tt class="xref docutils literal">read()</tt> and <tt class="xref docutils literal">write()</tt>, but they may not support <tt class="xref docutils literal">readline()</tt>, for example. Python 3.0 introduces a layered I/O library in the <a title="Core tools for working with streams." class="reference external" href="../library/io.html#module-io"><tt class="xref docutils literal">io</tt></a> module that separates buffering and text-handling features from the fundamental read and write operations. There are three levels of abstract base classes provided by the <a title="Core tools for working with streams." class="reference external" href="../library/io.html#module-io"><tt class="xref docutils literal">io</tt></a> module: <ul> <li><tt class="xref docutils literal">RawIOBase</tt> defines raw I/O operations: <tt class="xref docutils literal">read()</tt>, <tt class="xref docutils literal">readinto()</tt>, <tt class="xref docutils literal">write()</tt>, <tt class="xref docutils literal">seek()</tt>, <tt class="xref docutils literal">tell()</tt>, <tt class="xref docutils literal">truncate()</tt>, and <tt class="xref docutils literal">close()</tt>. Most of the methods of this class will often map to a single system call. There are also <tt class="xref docutils literal">readable()</tt>, <tt class="xref docutils literal">writable()</tt>, and <tt class="xref docutils literal">seekable()</tt> methods for determining what operations a given object will allow. Python 3.0 has concrete implementations of this class for files and sockets, but Python 2.6 hasn’t restructured its file and socket objects in this way. </li> <li><tt class="xref docutils literal">BufferedIOBase</tt> is an abstract base class that buffers data in memory to reduce the number of system calls used, making I/O processing more efficient. It supports all of the methods of <tt class="xref docutils literal">RawIOBase</tt>, and adds a <tt class="xref docutils literal">raw</tt> attribute holding the underlying raw object. There are five concrete classes implementing this ABC. <tt class="xref docutils literal">BufferedWriter</tt> and <tt class="xref docutils literal">BufferedReader</tt> are for objects that support write-only or read-only usage that have a <tt class="xref docutils literal">seek()</tt> method for random access. <tt class="xref docutils literal">BufferedRandom</tt> objects support read and write access upon the same underlying stream, and <tt class="xref docutils literal">BufferedRWPair</tt> is for objects such as TTYs that have both read and write operations acting upon unconnected streams of data. The <tt class="xref docutils literal">BytesIO</tt> class supports reading, writing, and seeking over an in-memory buffer. </li> <li><tt class="xref docutils literal">TextIOBase</tt>: Provides functions for reading and writing strings (remember, strings will be Unicode in Python 3.0), and supporting universal newlines. <tt class="xref docutils literal">TextIOBase</tt> defines the <tt class="xref docutils literal">readline()</tt> method and supports iteration upon objects. There are two concrete implementations. <tt class="xref docutils literal">TextIOWrapper</tt> wraps a buffered I/O object, supporting all of the methods for text I/O and adding a <tt class="xref docutils literal">buffer</tt> attribute for access to the underlying object. <tt class="xref docutils literal">StringIO</tt> simply buffers everything in memory without ever writing anything to disk. (In Python 2.6, <a title="io.StringIO" class="reference external" href="../library/io.html#io.StringIO"><tt class="xref docutils literal">io.StringIO</tt></a> is implemented in pure Python, so it’s pretty slow. You should therefore stick with the existing <tt class="xref docutils literal">StringIO</tt> module or <tt class="xref docutils literal">cStringIO</tt> for now. At some point Python 3.0’s <a title="Core tools for working with streams." class="reference external" href="../library/io.html#module-io"><tt class="xref docutils literal">io</tt></a> module will be rewritten into C for speed, and perhaps the C implementation will be backported to the 2.x releases.) </li> </ul> In Python 2.6, the underlying implementations haven’t been restructured to build on top of the <a title="Core tools for working with streams." class="reference external" href="../library/io.html#module-io"><tt class="xref docutils literal">io</tt></a> module’s classes. The module is being provided to make it easier to write code that’s forward-compatible with 3.0, and to save developers the effort of writing their own implementations of buffering and text I/O. <div class="admonition-see-also admonition seealso"> See also <dl class="last docutils"> <dt><a class="reference external" href="http://www.python.org/dev/peps/pep-3116">PEP 3116</a> - New I/O</dt> <dd>PEP written by Daniel Stutzbach, Mike Verdone, and Guido van Rossum. Code by Guido van Rossum, Georg Brandl, Walter Doerwald, Jeremy Hylton, Martin von Loewis, Tony Lownds, and others.</dd> </dl> </div> </div> <div class="section" id="pep-3118-revised-buffer-protocol"> <h2>PEP 3118: Revised Buffer Protocol<a class="headerlink" href="#pep-3118-revised-buffer-protocol" title="Permalink to this headline">露</a></h2> The buffer protocol is a C-level API that lets Python types exchange pointers into their internal representations. A memory-mapped file can be viewed as a buffer of characters, for example, and this lets another module such as <a title="Regular expression operations." class="reference external" href="../library/re.html#module-re"><tt class="xref docutils literal">re</tt></a> treat memory-mapped files as a string of characters to be searched. The primary users of the buffer protocol are numeric-processing packages such as NumPy, which expose the internal representation of arrays so that callers can write data directly into an array instead of going through a slower API. This PEP updates the buffer protocol in light of experience from NumPy development, adding a number of new features such as indicating the shape of an array or locking a memory region. The most important new C API function is <tt class="docutils literal">PyObject_GetBuffer(PyObject *obj, Py_buffer *view, int flags)</tt>, which takes an object and a set of flags, and fills in the <tt class="docutils literal">Py_buffer</tt> structure with information about the object’s memory representation. Objects can use this operation to lock memory in place while an external caller could be modifying the contents, so there’s a corresponding <tt class="docutils literal">PyBuffer_Release(Py_buffer *view)</tt> to indicate that the external caller is done. The flags argument to <a title="PyObject_GetBuffer" class="reference external" href="../c-api/buffer.html#PyObject_GetBuffer"><tt class="xref docutils literal">PyObject_GetBuffer()</tt></a> specifies constraints upon the memory returned. Some examples are: <blockquote> <ul class="simple"> <li><a title="PyBUF_WRITABLE" class="reference external" href="../c-api/buffer.html#PyBUF_WRITABLE"><tt class="xref docutils literal">PyBUF_WRITABLE</tt></a> indicates that the memory must be writable.</li> <li><tt class="xref docutils literal">PyBUF_LOCK</tt> requests a read-only or exclusive lock on the memory.</li> <li><a title="PyBUF_C_CONTIGUOUS" class="reference external" href="../c-api/buffer.html#PyBUF_C_CONTIGUOUS"><tt class="xref docutils literal">PyBUF_C_CONTIGUOUS</tt></a> and <a title="PyBUF_F_CONTIGUOUS" class="reference external" href="../c-api/buffer.html#PyBUF_F_CONTIGUOUS"><tt class="xref docutils literal">PyBUF_F_CONTIGUOUS</tt></a> requests a C-contiguous (last dimension varies the fastest) or Fortran-contiguous (first dimension varies the fastest) array layout.</li> </ul> </blockquote> Two new argument codes for <a title="PyArg_ParseTuple" class="reference external" href="../c-api/arg.html#PyArg_ParseTuple"><tt class="xref docutils literal">PyArg_ParseTuple()</tt></a>, <tt class="docutils literal">s*</tt> and <tt class="docutils literal">z*</tt>, return locked buffer objects for a parameter. <div class="admonition-see-also admonition seealso"> See also <dl class="last docutils"> <dt><a class="reference external" href="http://www.python.org/dev/peps/pep-3118">PEP 3118</a> - Revising the buffer protocol</dt> <dd>PEP written by Travis Oliphant and Carl Banks; implemented by Travis Oliphant.</dd> </dl> </div> </div> <div class="section" id="pep-3119-abstract-base-classes"> <h2>PEP 3119: Abstract Base Classes<a class="headerlink" href="#pep-3119-abstract-base-classes" title="Permalink to this headline">露</a></h2> Some object-oriented languages such as Java support interfaces, declaring that a class has a given set of methods or supports a given access protocol. Abstract Base Classes (or ABCs) are an equivalent feature for Python. The ABC support consists of an <a title="Abstract base classes according to PEP 3119." class="reference external" href="../library/abc.html#module-abc"><tt class="xref docutils literal">abc</tt></a> module containing a metaclass called <tt class="xref docutils literal">ABCMeta</tt>, special handling of this metaclass by the <a title="isinstance" class="reference external" href="../library/functions.html#isinstance"><tt class="xref docutils literal">isinstance()</tt></a> and <a title="issubclass" class="reference external" href="../library/functions.html#issubclass"><tt class="xref docutils literal">issubclass()</tt></a> builtins, and a collection of basic ABCs that the Python developers think will be widely useful. Future versions of Python will probably add more ABCs. Let’s say you have a particular class and wish to know whether it supports dictionary-style access. The phrase “dictionary-style” is vague, however. It probably means that accessing items with <tt class="docutils literal">obj[1]</tt> works. Does it imply that setting items with <tt class="docutils literal">obj[2] = value</tt> works? Or that the object will have <tt class="xref docutils literal">keys()</tt>, <tt class="xref docutils literal">values()</tt>, and <tt class="xref docutils literal">items()</tt> methods? What about the iterative variants such as <tt class="xref docutils literal">iterkeys()</tt>? <tt class="xref docutils literal">copy()</tt> and <tt class="xref docutils literal">update()</tt>? Iterating over the object with <a title="iter" class="reference external" href="../library/functions.html#iter"><tt class="xref docutils literal">iter()</tt></a>? The Python 2.6 <a title="Container datatypes" class="reference external" href="../library/collections.html#module-collections"><tt class="xref docutils literal">collections</tt></a> module includes a number of different ABCs that represent these distinctions. <tt class="xref docutils literal">Iterable</tt> indicates that a class defines <a title="object.__iter__" class="reference external" href="../reference/datamodel.html#object.__iter__"><tt class="xref docutils literal">__iter__()</tt></a>, and <tt class="xref docutils literal">Container</tt> means the class defines a <a title="object.__contains__" class="reference external" href="../reference/datamodel.html#object.__contains__"><tt class="xref docutils literal">__contains__()</tt></a> method and therefore supports <tt class="docutils literal">x in y</tt> expressions. The basic dictionary interface of getting items, setting items, and <tt class="xref docutils literal">keys()</tt>, <tt class="xref docutils literal">values()</tt>, and <tt class="xref docutils literal">items()</tt>, is defined by the <tt class="xref docutils literal">MutableMapping</tt> ABC. You can derive your own classes from a particular ABC to indicate they support that ABC’s interface: <div class="highlight-python3"><div class="highlight"><pre>import collections class Storage(collections.MutableMapping): ... </pre></div> </div> Alternatively, you could write the class without deriving from the desired ABC and instead register the class by calling the ABC’s <tt class="xref docutils literal">register()</tt> method: <div class="highlight-python3"><div class="highlight"><pre>import collections class Storage: ... collections.MutableMapping.register(Storage) </pre></div> </div> For classes that you write, deriving from the ABC is probably clearer. The <tt class="xref docutils literal">register()</tt> method is useful when you’ve written a new ABC that can describe an existing type or class, or if you want to declare that some third-party class implements an ABC. For example, if you defined a <tt class="xref docutils literal">PrintableType</tt> ABC, it’s legal to do: <div class="highlight-python3"><div class="highlight"><pre># Register Python's types PrintableType.register(int) PrintableType.register(float) PrintableType.register(str) </pre></div> </div> Classes should obey the semantics specified by an ABC, but Python can’t check this; it’s up to the class author to understand the ABC’s requirements and to implement the code accordingly. To check whether an object supports a particular interface, you can now write: <div class="highlight-python3"><div class="highlight"><pre>def func(d): if not isinstance(d, collections.MutableMapping): raise ValueError("Mapping object expected, not %r" % d) </pre></div> </div> Don’t feel that you must now begin writing lots of checks as in the above example. Python has a strong tradition of duck-typing, where explicit type-checking is never done and code simply calls methods on an object, trusting that those methods will be there and raising an exception if they aren’t. Be judicious in checking for ABCs and only do it where it’s absolutely necessary. You can write your own ABCs by using <tt class="docutils literal">abc.ABCMeta</tt> as the metaclass in a class definition: <div class="highlight-python3"><div class="highlight"><pre>from abc import ABCMeta, abstractmethod class Drawable(): __metaclass__ = ABCMeta @abstractmethod def draw(self, x, y, scale=1.0): pass def draw_doubled(self, x, y): self.draw(x, y, scale=2.0) class Square(Drawable): def draw(self, x, y, scale): ... </pre></div> </div> In the <tt class="xref docutils literal">Drawable</tt> ABC above, the <tt class="xref docutils literal">draw_doubled()</tt> method renders the object at twice its size and can be implemented in terms of other methods described in <tt class="xref docutils literal">Drawable</tt>. Classes implementing this ABC therefore don’t need to provide their own implementation of <tt class="xref docutils literal">draw_doubled()</tt>, though they can do so. An implementation of <tt class="xref docutils literal">draw()</tt> is necessary, though; the ABC can’t provide a useful generic implementation. You can apply the <tt class="docutils literal">@abstractmethod</tt> decorator to methods such as <tt class="xref docutils literal">draw()</tt> that must be implemented; Python will then raise an exception for classes that don’t define the method. Note that the exception is only raised when you actually try to create an instance of a subclass lacking the method: <div class="highlight-python3"><div class="highlight"><pre>>>> class Circle(Drawable): ... pass ... >>> c = Circle() Traceback (most recent call last): File "<stdin>", line 1, in <module> TypeError: Can't instantiate abstract class Circle with abstract methods draw >>> </pre></div> </div> Abstract data attributes can be declared using the <tt class="docutils literal">@abstractproperty</tt> decorator: <div class="highlight-python3"><div class="highlight"><pre>from abc import abstractproperty ... @abstractproperty def readonly(self): return self._x </pre></div> </div> Subclasses must then define a <a title="readonly" class="reference external" href="../c-api/buffer.html#readonly"><tt class="xref docutils literal">readonly()</tt></a> property. <div class="admonition-see-also admonition seealso"> See also <dl class="last docutils"> <dt><a class="reference external" href="http://www.python.org/dev/peps/pep-3119">PEP 3119</a> - Introducing Abstract Base Classes</dt> <dd>PEP written by Guido van Rossum and Talin. Implemented by Guido van Rossum. Backported to 2.6 by Benjamin Aranguren, with Alex Martelli.</dd> </dl> </div> </div> <div class="section" id="pep-3127-integer-literal-support-and-syntax"> <h2>PEP 3127: Integer Literal Support and Syntax<a class="headerlink" href="#pep-3127-integer-literal-support-and-syntax" title="Permalink to this headline">露</a></h2> Python 3.0 changes the syntax for octal (base-8) integer literals, prefixing them with “0o” or “0O” instead of a leading zero, and adds support for binary (base-2) integer literals, signalled by a “0b” or “0B” prefix. Python 2.6 doesn’t drop support for a leading 0 signalling an octal number, but it does add support for “0o” and “0b”: <div class="highlight-python3"><div class="highlight"><pre>>>> 0o21, 2*8 + 1 (17, 17) >>> 0b101111 47 </pre></div> </div> The <a title="oct" class="reference external" href="../library/functions.html#oct"><tt class="xref docutils literal">oct()</tt></a> builtin still returns numbers prefixed with a leading zero, and a new <a title="bin" class="reference external" href="../library/functions.html#bin"><tt class="xref docutils literal">bin()</tt></a> builtin returns the binary representation for a number: <div class="highlight-python3"><div class="highlight"><pre>>>> oct(42) '052' >>> future_builtins.oct(42) '0o52' >>> bin(173) '0b10101101' </pre></div> </div> The <a title="int" class="reference external" href="../library/functions.html#int"><tt class="xref docutils literal">int()</tt></a> and <tt class="xref docutils literal">long()</tt> builtins will now accept the “0o” and “0b” prefixes when base-8 or base-2 are requested, or when the base argument is zero (signalling that the base used should be determined from the string): <div class="highlight-python3"><div class="highlight"><pre>>>> int ('0o52', 0) 42 >>> int('1101', 2) 13 >>> int('0b1101', 2) 13 >>> int('0b1101', 0) 13 </pre></div> </div> <div class="admonition-see-also admonition seealso"> See also <dl class="last docutils"> <dt><a class="reference external" href="http://www.python.org/dev/peps/pep-3127">PEP 3127</a> - Integer Literal Support and Syntax</dt> <dd>PEP written by Patrick Maupin; backported to 2.6 by Eric Smith.</dd> </dl> </div> </div> <div class="section" id="pep-3129-class-decorators"> <h2>PEP 3129: Class Decorators<a class="headerlink" href="#pep-3129-class-decorators" title="Permalink to this headline">露</a></h2> Decorators have been extended from functions to classes. It’s now legal to write: <div class="highlight-python3"><div class="highlight"><pre>@foo @bar class A: pass </pre></div> </div> This is equivalent to: <div class="highlight-python3"><div class="highlight"><pre>class A: pass A = foo(bar(A)) </pre></div> </div> <div class="admonition-see-also admonition seealso"> See also <dl class="last docutils"> <dt><a class="reference external" href="http://www.python.org/dev/peps/pep-3129">PEP 3129</a> - Class Decorators</dt> <dd>PEP written by Collin Winter.</dd> </dl> </div> </div> <div class="section" id="pep-3141-a-type-hierarchy-for-numbers"> <h2>PEP 3141: A Type Hierarchy for Numbers<a class="headerlink" href="#pep-3141-a-type-hierarchy-for-numbers" title="Permalink to this headline">露</a></h2> Python 3.0 adds several abstract base classes for numeric types inspired by Scheme’s numeric tower. These classes were backported to 2.6 as the <a title="Numeric abstract base classes (Complex, Real, Integral, etc.)." class="reference external" href="../library/numbers.html#module-numbers"><tt class="xref docutils literal">numbers</tt></a> module. The most general ABC is <tt class="xref docutils literal">Number</tt>. It defines no operations at all, and only exists to allow checking if an object is a number by doing <tt class="docutils literal">isinstance(obj, Number)</tt>. <tt class="xref docutils literal">Complex</tt> is a subclass of <tt class="xref docutils literal">Number</tt>. Complex numbers can undergo the basic operations of addition, subtraction, multiplication, division, and exponentiation, and you can retrieve the real and imaginary parts and obtain a number’s conjugate. Python’s built-in complex type is an implementation of <tt class="xref docutils literal">Complex</tt>. <tt class="xref docutils literal">Real</tt> further derives from <tt class="xref docutils literal">Complex</tt>, and adds operations that only work on real numbers: <tt class="xref docutils literal">floor()</tt>, <tt class="xref docutils literal">trunc()</tt>, rounding, taking the remainder mod N, floor division, and comparisons. <tt class="xref docutils literal">Rational</tt> numbers derive from <tt class="xref docutils literal">Real</tt>, have <tt class="xref docutils literal">numerator</tt> and <tt class="xref docutils literal">denominator</tt> properties, and can be converted to floats. Python 2.6 adds a simple rational-number class, <tt class="xref docutils literal">Fraction</tt>, in the <a title="Rational numbers." class="reference external" href="../library/fractions.html#module-fractions"><tt class="xref docutils literal">fractions</tt></a> module. (It’s called <tt class="xref docutils literal">Fraction</tt> instead of <tt class="xref docutils literal">Rational</tt> to avoid a name clash with <a title="numbers.Rational" class="reference external" href="../library/numbers.html#numbers.Rational"><tt class="xref docutils literal">numbers.Rational</tt></a>.) <tt class="xref docutils literal">Integral</tt> numbers derive from <tt class="xref docutils literal">Rational</tt>, and can be shifted left and right with <tt class="docutils literal"><<</tt> and <tt class="docutils literal">>></tt>, combined using bitwise operations such as <tt class="docutils literal">&</tt> and <tt class="docutils literal">|</tt>, and can be used as array indexes and slice boundaries. In Python 3.0, the PEP slightly redefines the existing builtins <a title="round" class="reference external" href="../library/functions.html#round"><tt class="xref docutils literal">round()</tt></a>, <a title="math.floor" class="reference external" href="../library/math.html#math.floor"><tt class="xref docutils literal">math.floor()</tt></a>, <a title="math.ceil" class="reference external" href="../library/math.html#math.ceil"><tt class="xref docutils literal">math.ceil()</tt></a>, and adds a new one, <a title="math.trunc" class="reference external" href="../library/math.html#math.trunc"><tt class="xref docutils literal">math.trunc()</tt></a>, that’s been backported to Python 2.6. <a title="math.trunc" class="reference external" href="../library/math.html#math.trunc"><tt class="xref docutils literal">math.trunc()</tt></a> rounds toward zero, returning the closest <tt class="xref docutils literal">Integral</tt> that’s between the function’s argument and zero. <div class="admonition-see-also admonition seealso"> See also <dl class="docutils"> <dt><a class="reference external" href="http://www.python.org/dev/peps/pep-3141">PEP 3141</a> - A Type Hierarchy for Numbers</dt> <dd>PEP written by Jeffrey Yasskin.</dd> </dl> <a class="reference external" href="http://www.gnu.org/software/guile/manual/html_node/Numerical-Tower.html#Numerical-Tower">Scheme’s numerical tower</a>, from the Guile manual. <a class="reference external" href="http://schemers.org/Documents/Standards/R5RS/HTML/r5rs-Z-H-9.html#%_sec_6.2">Scheme’s number datatypes</a> from the R5RS Scheme specification. </div> <div class="section" id="the-fractions-module"> <h3>The <a title="Rational numbers." class="reference external" href="../library/fractions.html#module-fractions"><tt class="xref docutils literal">fractions</tt></a> Module<a class="headerlink" href="#the-fractions-module" title="Permalink to this headline">露</a></h3> To fill out the hierarchy of numeric types, the <a title="Rational numbers." class="reference external" href="../library/fractions.html#module-fractions"><tt class="xref docutils literal">fractions</tt></a> module provides a rational-number class. Rational numbers store their values as a numerator and denominator forming a fraction, and can exactly represent numbers such as <tt class="docutils literal">2/3</tt> that floating-point numbers can only approximate. The <tt class="xref docutils literal">Fraction</tt> constructor takes two <tt class="xref docutils literal">Integral</tt> values that will be the numerator and denominator of the resulting fraction. <div class="highlight-python3"><div class="highlight"><pre>>>> from fractions import Fraction >>> a = Fraction(2, 3) >>> b = Fraction(2, 5) >>> float(a), float(b) (0.66666666666666663, 0.40000000000000002) >>> a+b Fraction(16, 15) >>> a/b Fraction(5, 3) </pre></div> </div> For converting floating-point numbers to rationals, the float type now has an <tt class="xref docutils literal">as_integer_ratio()</tt> method that returns the numerator and denominator for a fraction that evaluates to the same floating-point value: <div class="highlight-python3"><div class="highlight"><pre>>>> (2.5) .as_integer_ratio() (5, 2) >>> (3.1415) .as_integer_ratio() (7074029114692207L, 2251799813685248L) >>> (1./3) .as_integer_ratio() (6004799503160661L, 18014398509481984L) </pre></div> </div> Note that values that can only be approximated by floating-point numbers, such as 1./3, are not simplified to the number being approximated; the fraction attempts to match the floating-point value exactly. The <a title="Rational numbers." class="reference external" href="../library/fractions.html#module-fractions"><tt class="xref docutils literal">fractions</tt></a> module is based upon an implementation by Sjoerd Mullender that was in Python’s <tt class="docutils literal">Demo/classes/</tt> directory for a long time. This implementation was significantly updated by Jeffrey Yasskin. </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">露</a></h2> Some smaller changes made to the core Python language are: <ul> <li>The <a title="hasattr" class="reference external" href="../library/functions.html#hasattr"><tt class="xref docutils literal">hasattr()</tt></a> function was catching and ignoring all errors, under the assumption that they meant a <a title="object.__getattr__" class="reference external" href="../reference/datamodel.html#object.__getattr__"><tt class="xref docutils literal">__getattr__()</tt></a> method was failing somehow and the return value of <a title="hasattr" class="reference external" href="../library/functions.html#hasattr"><tt class="xref docutils literal">hasattr()</tt></a> would therefore be <tt class="xref docutils literal">False</tt>. This logic shouldn’t be applied to <a title="KeyboardInterrupt" class="reference external" href="../library/exceptions.html#KeyboardInterrupt"><tt class="xref docutils literal">KeyboardInterrupt</tt></a> and <a title="SystemExit" class="reference external" href="../library/exceptions.html#SystemExit"><tt class="xref docutils literal">SystemExit</tt></a>, however; Python 2.6 will no longer discard such exceptions when <a title="hasattr" class="reference external" href="../library/functions.html#hasattr"><tt class="xref docutils literal">hasattr()</tt></a> encounters them. (Fixed by Benjamin Peterson; <a class="reference external" href="http://bugs.python.org/issue2196">issue 2196</a>.) </li> <li>When calling a function using the <tt class="docutils literal">**</tt> syntax to provide keyword arguments, you are no longer required to use a Python dictionary; any mapping will now work: <div class="highlight-python3"><div class="highlight"><pre>>>> def f(**kw): ... print sorted(kw) ... >>> ud=UserDict.UserDict() >>> ud['a'] = 1 >>> ud['b'] = 'string' >>> f(**ud) ['a', 'b'] </pre></div> </div> (Contributed by Alexander Belopolsky; <a class="reference external" href="http://bugs.python.org/issue1686487">issue 1686487</a>.) It’s also become legal to provide keyword arguments after a <tt class="docutils literal">*args</tt> argument to a function call. <div class="highlight-python3"><div class="highlight"><pre>>>> def f(*args, **kw): ... print args, kw ... >>> f(1,2,3, *(4,5,6), keyword=13) (1, 2, 3, 4, 5, 6) {'keyword': 13} </pre></div> </div> Previously this would have been a syntax error. (Contributed by Amaury Forgeot d’Arc; <a class="reference external" href="http://bugs.python.org/issue3473">issue 3473</a>.) </li> <li>A new builtin, <tt class="docutils literal">next(iterator, [default])</tt> returns the next item from the specified iterator. If the default argument is supplied, it will be returned if iterator has been exhausted; otherwise, the <a title="StopIteration" class="reference external" href="../library/exceptions.html#StopIteration"><tt class="xref docutils literal">StopIteration</tt></a> exception will be raised. (Backported in <a class="reference external" href="http://bugs.python.org/issue2719">issue 2719</a>.) </li> <li>Tuples now have <tt class="xref docutils literal">index()</tt> and <tt class="xref docutils literal">count()</tt> methods matching the list type’s <tt class="xref docutils literal">index()</tt> and <tt class="xref docutils literal">count()</tt> methods: <div class="highlight-python3"><div class="highlight"><pre>>>> t = (0,1,2,3,4,0,1,2) >>> t.index(3) 3 >>> t.count(0) 2 </pre></div> </div> (Contributed by Raymond Hettinger) </li> <li>The built-in types now have improved support for extended slicing syntax, accepting various combinations of <tt class="docutils literal">(start, stop, step)</tt>. Previously, the support was partial and certain corner cases wouldn’t work. (Implemented by Thomas Wouters.) </li> <li>Properties now have three attributes, <tt class="xref docutils literal">getter</tt>, <tt class="xref docutils literal">setter</tt> and <tt class="xref docutils literal">deleter</tt>, that are decorators providing useful shortcuts for adding a getter, setter or deleter function to an existing property. You would use them like this: <div class="highlight-python3"><div class="highlight"><pre>class C(object): @property def x(self): return self._x @x.setter def x(self, value): self._x = value @x.deleter def x(self): del self._x class D(C): @C.x.getter def x(self): return self._x * 2 @x.setter def x(self, value): self._x = value / 2 </pre></div> </div> </li> <li>Several methods of the built-in set types now accept multiple iterables: <tt class="xref docutils literal">intersection()</tt>, <tt class="xref docutils literal">intersection_update()</tt>, <tt class="xref docutils literal">union()</tt>, <tt class="xref docutils literal">update()</tt>, <tt class="xref docutils literal">difference()</tt> and <tt class="xref docutils literal">difference_update()</tt>. <div class="highlight-python3"><div class="highlight"><pre>>>> s=set('1234567890') >>> s.intersection('abc123', 'cdf246') # Intersection between all inputs set(['2']) >>> s.difference('246', '789') set(['1', '0', '3', '5']) </pre></div> </div> (Contributed by Raymond Hettinger.) </li> <li>Many floating-point features were added. The <a title="float" class="reference external" href="../library/functions.html#float"><tt class="xref docutils literal">float()</tt></a> function will now turn the string <tt class="docutils literal">nan</tt> into an IEEE 754 Not A Number value, and <tt class="docutils literal">+inf</tt> and <tt class="docutils literal">-inf</tt> into positive or negative infinity. This works on any platform with IEEE 754 semantics. (Contributed by Christian Heimes; <a class="reference external" href="http://bugs.python.org/issue1635">issue 1635</a>.) Other functions in the <a title="Mathematical functions (sin() etc.)." class="reference external" href="../library/math.html#module-math"><tt class="xref docutils literal">math</tt></a> module, <tt class="xref docutils literal">isinf()</tt> and <tt class="xref docutils literal">isnan()</tt>, return true if their floating-point argument is infinite or Not A Number. (<a class="reference external" href="http://bugs.python.org/issue1640">issue 1640</a>) Conversion functions were added to convert floating-point numbers into hexadecimal strings (<a class="reference external" href="http://bugs.python.org/issue3008">issue 3008</a>). These functions convert floats to and from a string representation without introducing rounding errors from the conversion between decimal and binary. Floats have a <a title="hex" class="reference external" href="../library/functions.html#hex"><tt class="xref docutils literal">hex()</tt></a> method that returns a string representation, and the <tt class="docutils literal">float.fromhex()</tt> method converts a string back into a number: <div class="highlight-python3"><div class="highlight"><pre>>>> a = 3.75 >>> a.hex() '0x1.e000000000000p+1' >>> float.fromhex('0x1.e000000000000p+1') 3.75 >>> b=1./3 >>> b.hex() '0x1.5555555555555p-2' </pre></div> </div> </li> <li>A numerical nicety: when creating a complex number from two floats on systems that support signed zeros (-0 and +0), the <a title="complex" class="reference external" href="../library/functions.html#complex"><tt class="xref docutils literal">complex()</tt></a> constructor will now preserve the sign of the zero. (Fixed by Mark T. Dickinson; <a class="reference external" href="http://bugs.python.org/issue1507">issue 1507</a>.) </li> <li>Classes that inherit a <a title="object.__hash__" class="reference external" href="../reference/datamodel.html#object.__hash__"><tt class="xref docutils literal">__hash__()</tt></a> method from a parent class can set <tt class="docutils literal">__hash__ = None</tt> to indicate that the class isn’t hashable. This will make <tt class="docutils literal">hash(obj)</tt> raise a <a title="TypeError" class="reference external" href="../library/exceptions.html#TypeError"><tt class="xref docutils literal">TypeError</tt></a> and the class will not be indicated as implementing the <tt class="xref docutils literal">Hashable</tt> ABC. You should do this when you’ve defined a <tt class="xref docutils literal">__cmp__()</tt> or <a title="object.__eq__" class="reference external" href="../reference/datamodel.html#object.__eq__"><tt class="xref docutils literal">__eq__()</tt></a> method that compares objects by their value rather than by identity. All objects have a default hash method that uses <tt class="docutils literal">id(obj)</tt> as the hash value. There’s no tidy way to remove the <a title="object.__hash__" class="reference external" href="../reference/datamodel.html#object.__hash__"><tt class="xref docutils literal">__hash__()</tt></a> method inherited from a parent class, so assigning <tt class="xref docutils literal">None</tt> was implemented as an override. At the C level, extensions can set <tt class="docutils literal">tp_hash</tt> to <a title="PyObject_HashNotImplemented" class="reference external" href="../c-api/object.html#PyObject_HashNotImplemented"><tt class="xref docutils literal">PyObject_HashNotImplemented()</tt></a>. (Fixed by Nick Coghlan and Amaury Forgeot d’Arc; <a class="reference external" href="http://bugs.python.org/issue2235">issue 2235</a>.) </li> <li>The <a title="GeneratorExit" class="reference external" href="../library/exceptions.html#GeneratorExit"><tt class="xref docutils literal">GeneratorExit</tt></a> exception now subclasses <a title="BaseException" class="reference external" href="../library/exceptions.html#BaseException"><tt class="xref docutils literal">BaseException</tt></a> instead of <a title="Exception" class="reference external" href="../library/exceptions.html#Exception"><tt class="xref docutils literal">Exception</tt></a>. This means that an exception handler that does <tt class="docutils literal">except Exception:</tt> will not inadvertently catch <a title="GeneratorExit" class="reference external" href="../library/exceptions.html#GeneratorExit"><tt class="xref docutils literal">GeneratorExit</tt></a>. (Contributed by Chad Austin; <a class="reference external" href="http://bugs.python.org/issue1537">issue 1537</a>.) </li> <li>Generator objects now have a <tt class="xref docutils literal">gi_code</tt> attribute that refers to the original code object backing the generator. (Contributed by Collin Winter; <a class="reference external" href="http://bugs.python.org/issue1473257">issue 1473257</a>.) </li> <li>The <a title="compile" class="reference external" href="../library/functions.html#compile"><tt class="xref docutils literal">compile()</tt></a> built-in function now accepts keyword arguments as well as positional parameters. (Contributed by Thomas Wouters; <a class="reference external" href="http://bugs.python.org/issue1444529">issue 1444529</a>.) </li> <li>The <a title="complex" class="reference external" href="../library/functions.html#complex"><tt class="xref docutils literal">complex()</tt></a> constructor now accepts strings containing parenthesized complex numbers, meaning that <tt class="docutils literal">complex(repr(cplx))</tt> will now round-trip values. For example, <tt class="docutils literal">complex('(3+4j)')</tt> now returns the value (3+4j). (<a class="reference external" href="http://bugs.python.org/issue1491866">issue 1491866</a>) </li> <li>The string <tt class="xref docutils literal">translate()</tt> method now accepts <tt class="xref docutils literal">None</tt> as the translation table parameter, which is treated as the identity transformation. This makes it easier to carry out operations that only delete characters. (Contributed by Bengt Richter and implemented by Raymond Hettinger; <a class="reference external" href="http://bugs.python.org/issue1193128">issue 1193128</a>.) </li> <li>The built-in <a title="dir" class="reference external" href="../library/functions.html#dir"><tt class="xref docutils literal">dir()</tt></a> function now checks for a <a title="object.__dir__" class="reference external" href="../reference/datamodel.html#object.__dir__"><tt class="xref docutils literal">__dir__()</tt></a> method on the objects it receives. This method must return a list of strings containing the names of valid attributes for the object, and lets the object control the value that <a title="dir" class="reference external" href="../library/functions.html#dir"><tt class="xref docutils literal">dir()</tt></a> produces. Objects that have <a title="object.__getattr__" class="reference external" href="../reference/datamodel.html#object.__getattr__"><tt class="xref docutils literal">__getattr__()</tt></a> or <a title="object.__getattribute__" class="reference external" href="../reference/datamodel.html#object.__getattribute__"><tt class="xref docutils literal">__getattribute__()</tt></a> methods can use this to advertise pseudo-attributes they will honor. (<a class="reference external" href="http://bugs.python.org/issue1591665">issue 1591665</a>) </li> <li>Instance method objects have new attributes for the object and function comprising the method; the new synonym for <tt class="xref docutils literal">im_self</tt> is <tt class="xref docutils literal">__self__</tt>, and <tt class="xref docutils literal">im_func</tt> is also available as <tt class="xref docutils literal">__func__</tt>. The old names are still supported in Python 2.6, but are gone in 3.0. </li> <li>An obscure change: when you use the <a title="locals" class="reference external" href="../library/functions.html#locals"><tt class="xref docutils literal">locals()</tt></a> function inside a <a class="reference external" href="../reference/compound_stmts.html#class"><tt class="xref docutils literal">class</tt></a> statement, the resulting dictionary no longer returns free variables. (Free variables, in this case, are variables referenced in the <a class="reference external" href="../reference/compound_stmts.html#class"><tt class="xref docutils literal">class</tt></a> statement that aren’t attributes of the class.) </li> </ul> <div class="section" id="optimizations"> <h3>Optimizations<a class="headerlink" href="#optimizations" title="Permalink to this headline">露</a></h3> <ul> <li>The <a title="Issue warning messages and control their disposition." class="reference external" href="../library/warnings.html#module-warnings"><tt class="xref docutils literal">warnings</tt></a> module has been rewritten in C. This makes it possible to invoke warnings from the parser, and may also make the interpreter’s startup faster. (Contributed by Neal Norwitz and Brett Cannon; <a class="reference external" href="http://bugs.python.org/issue1631171">issue 1631171</a>.) </li> <li>Type objects now have a cache of methods that can reduce the work required to find the correct method implementation for a particular class; once cached, the interpreter doesn’t need to traverse base classes to figure out the right method to call. The cache is cleared if a base class or the class itself is modified, so the cache should remain correct even in the face of Python’s dynamic nature. (Original optimization implemented by Armin Rigo, updated for Python 2.6 by Kevin Jacobs; <a class="reference external" href="http://bugs.python.org/issue1700288">issue 1700288</a>.) By default, this change is only applied to types that are included with the Python core. Extension modules may not necessarily be compatible with this cache, so they must explicitly add <tt class="xref docutils literal">Py_TPFLAGS_HAVE_VERSION_TAG</tt> to the module’s <tt class="docutils literal">tp_flags</tt> field to enable the method cache. (To be compatible with the method cache, the extension module’s code must not directly access and modify the <tt class="docutils literal">tp_dict</tt> member of any of the types it implements. Most modules don’t do this, but it’s impossible for the Python interpreter to determine that. See <a class="reference external" href="http://bugs.python.org/issue1878">issue 1878</a> for some discussion.) </li> <li>Function calls that use keyword arguments are significantly faster by doing a quick pointer comparison, usually saving the time of a full string comparison. (Contributed by Raymond Hettinger, after an initial implementation by Antoine Pitrou; <a class="reference external" href="http://bugs.python.org/issue1819">issue 1819</a>.) </li> <li>All of the functions in the <a title="Interpret bytes as packed binary data." class="reference external" href="../library/struct.html#module-struct"><tt class="xref docutils literal">struct</tt></a> module have been rewritten in C, thanks to work at the Need For Speed sprint. (Contributed by Raymond Hettinger.) </li> <li>Some of the standard built-in types now set a bit in their type objects. This speeds up checking whether an object is a subclass of one of these types. (Contributed by Neal Norwitz.) </li> <li>Unicode strings now use faster code for detecting whitespace and line breaks; this speeds up the <tt class="xref docutils literal">split()</tt> method by about 25% and <tt class="xref docutils literal">splitlines()</tt> by 35%. (Contributed by Antoine Pitrou.) Memory usage is reduced by using pymalloc for the Unicode string’s data. </li> <li>The <tt class="docutils literal">with</tt> statement now stores the <a title="object.__exit__" class="reference external" href="../reference/datamodel.html#object.__exit__"><tt class="xref docutils literal">__exit__()</tt></a> method on the stack, producing a small speedup. (Implemented by Jeffrey Yasskin.) </li> <li>To reduce memory usage, the garbage collector will now clear internal free lists when garbage-collecting the highest generation of objects. This may return memory to the operating system sooner. </li> </ul> </div> <div class="section" id="interpreter-changes"> <h3>Interpreter Changes<a class="headerlink" href="#interpreter-changes" title="Permalink to this headline">露</a></h3> Two command-line options have been reserved for use by other Python implementations. The <a class="reference external" href="../using/cmdline.html#cmdoption-trace-J">-J</a> switch has been reserved for use by Jython for Jython-specific options, such as switches that are passed to the underlying JVM. <a class="reference external" href="../using/cmdline.html#cmdoption-trace-X">-X</a> has been reserved for options specific to a particular implementation of Python such as CPython, Jython, or IronPython. If either option is used with Python 2.6, the interpreter will report that the option isn’t currently used. Python can now be prevented from writing <tt class="docutils literal">.pyc</tt> or <tt class="docutils literal">.pyo</tt> files by supplying the <a class="reference external" href="../using/cmdline.html#cmdoption-trace-B">-B</a> switch to the Python interpreter, or by setting the <a class="reference external" href="../using/cmdline.html#envvar-PYTHONDONTWRITEBYTECODE">PYTHONDONTWRITEBYTECODE</a> environment variable before running the interpreter. This setting is available to Python programs as the <tt class="docutils literal">sys.dont_write_bytecode</tt> variable, and Python code can change the value to modify the interpreter’s behaviour. (Contributed by Neal Norwitz and Georg Brandl.) The encoding used for standard input, output, and standard error can be specified by setting the <a class="reference external" href="../using/cmdline.html#envvar-PYTHONIOENCODING">PYTHONIOENCODING</a> environment variable before running the interpreter. The value should be a string in the form <tt class="docutils literal"><encoding></tt> or <tt class="docutils literal"><encoding>:<errorhandler></tt>. The encoding part specifies the encoding’s name, e.g. <tt class="docutils literal">utf-8</tt> or <tt class="docutils literal">latin-1</tt>; the optional errorhandler part specifies what to do with characters that can’t be handled by the encoding, and should be one of “error”, “ignore”, or “replace”. (Contributed by Martin von Loewis.) </div> </div> <div class="section" id="new-and-improved-modules"> <h2>New and Improved Modules<a class="headerlink" href="#new-and-improved-modules" title="Permalink to this headline">露</a></h2> As in every release, Python’s standard library received a number of enhancements and bug fixes. Here’s a partial list of the most notable changes, sorted alphabetically by module name. Consult the <tt class="docutils literal">Misc/NEWS</tt> file in the source tree for a more complete list of changes, or look through the Subversion logs for all the details. <ul> <li>The <a title="A base class for developing asynchronous socket handling services." class="reference external" href="../library/asyncore.html#module-asyncore"><tt class="xref docutils literal">asyncore</tt></a> and <a title="Support for asynchronous command/response protocols." class="reference external" href="../library/asynchat.html#module-asynchat"><tt class="xref docutils literal">asynchat</tt></a> modules are being actively maintained again, and a number of patches and bugfixes were applied. (Maintained by Josiah Carlson; see <a class="reference external" href="http://bugs.python.org/issue1736190">issue 1736190</a> for one patch.) </li> <li>The <tt class="xref docutils literal">bsddb</tt> module also has a new maintainer, Jes煤s Cea, and the package is now available as a standalone package. The web page for the package is <a class="reference external" href="http://www.jcea.es/programacion/pybsddb.htm">www.jcea.es/programacion/pybsddb.htm</a>. The plan is to remove the package from the standard library in Python 3.0, because its pace of releases is much more frequent than Python’s. The <tt class="xref docutils literal">bsddb.dbshelve</tt> module now uses the highest pickling protocol available, instead of restricting itself to protocol 1. (Contributed by W. Barnes.) </li> <li>The <a title="Helpers for running Python scripts via the Common Gateway Interface." class="reference external" href="../library/cgi.html#module-cgi"><tt class="xref docutils literal">cgi</tt></a> module will now read variables from the query string of an HTTP POST request. This makes it possible to use form actions with URLs that include query strings such as “/cgi-bin/add.py?category=1”. (Contributed by Alexandre Fiori and Nubis; <a class="reference external" href="http://bugs.python.org/issue1817">issue 1817</a>.) The <tt class="xref docutils literal">parse_qs()</tt> and <tt class="xref docutils literal">parse_qsl()</tt> functions have been relocated from the <a title="Helpers for running Python scripts via the Common Gateway Interface." class="reference external" href="../library/cgi.html#module-cgi"><tt class="xref docutils literal">cgi</tt></a> module to the <tt class="xref docutils literal">urlparse</tt> module. The versions still available in the <a title="Helpers for running Python scripts via the Common Gateway Interface." class="reference external" href="../library/cgi.html#module-cgi"><tt class="xref docutils literal">cgi</tt></a> module will trigger <a title="PendingDeprecationWarning" class="reference external" href="../library/exceptions.html#PendingDeprecationWarning"><tt class="xref docutils literal">PendingDeprecationWarning</tt></a> messages in 2.6 (<a class="reference external" href="http://bugs.python.org/issue600362">issue 600362</a>). </li> <li>The <a title="Mathematical functions for complex numbers." class="reference external" href="../library/cmath.html#module-cmath"><tt class="xref docutils literal">cmath</tt></a> module underwent extensive revision, contributed by Mark Dickinson and Christian Heimes. Five new functions were added: <ul class="simple"> <li><tt class="xref docutils literal">polar()</tt> converts a complex number to polar form, returning the modulus and argument of the complex number.</li> <li><tt class="xref docutils literal">rect()</tt> does the opposite, turning a modulus, argument pair back into the corresponding complex number.</li> <li><tt class="xref docutils literal">phase()</tt> returns the argument (also called the angle) of a complex number.</li> <li><tt class="xref docutils literal">isnan()</tt> returns True if either the real or imaginary part of its argument is a NaN.</li> <li><tt class="xref docutils literal">isinf()</tt> returns True if either the real or imaginary part of its argument is infinite.</li> </ul> The revisions also improved the numerical soundness of the <a title="Mathematical functions for complex numbers." class="reference external" href="../library/cmath.html#module-cmath"><tt class="xref docutils literal">cmath</tt></a> module. For all functions, the real and imaginary parts of the results are accurate to within a few units of least precision (ulps) whenever possible. See <a class="reference external" href="http://bugs.python.org/issue1381">issue 1381</a> for the details. The branch cuts for <tt class="xref docutils literal">asinh()</tt>, <tt class="xref docutils literal">atanh()</tt>: and <tt class="xref docutils literal">atan()</tt> have also been corrected. The tests for the module have been greatly expanded; nearly 2000 new test cases exercise the algebraic functions. On IEEE 754 platforms, the <a title="Mathematical functions for complex numbers." class="reference external" href="../library/cmath.html#module-cmath"><tt class="xref docutils literal">cmath</tt></a> module now handles IEEE 754 special values and floating-point exceptions in a manner consistent with Annex ‘G’ of the C99 standard. </li> <li>A new data type in the <a title="Container datatypes" class="reference external" href="../library/collections.html#module-collections"><tt class="xref docutils literal">collections</tt></a> module: <tt class="xref docutils literal">namedtuple(typename, fieldnames)</tt> is a factory function that creates subclasses of the standard tuple whose fields are accessible by name as well as index. For example: <div class="highlight-python3"><div class="highlight"><pre>>>> var_type = collections.namedtuple('variable', ... 'id name type size') >>> # Names are separated by spaces or commas. >>> # 'id, name, type, size' would also work. >>> var_type._fields ('id', 'name', 'type', 'size') >>> var = var_type(1, 'frequency', 'int', 4) >>> print var[0], var.id # Equivalent 1 1 >>> print var[2], var.type # Equivalent int int >>> var._asdict() {'size': 4, 'type': 'int', 'id': 1, 'name': 'frequency'} >>> v2 = var._replace(name='amplitude') >>> v2 variable(id=1, name='amplitude', type='int', size=4) </pre></div> </div> Several places in the standard library that returned tuples have been modified to return <tt class="xref docutils literal">namedtuple</tt> instances. For example, the <tt class="xref docutils literal">Decimal.as_tuple()</tt> method now returns a named tuple with <tt class="xref docutils literal">sign</tt>, <tt class="xref docutils literal">digits</tt>, and <tt class="xref docutils literal">exponent</tt> fields. (Contributed by Raymond Hettinger.) </li> <li>Another change to the <a title="Container datatypes" class="reference external" href="../library/collections.html#module-collections"><tt class="xref docutils literal">collections</tt></a> module is that the <tt class="xref docutils literal">deque</tt> type now supports an optional maxlen parameter; if supplied, the deque’s size will be restricted to no more than maxlen items. Adding more items to a full deque causes old items to be discarded. <div class="highlight-python3"><div class="highlight"><pre>>>> from collections import deque >>> dq=deque(maxlen=3) >>> dq deque([], maxlen=3) >>> dq.append(1) ; dq.append(2) ; dq.append(3) >>> dq deque([1, 2, 3], maxlen=3) >>> dq.append(4) >>> dq deque([2, 3, 4], maxlen=3) </pre></div> </div> (Contributed by Raymond Hettinger.) </li> <li>The <tt class="xref docutils literal">Cookie</tt> module’s <tt class="xref docutils literal">Morsel</tt> objects now support an <tt class="xref docutils literal">httponly</tt> attribute. In some browsers. cookies with this attribute set cannot be accessed or manipulated by JavaScript code. (Contributed by Arvin Schnell; <a class="reference external" href="http://bugs.python.org/issue1638033">issue 1638033</a>.) </li> <li>A new window method in the <a title="(Unix) An interface to the curses library, providing portable terminal handling." class="reference external" href="../library/curses.html#module-curses"><tt class="xref docutils literal">curses</tt></a> module, <tt class="xref docutils literal">chgat()</tt>, changes the display attributes for a certain number of characters on a single line. (Contributed by Fabian Kreutz.) <div class="highlight-python3"><div class="highlight"><pre># Boldface text starting at y=0,x=21 # and affecting the rest of the line. stdscr.chgat(0, 21, curses.A_BOLD) </pre></div> </div> The <tt class="xref docutils literal">Textbox</tt> class in the <a title="Emacs-like input editing in a curses window." class="reference external" href="../library/curses.html#module-curses.textpad"><tt class="xref docutils literal">curses.textpad</tt></a> module now supports editing in insert mode as well as overwrite mode. Insert mode is enabled by supplying a true value for the insert_mode parameter when creating the <tt class="xref docutils literal">Textbox</tt> instance. </li> <li>The <a title="Basic date and time types." class="reference external" href="../library/datetime.html#module-datetime"><tt class="xref docutils literal">datetime</tt></a> module’s <tt class="xref docutils literal">strftime()</tt> methods now support a <tt class="docutils literal">%f</tt> format code that expands to the number of microseconds in the object, zero-padded on the left to six places. (Contributed by Skip Montanaro; <a class="reference external" href="http://bugs.python.org/issue1158">issue 1158</a>.) </li> <li>The <a title="Implementation of the General Decimal Arithmetic Specification." class="reference external" href="../library/decimal.html#module-decimal"><tt class="xref docutils literal">decimal</tt></a> module was updated to version 1.66 of <a class="reference external" href="http://www2.hursley.ibm.com/decimal/decarith.html">the General Decimal Specification</a>. New features include some methods for some basic mathematical functions such as <tt class="xref docutils literal">exp()</tt> and <tt class="xref docutils literal">log10()</tt>: <div class="highlight-python3"><div class="highlight"><pre>>>> Decimal(1).exp() Decimal("2.718281828459045235360287471") >>> Decimal("2.7182818").ln() Decimal("0.9999999895305022877376682436") >>> Decimal(1000).log10() Decimal("3") </pre></div> </div> The <tt class="xref docutils literal">as_tuple()</tt> method of <tt class="xref docutils literal">Decimal</tt> objects now returns a named tuple with <tt class="xref docutils literal">sign</tt>, <tt class="xref docutils literal">digits</tt>, and <tt class="xref docutils literal">exponent</tt> fields. (Implemented by Facundo Batista and Mark Dickinson. Named tuple support added by Raymond Hettinger.) </li> <li>The <a title="Helpers for computing differences between objects." class="reference external" href="../library/difflib.html#module-difflib"><tt class="xref docutils literal">difflib</tt></a> module’s <tt class="xref docutils literal">SequenceMatcher</tt> class now returns named tuples representing matches, with <tt class="xref docutils literal">a</tt>, <tt class="xref docutils literal">b</tt>, and <tt class="xref docutils literal">size</tt> attributes. (Contributed by Raymond Hettinger.) </li> <li>An optional <tt class="docutils literal">timeout</tt> parameter, specifying a timeout measured in seconds, was added to the <a title="ftplib.FTP" class="reference external" href="../library/ftplib.html#ftplib.FTP"><tt class="xref docutils literal">ftplib.FTP</tt></a> class constructor as well as the <tt class="xref docutils literal">connect()</tt> method. (Added by Facundo Batista.) Also, the <tt class="xref docutils literal">FTP</tt> class’s <tt class="xref docutils literal">storbinary()</tt> and <tt class="xref docutils literal">storlines()</tt> now take an optional callback parameter that will be called with each block of data after the data has been sent. (Contributed by Phil Schwartz; <a class="reference external" href="http://bugs.python.org/issue1221598">issue 1221598</a>.) </li> <li>The <tt class="xref docutils literal">reduce()</tt> built-in function is also available in the <a title="Higher order functions and operations on callable objects." class="reference external" href="../library/functools.html#module-functools"><tt class="xref docutils literal">functools</tt></a> module. In Python 3.0, the builtin has been dropped and <tt class="xref docutils literal">reduce()</tt> is only available from <a title="Higher order functions and operations on callable objects." class="reference external" href="../library/functools.html#module-functools"><tt class="xref docutils literal">functools</tt></a>; currently there are no plans to drop the builtin in the 2.x series. (Patched by Christian Heimes; <a class="reference external" href="http://bugs.python.org/issue1739906">issue 1739906</a>.) </li> <li>When possible, the <a title="Portable reading of passwords and retrieval of the userid." class="reference external" href="../library/getpass.html#module-getpass"><tt class="xref docutils literal">getpass</tt></a> module will now use <tt class="docutils literal">/dev/tty</tt> to print a prompt message and read the password, falling back to standard error and standard input. If the password may be echoed to the terminal, a warning is printed before the prompt is displayed. (Contributed by Gregory P. Smith.) </li> <li>The <a title="glob.glob" class="reference external" href="../library/glob.html#glob.glob"><tt class="xref docutils literal">glob.glob()</tt></a> function can now return Unicode filenames if a Unicode path was used and Unicode filenames are matched within the directory. (<a class="reference external" href="http://bugs.python.org/issue1001604">issue 1001604</a>) </li> <li>A new function in the <a title="Heap queue algorithm (a.k.a. priority queue)." class="reference external" href="../library/heapq.html#module-heapq"><tt class="xref docutils literal">heapq</tt></a> module, <tt class="docutils literal">merge(iter1, iter2, ...)</tt>, takes any number of iterables returning data in sorted order, and returns a new generator that returns the contents of all the iterators, also in sorted order. For example: <div class="highlight-python3"><div class="highlight"><pre>>>> list(heapq.merge([1, 3, 5, 9], [2, 8, 16])) [1, 2, 3, 5, 8, 9, 16] </pre></div> </div> Another new function, <tt class="docutils literal">heappushpop(heap, item)</tt>, pushes item onto heap, then pops off and returns the smallest item. This is more efficient than making a call to <tt class="xref docutils literal">heappush()</tt> and then <tt class="xref docutils literal">heappop()</tt>. <a title="Heap queue algorithm (a.k.a. priority queue)." class="reference external" href="../library/heapq.html#module-heapq"><tt class="xref docutils literal">heapq</tt></a> is now implemented to only use less-than comparison, instead of the less-than-or-equal comparison it previously used. This makes <a title="Heap queue algorithm (a.k.a. priority queue)." class="reference external" href="../library/heapq.html#module-heapq"><tt class="xref docutils literal">heapq</tt></a>‘s usage of a type match the <tt class="xref docutils literal">list.sort()</tt> method. (Contributed by Raymond Hettinger.) </li> <li>An optional <tt class="docutils literal">timeout</tt> parameter, specifying a timeout measured in seconds, was added to the <tt class="xref docutils literal">httplib.HTTPConnection</tt> and <tt class="xref docutils literal">HTTPSConnection</tt> class constructors. (Added by Facundo Batista.) </li> <li>Most of the <a title="Extract information and source code from live objects." class="reference external" href="../library/inspect.html#module-inspect"><tt class="xref docutils literal">inspect</tt></a> module’s functions, such as <tt class="xref docutils literal">getmoduleinfo()</tt> and <tt class="xref docutils literal">getargs()</tt>, now return named tuples. In addition to behaving like tuples, the elements of the return value can also be accessed as attributes. (Contributed by Raymond Hettinger.) Some new functions in the module include <tt class="xref docutils literal">isgenerator()</tt>, <tt class="xref docutils literal">isgeneratorfunction()</tt>, and <tt class="xref docutils literal">isabstract()</tt>. </li> <li>The <a title="Functions creating iterators for efficient looping." class="reference external" href="../library/itertools.html#module-itertools"><tt class="xref docutils literal">itertools</tt></a> module gained several new functions. <tt class="docutils literal">izip_longest(iter1, iter2, ...[, fillvalue])</tt> makes tuples from each of the elements; if some of the iterables are shorter than others, the missing values are set to fillvalue. For example: <div class="highlight-python3"><div class="highlight"><pre>>>> tuple(itertools.izip_longest([1,2,3], [1,2,3,4,5])) ((1, 1), (2, 2), (3, 3), (None, 4), (None, 5)) </pre></div> </div> <tt class="docutils literal">product(iter1, iter2, ..., [repeat=N])</tt> returns the Cartesian product of the supplied iterables, a set of tuples containing every possible combination of the elements returned from each iterable. <div class="highlight-python3"><div class="highlight"><pre>>>> list(itertools.product([1,2,3], [4,5,6])) [(1, 4), (1, 5), (1, 6), (2, 4), (2, 5), (2, 6), (3, 4), (3, 5), (3, 6)] </pre></div> </div> The optional repeat keyword argument is used for taking the product of an iterable or a set of iterables with themselves, repeated N times. With a single iterable argument, N-tuples are returned: <div class="highlight-python3"><div class="highlight"><pre>>>> list(itertools.product([1,2], repeat=3)) [(1, 1, 1), (1, 1, 2), (1, 2, 1), (1, 2, 2), (2, 1, 1), (2, 1, 2), (2, 2, 1), (2, 2, 2)] </pre></div> </div> With two iterables, 2N-tuples are returned. <div class="highlight-python3"><div class="highlight"><pre>>>> list(itertools.product([1,2], [3,4], repeat=2)) [(1, 3, 1, 3), (1, 3, 1, 4), (1, 3, 2, 3), (1, 3, 2, 4), (1, 4, 1, 3), (1, 4, 1, 4), (1, 4, 2, 3), (1, 4, 2, 4), (2, 3, 1, 3), (2, 3, 1, 4), (2, 3, 2, 3), (2, 3, 2, 4), (2, 4, 1, 3), (2, 4, 1, 4), (2, 4, 2, 3), (2, 4, 2, 4)] </pre></div> </div> <tt class="docutils literal">combinations(iterable, r)</tt> returns sub-sequences of length r from the elements of iterable. <div class="highlight-python3"><div class="highlight"><pre>>>> list(itertools.combinations('123', 2)) [('1', '2'), ('1', '3'), ('2', '3')] >>> list(itertools.combinations('123', 3)) [('1', '2', '3')] >>> list(itertools.combinations('1234', 3)) [('1', '2', '3'), ('1', '2', '4'), ('1', '3', '4'), ('2', '3', '4')] </pre></div> </div> <tt class="docutils literal">permutations(iter[, r])</tt> returns all the permutations of length r of the iterable’s elements. If r is not specified, it will default to the number of elements produced by the iterable. <div class="highlight-python3"><div class="highlight"><pre>>>> list(itertools.permutations([1,2,3,4], 2)) [(1, 2), (1, 3), (1, 4), (2, 1), (2, 3), (2, 4), (3, 1), (3, 2), (3, 4), (4, 1), (4, 2), (4, 3)] </pre></div> </div> <tt class="docutils literal">itertools.chain(*iterables)</tt> is an existing function in <a title="Functions creating iterators for efficient looping." class="reference external" href="../library/itertools.html#module-itertools"><tt class="xref docutils literal">itertools</tt></a> that gained a new constructor in Python 2.6. <tt class="docutils literal">itertools.chain.from_iterable(iterable)</tt> takes a single iterable that should return other iterables. <tt class="xref docutils literal">chain()</tt> will then return all the elements of the first iterable, then all the elements of the second, and so on. <div class="highlight-python3"><div class="highlight"><pre>>>> list(itertools.chain.from_iterable([[1,2,3], [4,5,6]])) [1, 2, 3, 4, 5, 6] </pre></div> </div> (All contributed by Raymond Hettinger.) </li> <li>The <a title="Flexible error logging system for applications." class="reference external" href="../library/logging.html#module-logging"><tt class="xref docutils literal">logging</tt></a> module’s <tt class="xref docutils literal">FileHandler</tt> class and its subclasses <tt class="xref docutils literal">WatchedFileHandler</tt>, <tt class="xref docutils literal">RotatingFileHandler</tt>, and <tt class="xref docutils literal">TimedRotatingFileHandler</tt> now have an optional delay parameter to their constructors. If delay is true, opening of the log file is deferred until the first <tt class="xref docutils literal">emit()</tt> call is made. (Contributed by Vinay Sajip.) <tt class="xref docutils literal">TimedRotatingFileHandler</tt> also has a utc constructor parameter. If the argument is true, UTC time will be used in determining when midnight occurs and in generating filenames; otherwise local time will be used. </li> <li>Several new functions were added to the <a title="Mathematical functions (sin() etc.)." class="reference external" href="../library/math.html#module-math"><tt class="xref docutils literal">math</tt></a> module: <ul class="simple"> <li><a title="math.isinf" class="reference external" href="../library/math.html#math.isinf"><tt class="xref docutils literal">isinf()</tt></a> and <a title="math.isnan" class="reference external" href="../library/math.html#math.isnan"><tt class="xref docutils literal">isnan()</tt></a> determine whether a given float is a (positive or negative) infinity or a NaN (Not a Number), respectively.</li> <li><a title="math.copysign" class="reference external" href="../library/math.html#math.copysign"><tt class="xref docutils literal">copysign()</tt></a> copies the sign bit of an IEEE 754 number, returning the absolute value of x combined with the sign bit of y. For example, <tt class="docutils literal">math.copysign(1, -0.0)</tt> returns -1.0. (Contributed by Christian Heimes.)</li> <li><a title="math.factorial" class="reference external" href="../library/math.html#math.factorial"><tt class="xref docutils literal">factorial()</tt></a> computes the factorial of a number. (Contributed by Raymond Hettinger; <a class="reference external" href="http://bugs.python.org/issue2138">issue 2138</a>.)</li> <li><a title="math.fsum" class="reference external" href="../library/math.html#math.fsum"><tt class="xref docutils literal">fsum()</tt></a> adds up the stream of numbers from an iterable, and is careful to avoid loss of precision through using partial sums. (Contributed by Jean Brouwers, Raymond Hettinger, and Mark Dickinson; <a class="reference external" href="http://bugs.python.org/issue2819">issue 2819</a>.)</li> <li><a title="math.acosh" class="reference external" href="../library/math.html#math.acosh"><tt class="xref docutils literal">acosh()</tt></a>, <a title="math.asinh" class="reference external" href="../library/math.html#math.asinh"><tt class="xref docutils literal">asinh()</tt></a> and <a title="math.atanh" class="reference external" href="../library/math.html#math.atanh"><tt class="xref docutils literal">atanh()</tt></a> compute the inverse hyperbolic functions.</li> <li><a title="math.log1p" class="reference external" href="../library/math.html#math.log1p"><tt class="xref docutils literal">log1p()</tt></a> returns the natural logarithm of 1+x (base e).</li> <li><tt class="xref docutils literal">trunc()</tt> rounds a number toward zero, returning the closest <tt class="xref docutils literal">Integral</tt> that’s between the function’s argument and zero. Added as part of the backport of <a class="reference external" href="#pep-3141">PEP 3141’s type hierarchy for numbers</a>.</li> </ul> </li> <li>The <a title="Mathematical functions (sin() etc.)." class="reference external" href="../library/math.html#module-math"><tt class="xref docutils literal">math</tt></a> module has been improved to give more consistent behaviour across platforms, especially with respect to handling of floating-point exceptions and IEEE 754 special values. Whenever possible, the module follows the recommendations of the C99 standard about 754’s special values. For example, <tt class="docutils literal">sqrt(-1.)</tt> should now give a <a title="ValueError" class="reference external" href="../library/exceptions.html#ValueError"><tt class="xref docutils literal">ValueError</tt></a> across almost all platforms, while <tt class="docutils literal">sqrt(float('NaN'))</tt> should return a NaN on all IEEE 754 platforms. Where Annex ‘F’ of the C99 standard recommends signaling ‘divide-by-zero’ or ‘invalid’, Python will raise <a title="ValueError" class="reference external" href="../library/exceptions.html#ValueError"><tt class="xref docutils literal">ValueError</tt></a>. Where Annex ‘F’ of the C99 standard recommends signaling ‘overflow’, Python will raise <a title="OverflowError" class="reference external" href="../library/exceptions.html#OverflowError"><tt class="xref docutils literal">OverflowError</tt></a>. (See <a class="reference external" href="http://bugs.python.org/issue711019">issue 711019</a> and <a class="reference external" href="http://bugs.python.org/issue1640">issue 1640</a>.) (Contributed by Christian Heimes and Mark Dickinson.) </li> <li><tt class="xref docutils literal">mmap</tt> objects now have a <tt class="xref docutils literal">rfind()</tt> method that searches for a substring beginning at the end of the string and searching backwards. The <tt class="xref docutils literal">find()</tt> method also gained an end parameter giving an index at which to stop searching. (Contributed by John Lenton.) </li> <li>The <a title="Functions corresponding to the standard operators." class="reference external" href="../library/operator.html#module-operator"><tt class="xref docutils literal">operator</tt></a> module gained a <tt class="xref docutils literal">methodcaller()</tt> function that takes a name and an optional set of arguments, returning a callable that will call the named function on any arguments passed to it. For example: <div class="highlight-python3"><div class="highlight"><pre>>>> # Equivalent to lambda s: s.replace('old', 'new') >>> replacer = operator.methodcaller('replace', 'old', 'new') >>> replacer('old wine in old bottles') 'new wine in new bottles' </pre></div> </div> (Contributed by Georg Brandl, after a suggestion by Gregory Petrosyan.) The <tt class="xref docutils literal">attrgetter()</tt> function now accepts dotted names and performs the corresponding attribute lookups: <div class="highlight-python3"><div class="highlight"><pre>>>> inst_name = operator.attrgetter( ... '__class__.__name__') >>> inst_name('') 'str' >>> inst_name(help) '_Helper' </pre></div> </div> (Contributed by Georg Brandl, after a suggestion by Barry Warsaw.) </li> <li>The <a title="Miscellaneous operating system interfaces." class="reference external" href="../library/os.html#module-os"><tt class="xref docutils literal">os</tt></a> module now wraps several new system calls. <tt class="docutils literal">fchmod(fd, mode)</tt> and <tt class="docutils literal">fchown(fd, uid, gid)</tt> change the mode and ownership of an opened file, and <tt class="docutils literal">lchmod(path, mode)</tt> changes the mode of a symlink. (Contributed by Georg Brandl and Christian Heimes.) <tt class="xref docutils literal">chflags()</tt> and <tt class="xref docutils literal">lchflags()</tt> are wrappers for the corresponding system calls (where they’re available), changing the flags set on a file. Constants for the flag values are defined in the <a title="Utilities for interpreting the results of os.stat(), os.lstat() and os.fstat()." class="reference external" href="../library/stat.html#module-stat"><tt class="xref docutils literal">stat</tt></a> module; some possible values include <tt class="xref docutils literal">UF_IMMUTABLE</tt> to signal the file may not be changed and <tt class="xref docutils literal">UF_APPEND</tt> to indicate that data can only be appended to the file. (Contributed by M. Levinson.) <tt class="docutils literal">os.closerange(low, high)</tt> efficiently closes all file descriptors from low to high, ignoring any errors and not including high itself. This function is now used by the <a title="Subprocess management." class="reference external" href="../library/subprocess.html#module-subprocess"><tt class="xref docutils literal">subprocess</tt></a> module to make starting processes faster. (Contributed by Georg Brandl; <a class="reference external" href="http://bugs.python.org/issue1663329">issue 1663329</a>.) </li> <li>The <tt class="docutils literal">os.environ</tt> object’s <tt class="xref docutils literal">clear()</tt> method will now unset the environment variables using <a title="os.unsetenv" class="reference external" href="../library/os.html#os.unsetenv"><tt class="xref docutils literal">os.unsetenv()</tt></a> in addition to clearing the object’s keys. (Contributed by Martin Horcicka; <a class="reference external" href="http://bugs.python.org/issue1181">issue 1181</a>.) </li> <li>The <a title="os.walk" class="reference external" href="../library/os.html#os.walk"><tt class="xref docutils literal">os.walk()</tt></a> function now has a <tt class="docutils literal">followlinks</tt> parameter. If set to True, it will follow symlinks pointing to directories and visit the directory’s contents. For backward compatibility, the parameter’s default value is false. Note that the function can fall into an infinite recursion if there’s a symlink that points to a parent directory. (<a class="reference external" href="http://bugs.python.org/issue1273829">issue 1273829</a>) </li> <li>In the <a title="Operations on pathnames." class="reference external" href="../library/os.path.html#module-os.path"><tt class="xref docutils literal">os.path</tt></a> module, the <tt class="xref docutils literal">splitext()</tt> function has been changed to not split on leading period characters. This produces better results when operating on Unix’s dot-files. For example, <tt class="docutils literal">os.path.splitext('.ipython')</tt> now returns <tt class="docutils literal">('.ipython', '')</tt> instead of <tt class="docutils literal">('', '.ipython')</tt>. (<a class="reference external" href="http://bugs.python.org/issue1115886">issue 1115886</a>) A new function, <tt class="docutils literal">os.path.relpath(path, start='.')</tt>, returns a relative path from the <tt class="docutils literal">start</tt> path, if it’s supplied, or from the current working directory to the destination <tt class="docutils literal">path</tt>. (Contributed by Richard Barran; <a class="reference external" href="http://bugs.python.org/issue1339796">issue 1339796</a>.) On Windows, <a title="os.path.expandvars" class="reference external" href="../library/os.path.html#os.path.expandvars"><tt class="xref docutils literal">os.path.expandvars()</tt></a> will now expand environment variables given in the form “%var%”, and “~user” will be expanded into the user’s home directory path. (Contributed by Josiah Carlson; <a class="reference external" href="http://bugs.python.org/issue957650">issue 957650</a>.) </li> <li>The Python debugger provided by the <a title="The Python debugger for interactive interpreters." class="reference external" href="../library/pdb.html#module-pdb"><tt class="xref docutils literal">pdb</tt></a> module gained a new command: “run” restarts the Python program being debugged and can optionally take new command-line arguments for the program. (Contributed by Rocky Bernstein; <a class="reference external" href="http://bugs.python.org/issue1393667">issue 1393667</a>.) </li> <li>The <a title="pdb.post_mortem" class="reference external" href="../library/pdb.html#pdb.post_mortem"><tt class="xref docutils literal">pdb.post_mortem()</tt></a> function, used to begin debugging a traceback, will now use the traceback returned by <a title="sys.exc_info" class="reference external" href="../library/sys.html#sys.exc_info"><tt class="xref docutils literal">sys.exc_info()</tt></a> if no traceback is supplied. (Contributed by Facundo Batista; <a class="reference external" href="http://bugs.python.org/issue1106316">issue 1106316</a>.) </li> <li>The <a title="Contains extensive comments about the pickle protocols and pickle-machine opcodes, as well as some useful functions." class="reference external" href="../library/pickletools.html#module-pickletools"><tt class="xref docutils literal">pickletools</tt></a> module now has an <tt class="xref docutils literal">optimize()</tt> function that takes a string containing a pickle and removes some unused opcodes, returning a shorter pickle that contains the same data structure. (Contributed by Raymond Hettinger.) </li> <li>A <tt class="xref docutils literal">get_data()</tt> function was added to the <a title="Utilities for the import system." class="reference external" href="../library/pkgutil.html#module-pkgutil"><tt class="xref docutils literal">pkgutil</tt></a> module that returns the contents of resource files included with an installed Python package. For example: <div class="highlight-python3"><div class="highlight"><pre>>>> import pkgutil >>> print pkgutil.get_data('test', 'exception_hierarchy.txt') BaseException +-- SystemExit +-- KeyboardInterrupt +-- GeneratorExit +-- Exception +-- StopIteration +-- StandardError ... </pre></div> </div> (Contributed by Paul Moore; <a class="reference external" href="http://bugs.python.org/issue2439">issue 2439</a>.) </li> <li>The <tt class="xref docutils literal">pyexpat</tt> module’s <tt class="xref docutils literal">Parser</tt> objects now allow setting their <tt class="xref docutils literal">buffer_size</tt> attribute to change the size of the buffer used to hold character data. (Contributed by Achim Gaedke; <a class="reference external" href="http://bugs.python.org/issue1137">issue 1137</a>.) </li> <li>The <tt class="xref docutils literal">Queue</tt> module now provides queue variants that retrieve entries in different orders. The <tt class="xref docutils literal">PriorityQueue</tt> class stores queued items in a heap and retrieves them in priority order, and <tt class="xref docutils literal">LifoQueue</tt> retrieves the most recently added entries first, meaning that it behaves like a stack. (Contributed by Raymond Hettinger.) </li> <li>The <a title="Generate pseudo-random numbers with various common distributions." class="reference external" href="../library/random.html#module-random"><tt class="xref docutils literal">random</tt></a> module’s <tt class="xref docutils literal">Random</tt> objects can now be pickled on a 32-bit system and unpickled on a 64-bit system, and vice versa. Unfortunately, this change also means that Python 2.6’s <tt class="xref docutils literal">Random</tt> objects can’t be unpickled correctly on earlier versions of Python. (Contributed by Shawn Ligocki; <a class="reference external" href="http://bugs.python.org/issue1727780">issue 1727780</a>.) The new <tt class="docutils literal">triangular(low, high, mode)</tt> function returns random numbers following a triangular distribution. The returned values are between low and high, not including high itself, and with mode as the most frequently occurring value in the distribution. (Contributed by Wladmir van der Laan and Raymond Hettinger; <a class="reference external" href="http://bugs.python.org/issue1681432">issue 1681432</a>.) </li> <li>Long regular expression searches carried out by the <a title="Regular expression operations." class="reference external" href="../library/re.html#module-re"><tt class="xref docutils literal">re</tt></a> module will check for signals being delivered, so time-consuming searches can now be interrupted. (Contributed by Josh Hoyt and Ralf Schmitt; <a class="reference external" href="http://bugs.python.org/issue846388">issue 846388</a>.) The regular expression module is implemented by compiling bytecodes for a tiny regex-specific virtual machine. Untrusted code could create malicious strings of bytecode directly and cause crashes, so Python 2.6 includes a verifier for the regex bytecode. (Contributed by Guido van Rossum from work for Google App Engine; <a class="reference external" href="http://bugs.python.org/issue3487">issue 3487</a>.) </li> <li>The <a title="Python identifier completion, suitable for the GNU readline library." class="reference external" href="../library/rlcompleter.html#module-rlcompleter"><tt class="xref docutils literal">rlcompleter</tt></a> module’s <tt class="xref docutils literal">Completer.complete()</tt> method will now ignore exceptions triggered while evaluating a name. (Fixed by Lorenz Quack; <a class="reference external" href="http://bugs.python.org/issue2250">issue 2250</a>.) </li> <li>The <a title="General purpose event scheduler." class="reference external" href="../library/sched.html#module-sched"><tt class="xref docutils literal">sched</tt></a> module’s <tt class="xref docutils literal">scheduler</tt> instances now have a read-only <tt class="xref docutils literal">queue</tt> attribute that returns the contents of the scheduler’s queue, represented as a list of named tuples with the fields <tt class="docutils literal">(time, priority, action, argument)</tt>. (Contributed by Raymond Hettinger; <a class="reference external" href="http://bugs.python.org/issue1861">issue 1861</a>.) </li> <li>The <a title="Wait for I/O completion on multiple streams." class="reference external" href="../library/select.html#module-select"><tt class="xref docutils literal">select</tt></a> module now has wrapper functions for the Linux <tt class="xref docutils literal">epoll()</tt> and BSD <tt class="xref docutils literal">kqueue()</tt> system calls. <tt class="xref docutils literal">modify()</tt> method was added to the existing <tt class="xref docutils literal">poll</tt> objects; <tt class="docutils literal">pollobj.modify(fd, eventmask)</tt> takes a file descriptor or file object and an event mask, modifying the recorded event mask for that file. (Contributed by Christian Heimes; <a class="reference external" href="http://bugs.python.org/issue1657">issue 1657</a>.) </li> <li>The <a title="shutil.copytree" class="reference external" href="../library/shutil.html#shutil.copytree"><tt class="xref docutils literal">shutil.copytree()</tt></a> function now has an optional ignore argument that takes a callable object. This callable will receive each directory path and a list of the directory’s contents, and returns a list of names that will be ignored, not copied. The <a title="High-level file operations, including copying." class="reference external" href="../library/shutil.html#module-shutil"><tt class="xref docutils literal">shutil</tt></a> module also provides an <tt class="xref docutils literal">ignore_patterns()</tt> function for use with this new parameter. <tt class="xref docutils literal">ignore_patterns()</tt> takes an arbitrary number of glob-style patterns and returns a callable that will ignore any files and directories that match any of these patterns. The following example copies a directory tree, but skips both <tt class="docutils literal">.svn</tt> directories and Emacs backup files, which have names ending with ‘~’: <div class="highlight-python3"><div class="highlight"><pre>shutil.copytree('Doc/library', '/tmp/library', ignore=shutil.ignore_patterns('*~', '.svn')) </pre></div> </div> (Contributed by Tarek Ziad茅; <a class="reference external" href="http://bugs.python.org/issue2663">issue 2663</a>.) </li> <li>Integrating signal handling with GUI handling event loops like those used by Tkinter or GTk+ has long been a problem; most software ends up polling, waking up every fraction of a second to check if any GUI events have occurred. The <a title="Set handlers for asynchronous events." class="reference external" href="../library/signal.html#module-signal"><tt class="xref docutils literal">signal</tt></a> module can now make this more efficient. Calling <tt class="docutils literal">signal.set_wakeup_fd(fd)</tt> sets a file descriptor to be used; when a signal is received, a byte is written to that file descriptor. There’s also a C-level function, <a title="PySignal_SetWakeupFd" class="reference external" href="../c-api/exceptions.html#PySignal_SetWakeupFd"><tt class="xref docutils literal">PySignal_SetWakeupFd()</tt></a>, for setting the descriptor. Event loops will use this by opening a pipe to create two descriptors, one for reading and one for writing. The writable descriptor will be passed to <tt class="xref docutils literal">set_wakeup_fd()</tt>, and the readable descriptor will be added to the list of descriptors monitored by the event loop via <tt class="xref docutils literal">select()</tt> or <tt class="xref docutils literal">poll()</tt>. On receiving a signal, a byte will be written and the main event loop will be woken up, avoiding the need to poll. (Contributed by Adam Olsen; <a class="reference external" href="http://bugs.python.org/issue1583">issue 1583</a>.) The <tt class="xref docutils literal">siginterrupt()</tt> function is now available from Python code, and allows changing whether signals can interrupt system calls or not. (Contributed by Ralf Schmitt.) The <tt class="xref docutils literal">setitimer()</tt> and <tt class="xref docutils literal">getitimer()</tt> functions have also been added (where they’re available). <tt class="xref docutils literal">setitimer()</tt> allows setting interval timers that will cause a signal to be delivered to the process after a specified time, measured in wall-clock time, consumed process time, or combined process+system time. (Contributed by Guilherme Polo; <a class="reference external" href="http://bugs.python.org/issue2240">issue 2240</a>.) </li> <li>The <a title="SMTP protocol client (requires sockets)." class="reference external" href="../library/smtplib.html#module-smtplib"><tt class="xref docutils literal">smtplib</tt></a> module now supports SMTP over SSL thanks to the addition of the <tt class="xref docutils literal">SMTP_SSL</tt> class. This class supports an interface identical to the existing <tt class="xref docutils literal">SMTP</tt> class. (Contributed by Monty Taylor.) Both class constructors also have an optional <tt class="docutils literal">timeout</tt> parameter that specifies a timeout for the initial connection attempt, measured in seconds. (Contributed by Facundo Batista.) An implementation of the LMTP protocol (<a class="reference external" href="http://tools.ietf.org/html/rfc2033.html">RFC 2033</a>) was also added to the module. LMTP is used in place of SMTP when transferring e-mail between agents that don’t manage a mail queue. (LMTP implemented by Leif Hedstrom; <a class="reference external" href="http://bugs.python.org/issue957003">issue 957003</a>.) <tt class="xref docutils literal">SMTP.starttls()</tt> now complies with <a class="reference external" href="http://tools.ietf.org/html/rfc3207.html">RFC 3207</a> and forgets any knowledge obtained from the server not obtained from the TLS negotiation itself. (Patch contributed by Bill Fenner; <a class="reference external" href="http://bugs.python.org/issue829951">issue 829951</a>.) </li> <li>The <a title="Low-level networking interface." class="reference external" href="../library/socket.html#module-socket"><tt class="xref docutils literal">socket</tt></a> module now supports TIPC (<a class="reference external" href="http://tipc.sf.net">http://tipc.sf.net</a>), a high-performance non-IP-based protocol designed for use in clustered environments. TIPC addresses are 4- or 5-tuples. (Contributed by Alberto Bertogli; <a class="reference external" href="http://bugs.python.org/issue1646">issue 1646</a>.) A new function, <tt class="xref docutils literal">create_connection()</tt>, takes an address and connects to it using an optional timeout value, returning the connected socket object. This function also looks up the address’s type and connects to it using IPv4 or IPv6 as appropriate. Changing your code to use <tt class="xref docutils literal">create_connection()</tt> instead of <tt class="docutils literal">socket(socket.AF_INET, ...)</tt> may be all that’s required to make your code work with IPv6. </li> <li>The base classes in the <tt class="xref docutils literal">SocketServer</tt> module now support calling a <tt class="xref docutils literal">handle_timeout()</tt> method after a span of inactivity specified by the server’s <tt class="xref docutils literal">timeout</tt> attribute. (Contributed by Michael Pomraning.) The <tt class="xref docutils literal">serve_forever()</tt> method now takes an optional poll interval measured in seconds, controlling how often the server will check for a shutdown request. (Contributed by Pedro Werneck and Jeffrey Yasskin; <a class="reference external" href="http://bugs.python.org/issue742598">issue 742598</a>, <a class="reference external" href="http://bugs.python.org/issue1193577">issue 1193577</a>.) </li> <li>The <a title="A DB-API 2.0 implementation using SQLite 3.x." class="reference external" href="../library/sqlite3.html#module-sqlite3"><tt class="xref docutils literal">sqlite3</tt></a> module, maintained by Gerhard Haering, has been updated from version 2.3.2 in Python 2.5 to version 2.4.1. </li> <li>The <a title="Interpret bytes as packed binary data." class="reference external" href="../library/struct.html#module-struct"><tt class="xref docutils literal">struct</tt></a> module now supports the C99 <tt class="xref docutils literal">_Bool</tt> type, using the format character <tt class="docutils literal">'?'</tt>. (Contributed by David Remahl.) </li> <li>The <tt class="xref docutils literal">Popen</tt> objects provided by the <a title="Subprocess management." class="reference external" href="../library/subprocess.html#module-subprocess"><tt class="xref docutils literal">subprocess</tt></a> module now have <tt class="xref docutils literal">terminate()</tt>, <tt class="xref docutils literal">kill()</tt>, and <tt class="xref docutils literal">send_signal()</tt> methods. On Windows, <tt class="xref docutils literal">send_signal()</tt> only supports the <tt class="xref docutils literal">SIGTERM</tt> signal, and all these methods are aliases for the Win32 API function <tt class="xref docutils literal">TerminateProcess()</tt>. (Contributed by Christian Heimes.) </li> <li>A new variable in the <a title="Access system-specific parameters and functions." class="reference external" href="../library/sys.html#module-sys"><tt class="xref docutils literal">sys</tt></a> module, <tt class="xref docutils literal">float_info</tt>, is an object containing information derived from the <tt class="docutils literal">float.h</tt> file about the platform’s floating-point support. Attributes of this object include <tt class="xref docutils literal">mant_dig</tt> (number of digits in the mantissa), <tt class="xref docutils literal">epsilon</tt> (smallest difference between 1.0 and the next largest value representable), and several others. (Contributed by Christian Heimes; <a class="reference external" href="http://bugs.python.org/issue1534">issue 1534</a>.) Another new variable, <tt class="xref docutils literal">dont_write_bytecode</tt>, controls whether Python writes any <tt class="docutils literal">.pyc</tt> or <tt class="docutils literal">.pyo</tt> files on importing a module. If this variable is true, the compiled files are not written. The variable is initially set on start-up by supplying the <a class="reference external" href="../using/cmdline.html#cmdoption-trace-B">-B</a> switch to the Python interpreter, or by setting the <a class="reference external" href="../using/cmdline.html#envvar-PYTHONDONTWRITEBYTECODE">PYTHONDONTWRITEBYTECODE</a> environment variable before running the interpreter. Python code can subsequently change the value of this variable to control whether bytecode files are written or not. (Contributed by Neal Norwitz and Georg Brandl.) Information about the command-line arguments supplied to the Python interpreter is available by reading attributes of a named tuple available as <tt class="docutils literal">sys.flags</tt>. For example, the <tt class="xref docutils literal">verbose</tt> attribute is true if Python was executed in verbose mode, <tt class="xref docutils literal">debug</tt> is true in debugging mode, etc. These attributes are all read-only. (Contributed by Christian Heimes.) A new function, <tt class="xref docutils literal">getsizeof()</tt>, takes a Python object and returns the amount of memory used by the object, measured in bytes. Built-in objects return correct results; third-party extensions may not, but can define a <tt class="xref docutils literal">__sizeof__()</tt> method to return the object’s size. (Contributed by Robert Schuppenies; <a class="reference external" href="http://bugs.python.org/issue2898">issue 2898</a>.) It’s now possible to determine the current profiler and tracer functions by calling <a title="sys.getprofile" class="reference external" href="../library/sys.html#sys.getprofile"><tt class="xref docutils literal">sys.getprofile()</tt></a> and <a title="sys.gettrace" class="reference external" href="../library/sys.html#sys.gettrace"><tt class="xref docutils literal">sys.gettrace()</tt></a>. (Contributed by Georg Brandl; <a class="reference external" href="http://bugs.python.org/issue1648">issue 1648</a>.) </li> <li>The <a title="Read and write tar-format archive files." class="reference external" href="../library/tarfile.html#module-tarfile"><tt class="xref docutils literal">tarfile</tt></a> module now supports POSIX.1-2001 (pax) tarfiles in addition to the POSIX.1-1988 (ustar) and GNU tar formats that were already supported. The default format is GNU tar; specify the <tt class="docutils literal">format</tt> parameter to open a file using a different format: <div class="highlight-python3"><div class="highlight"><pre>tar = tarfile.open("output.tar", "w", format=tarfile.PAX_FORMAT) </pre></div> </div> The new <tt class="docutils literal">encoding</tt> and <tt class="docutils literal">errors</tt> parameters specify an encoding and an error handling scheme for character conversions. <tt class="docutils literal">'strict'</tt>, <tt class="docutils literal">'ignore'</tt>, and <tt class="docutils literal">'replace'</tt> are the three standard ways Python can handle errors,; <tt class="docutils literal">'utf-8'</tt> is a special value that replaces bad characters with their UTF-8 representation. (Character conversions occur because the PAX format supports Unicode filenames, defaulting to UTF-8 encoding.) The <tt class="xref docutils literal">TarFile.add()</tt> method now accepts an <tt class="docutils literal">exclude</tt> argument that’s a function that can be used to exclude certain filenames from an archive. The function must take a filename and return true if the file should be excluded or false if it should be archived. The function is applied to both the name initially passed to <tt class="xref docutils literal">add()</tt> and to the names of files in recursively-added directories. (All changes contributed by Lars Gust盲bel). </li> <li>An optional <tt class="docutils literal">timeout</tt> parameter was added to the <a title="telnetlib.Telnet" class="reference external" href="../library/telnetlib.html#telnetlib.Telnet"><tt class="xref docutils literal">telnetlib.Telnet</tt></a> class constructor, specifying a timeout measured in seconds. (Added by Facundo Batista.) </li> <li>The <a title="tempfile.NamedTemporaryFile" class="reference external" href="../library/tempfile.html#tempfile.NamedTemporaryFile"><tt class="xref docutils literal">tempfile.NamedTemporaryFile</tt></a> class usually deletes the temporary file it created when the file is closed. This behaviour can now be changed by passing <tt class="docutils literal">delete=False</tt> to the constructor. (Contributed by Damien Miller; <a class="reference external" href="http://bugs.python.org/issue1537850">issue 1537850</a>.) A new class, <tt class="xref docutils literal">SpooledTemporaryFile</tt>, behaves like a temporary file but stores its data in memory until a maximum size is exceeded. On reaching that limit, the contents will be written to an on-disk temporary file. (Contributed by Dustin J. Mitchell.) The <tt class="xref docutils literal">NamedTemporaryFile</tt> and <tt class="xref docutils literal">SpooledTemporaryFile</tt> classes both work as context managers, so you can write <tt class="docutils literal">with tempfile.NamedTemporaryFile() as tmp: ...</tt>. (Contributed by Alexander Belopolsky; <a class="reference external" href="http://bugs.python.org/issue2021">issue 2021</a>.) </li> <li>The <tt class="xref docutils literal">test.test_support</tt> module gained a number of context managers useful for writing tests. <tt class="xref docutils literal">EnvironmentVarGuard()</tt> is a context manager that temporarily changes environment variables and automatically restores them to their old values. Another context manager, <tt class="xref docutils literal">TransientResource</tt>, can surround calls to resources that may or may not be available; it will catch and ignore a specified list of exceptions. For example, a network test may ignore certain failures when connecting to an external web site: <div class="highlight-python3"><div class="highlight"><pre>with test_support.TransientResource(IOError, errno=errno.ETIMEDOUT): f = urllib.urlopen('https://sf.net') ... </pre></div> </div> Finally, <tt class="xref docutils literal">check_warnings()</tt> resets the <tt class="xref docutils literal">warning</tt> module’s warning filters and returns an object that will record all warning messages triggered (<a class="reference external" href="http://bugs.python.org/issue3781">issue 3781</a>): <div class="highlight-python3"><div class="highlight"><pre>with test_support.check_warnings() as wrec: warnings.simplefilter("always") # ... code that triggers a warning ... assert str(wrec.message) == "function is outdated" assert len(wrec.warnings) == 1, "Multiple warnings raised" </pre></div> </div> (Contributed by Brett Cannon.) </li> <li>The <a title="Text wrapping and filling" class="reference external" href="../library/textwrap.html#module-textwrap"><tt class="xref docutils literal">textwrap</tt></a> module can now preserve existing whitespace at the beginnings and ends of the newly-created lines by specifying <tt class="docutils literal">drop_whitespace=False</tt> as an argument: <div class="highlight-python3"><div class="highlight"><pre>>>> S = """This sentence has a bunch of ... extra whitespace.""" >>> print textwrap.fill(S, width=15) This sentence has a bunch of extra whitespace. >>> print textwrap.fill(S, drop_whitespace=False, width=15) This sentence has a bunch of extra whitespace. >>> </pre></div> </div> (Contributed by Dwayne Bailey; <a class="reference external" href="http://bugs.python.org/issue1581073">issue 1581073</a>.) </li> <li>The <a title="Thread-based parallelism." class="reference external" href="../library/threading.html#module-threading"><tt class="xref docutils literal">threading</tt></a> module API is being changed to use properties such as <tt class="xref docutils literal">daemon</tt> instead of <tt class="xref docutils literal">setDaemon()</tt> and <tt class="xref docutils literal">isDaemon()</tt> methods, and some methods have been renamed to use underscores instead of camel-case; for example, the <tt class="xref docutils literal">activeCount()</tt> method is renamed to <tt class="xref docutils literal">active_count()</tt>. Both the 2.6 and 3.0 versions of the module support the same properties and renamed methods, but don’t remove the old methods. No date has been set for the deprecation of the old APIs in Python 3.x; the old APIs won’t be removed in any 2.x version. (Carried out by several people, most notably Benjamin Peterson.) The <a title="Thread-based parallelism." class="reference external" href="../library/threading.html#module-threading"><tt class="xref docutils literal">threading</tt></a> module’s <tt class="xref docutils literal">Thread</tt> objects gained an <tt class="xref docutils literal">ident</tt> property that returns the thread’s identifier, a nonzero integer. (Contributed by Gregory P. Smith; <a class="reference external" href="http://bugs.python.org/issue2871">issue 2871</a>.) </li> <li>The <a title="Measure the execution time of small code snippets." class="reference external" href="../library/timeit.html#module-timeit"><tt class="xref docutils literal">timeit</tt></a> module now accepts callables as well as strings for the statement being timed and for the setup code. Two convenience functions were added for creating <tt class="xref docutils literal">Timer</tt> instances: <tt class="docutils literal">repeat(stmt, setup, time, repeat, number)</tt> and <tt class="docutils literal">timeit(stmt, setup, time, number)</tt> create an instance and call the corresponding method. (Contributed by Erik Demaine; <a class="reference external" href="http://bugs.python.org/issue1533909">issue 1533909</a>.) </li> <li>The <tt class="xref docutils literal">Tkinter</tt> module now accepts lists and tuples for options, separating the elements by spaces before passing the resulting value to Tcl/Tk. (Contributed by Guilherme Polo; <a class="reference external" href="http://bugs.python.org/issue2906">issue 2906</a>.) </li> <li>The <a title="An educational framework for simple graphics applications" class="reference external" href="../library/turtle.html#module-turtle"><tt class="xref docutils literal">turtle</tt></a> module for turtle graphics was greatly enhanced by Gregor Lingl. New features in the module include: <ul class="simple"> <li>Better animation of turtle movement and rotation.</li> <li>Control over turtle movement using the new <tt class="xref docutils literal">delay()</tt>, <tt class="xref docutils literal">tracer()</tt>, and <tt class="xref docutils literal">speed()</tt> methods.</li> <li>The ability to set new shapes for the turtle, and to define a new coordinate system.</li> <li>Turtles now have an <tt class="xref docutils literal">undo()</tt> method that can roll back actions.</li> <li>Simple support for reacting to input events such as mouse and keyboard activity, making it possible to write simple games.</li> <li>A <tt class="docutils literal">turtle.cfg</tt> file can be used to customize the starting appearance of the turtle’s screen.</li> <li>The module’s docstrings can be replaced by new docstrings that have been translated into another language.</li> </ul> (<a class="reference external" href="http://bugs.python.org/issue1513695">issue 1513695</a>) </li> <li>An optional <tt class="docutils literal">timeout</tt> parameter was added to the <tt class="xref docutils literal">urllib.urlopen()</tt> function and the <tt class="xref docutils literal">urllib.ftpwrapper</tt> class constructor, as well as the <tt class="xref docutils literal">urllib2.urlopen()</tt> function. The parameter specifies a timeout measured in seconds. For example: <div class="highlight-python3"><div class="highlight"><pre>>>> u = urllib2.urlopen("http://slow.example.com", timeout=3) Traceback (most recent call last): ... urllib2.URLError: <urlopen error timed out> >>> </pre></div> </div> (Added by Facundo Batista.) </li> <li>The Unicode database provided by the <a title="Access the Unicode Database." class="reference external" href="../library/unicodedata.html#module-unicodedata"><tt class="xref docutils literal">unicodedata</tt></a> module has been updated to version 5.1.0. (Updated by Martin von Loewis; <a class="reference external" href="http://bugs.python.org/issue3811">issue 3811</a>.) </li> <li>The <a title="Issue warning messages and control their disposition." class="reference external" href="../library/warnings.html#module-warnings"><tt class="xref docutils literal">warnings</tt></a> module’s <tt class="xref docutils literal">formatwarning()</tt> and <tt class="xref docutils literal">showwarning()</tt> gained an optional line argument that can be used to supply the line of source code. (Added as part of <a class="reference external" href="http://bugs.python.org/issue1631171">issue 1631171</a>, which re-implemented part of the <a title="Issue warning messages and control their disposition." class="reference external" href="../library/warnings.html#module-warnings"><tt class="xref docutils literal">warnings</tt></a> module in C code.) A new function, <tt class="xref docutils literal">catch_warnings()</tt>, is a context manager intended for testing purposes that lets you temporarily modify the warning filters and then restore their original values (<a class="reference external" href="http://bugs.python.org/issue3781">issue 3781</a>). </li> <li>The XML-RPC <tt class="xref docutils literal">SimpleXMLRPCServer</tt> and <tt class="xref docutils literal">DocXMLRPCServer</tt> classes can now be prevented from immediately opening and binding to their socket by passing True as the <tt class="docutils literal">bind_and_activate</tt> constructor parameter. This can be used to modify the instance’s <tt class="xref docutils literal">allow_reuse_address</tt> attribute before calling the <tt class="xref docutils literal">server_bind()</tt> and <tt class="xref docutils literal">server_activate()</tt> methods to open the socket and begin listening for connections. (Contributed by Peter Parente; <a class="reference external" href="http://bugs.python.org/issue1599845">issue 1599845</a>.) <tt class="xref docutils literal">SimpleXMLRPCServer</tt> also has a <tt class="xref docutils literal">_send_traceback_header</tt> attribute; if true, the exception and formatted traceback are returned as HTTP headers “X-Exception” and “X-Traceback”. This feature is for debugging purposes only and should not be used on production servers because the tracebacks might reveal passwords or other sensitive information. (Contributed by Alan McIntyre as part of his project for Google’s Summer of Code 2007.) </li> <li>The <tt class="xref docutils literal">xmlrpclib</tt> module no longer automatically converts <a title="datetime.date" class="reference external" href="../library/datetime.html#datetime.date"><tt class="xref docutils literal">datetime.date</tt></a> and <a title="datetime.time" class="reference external" href="../library/datetime.html#datetime.time"><tt class="xref docutils literal">datetime.time</tt></a> to the <tt class="xref docutils literal">xmlrpclib.DateTime</tt> type; the conversion semantics were not necessarily correct for all applications. Code using <tt class="xref docutils literal">xmlrpclib</tt> should convert <tt class="xref docutils literal">date</tt> and <tt class="xref docutils literal">time</tt> instances. (<a class="reference external" href="http://bugs.python.org/issue1330538">issue 1330538</a>) The code can also handle dates before 1900 (contributed by Ralf Schmitt; <a class="reference external" href="http://bugs.python.org/issue2014">issue 2014</a>) and 64-bit integers represented by using <tt class="docutils literal"><i8></tt> in XML-RPC responses (contributed by Riku Lindblad; <a class="reference external" href="http://bugs.python.org/issue2985">issue 2985</a>). </li> <li>The <a title="Read and write ZIP-format archive files." class="reference external" href="../library/zipfile.html#module-zipfile"><tt class="xref docutils literal">zipfile</tt></a> module’s <tt class="xref docutils literal">ZipFile</tt> class now has <tt class="xref docutils literal">extract()</tt> and <tt class="xref docutils literal">extractall()</tt> methods that will unpack a single file or all the files in the archive to the current directory, or to a specified directory: <div class="highlight-python3"><div class="highlight"><pre>z = zipfile.ZipFile('python-251.zip') # Unpack a single file, writing it relative # to the /tmp directory. z.extract('Python/sysmodule.c', '/tmp') # Unpack all the files in the archive. z.extractall() </pre></div> </div> (Contributed by Alan McIntyre; <a class="reference external" href="http://bugs.python.org/issue467924">issue 467924</a>.) The <a title="open" class="reference external" href="../library/functions.html#open"><tt class="xref docutils literal">open()</tt></a>, <tt class="xref docutils literal">read()</tt> and <tt class="xref docutils literal">extract()</tt> methods can now take either a filename or a <tt class="xref docutils literal">ZipInfo</tt> object. This is useful when an archive accidentally contains a duplicated filename. (Contributed by Graham Horler; <a class="reference external" href="http://bugs.python.org/issue1775025">issue 1775025</a>.) Finally, <a title="Read and write ZIP-format archive files." class="reference external" href="../library/zipfile.html#module-zipfile"><tt class="xref docutils literal">zipfile</tt></a> now supports using Unicode filenames for archived files. (Contributed by Alexey Borzenkov; <a class="reference external" href="http://bugs.python.org/issue1734346">issue 1734346</a>.) </li> </ul> <div class="section" id="the-ast-module"> <h3>The <a title="Abstract Syntax Tree classes and manipulation." class="reference external" href="../library/ast.html#module-ast"><tt class="xref docutils literal">ast</tt></a> module<a class="headerlink" href="#the-ast-module" title="Permalink to this headline">露</a></h3> The <a title="Abstract Syntax Tree classes and manipulation." class="reference external" href="../library/ast.html#module-ast"><tt class="xref docutils literal">ast</tt></a> module provides an Abstract Syntax Tree representation of Python code, and Armin Ronacher contributed a set of helper functions that perform a variety of common tasks. These will be useful for HTML templating packages, code analyzers, and similar tools that process Python code. The <tt class="xref docutils literal">parse()</tt> function takes an expression and returns an AST. The <tt class="xref docutils literal">dump()</tt> function outputs a representation of a tree, suitable for debugging: <div class="highlight-python3"><div class="highlight"><pre>import ast t = ast.parse(""" d = {} for i in 'abcdefghijklm': d[i + i] = ord(i) - ord('a') + 1 print d """) print ast.dump(t) </pre></div> </div> This outputs a deeply nested tree: <div class="highlight-python3"><div class="highlight"><pre>Module(body=[ Assign(targets=[ Name(id='d', ctx=Store()) ], value=Dict(keys=[], values=[])) For(target=Name(id='i', ctx=Store()), iter=Str(s='abcdefghijklm'), body=[ Assign(targets=[ Subscript(value= Name(id='d', ctx=Load()), slice= Index(value= BinOp(left=Name(id='i', ctx=Load()), op=Add(), right=Name(id='i', ctx=Load()))), ctx=Store()) ], value= BinOp(left= BinOp(left= Call(func= Name(id='ord', ctx=Load()), args=[ Name(id='i', ctx=Load()) ], keywords=[], starargs=None, kwargs=None), op=Sub(), right=Call(func= Name(id='ord', ctx=Load()), args=[ Str(s='a') ], keywords=[], starargs=None, kwargs=None)), op=Add(), right=Num(n=1))) ], orelse=[]) Print(dest=None, values=[ Name(id='d', ctx=Load()) ], nl=True) ]) </pre></div> </div> The <tt class="xref docutils literal">literal_eval()</tt> method takes a string or an AST representing a literal expression, parses and evaluates it, and returns the resulting value. A literal expression is a Python expression containing only strings, numbers, dictionaries, etc. but no statements or function calls. If you need to evaluate an expression but cannot accept the security risk of using an <a title="eval" class="reference external" href="../library/functions.html#eval"><tt class="xref docutils literal">eval()</tt></a> call, <tt class="xref docutils literal">literal_eval()</tt> will handle it safely: <div class="highlight-python3"><div class="highlight"><pre>>>> literal = '("a", "b", {2:4, 3:8, 1:2})' >>> print ast.literal_eval(literal) ('a', 'b', {1: 2, 2: 4, 3: 8}) >>> print ast.literal_eval('"a" + "b"') Traceback (most recent call last): ... ValueError: malformed string </pre></div> </div> The module also includes <tt class="xref docutils literal">NodeVisitor</tt> and <tt class="xref docutils literal">NodeTransformer</tt> classes for traversing and modifying an AST, and functions for common transformations such as changing line numbers. </div> <div class="section" id="the-future-builtins-module"> <h3>The <tt class="xref docutils literal">future_builtins</tt> module<a class="headerlink" href="#the-future-builtins-module" title="Permalink to this headline">露</a></h3> Python 3.0 makes many changes to the repertoire of built-in functions, and most of the changes can’t be introduced in the Python 2.x series because they would break compatibility. The <tt class="xref docutils literal">future_builtins</tt> module provides versions of these built-in functions that can be imported when writing 3.0-compatible code. The functions in this module currently include: <ul class="simple"> <li><tt class="docutils literal">ascii(obj)</tt>: equivalent to <a title="repr" class="reference external" href="../library/functions.html#repr"><tt class="xref docutils literal">repr()</tt></a>. In Python 3.0, <a title="repr" class="reference external" href="../library/functions.html#repr"><tt class="xref docutils literal">repr()</tt></a> will return a Unicode string, while <a title="ascii" class="reference external" href="../library/functions.html#ascii"><tt class="xref docutils literal">ascii()</tt></a> will return a pure ASCII bytestring.</li> <li><tt class="docutils literal">filter(predicate, iterable)</tt>, <tt class="docutils literal">map(func, iterable1, ...)</tt>: the 3.0 versions return iterators, unlike the 2.x builtins which return lists.</li> <li><tt class="docutils literal">hex(value)</tt>, <tt class="docutils literal">oct(value)</tt>: instead of calling the <tt class="xref docutils literal">__hex__()</tt> or <tt class="xref docutils literal">__oct__()</tt> methods, these versions will call the <a title="object.__index__" class="reference external" href="../reference/datamodel.html#object.__index__"><tt class="xref docutils literal">__index__()</tt></a> method and convert the result to hexadecimal or octal. <a title="oct" class="reference external" href="../library/functions.html#oct"><tt class="xref docutils literal">oct()</tt></a> will use the new <tt class="docutils literal">0o</tt> notation for its result.</li> </ul> </div> <div class="section" id="the-json-module-javascript-object-notation"> <h3>The <a title="Encode and decode the JSON format." class="reference external" href="../library/json.html#module-json"><tt class="xref docutils literal">json</tt></a> module: JavaScript Object Notation<a class="headerlink" href="#the-json-module-javascript-object-notation" title="Permalink to this headline">露</a></h3> The new <a title="Encode and decode the JSON format." class="reference external" href="../library/json.html#module-json"><tt class="xref docutils literal">json</tt></a> module supports the encoding and decoding of Python types in JSON (Javascript Object Notation). JSON is a lightweight interchange format often used in web applications. For more information about JSON, see <a class="reference external" href="http://www.json.org">http://www.json.org</a>. <a title="Encode and decode the JSON format." class="reference external" href="../library/json.html#module-json"><tt class="xref docutils literal">json</tt></a> comes with support for decoding and encoding most built-in Python types. The following example encodes and decodes a dictionary: <div class="highlight-python3"><div class="highlight"><pre>>>> import json >>> data = {"spam" : "foo", "parrot" : 42} >>> in_json = json.dumps(data) # Encode the data >>> in_json '{"parrot": 42, "spam": "foo"}' >>> json.loads(in_json) # Decode into a Python object {"spam" : "foo", "parrot" : 42} </pre></div> </div> It’s also possible to write your own decoders and encoders to support more types. Pretty-printing of the JSON strings is also supported. <a title="Encode and decode the JSON format." class="reference external" href="../library/json.html#module-json"><tt class="xref docutils literal">json</tt></a> (originally called simplejson) was written by Bob Ippolito. </div> <div class="section" id="the-plistlib-module-a-property-list-parser"> <h3>The <a title="Generate and parse Mac OS X plist files." class="reference external" href="../library/plistlib.html#module-plistlib"><tt class="xref docutils literal">plistlib</tt></a> module: A Property-List Parser<a class="headerlink" href="#the-plistlib-module-a-property-list-parser" title="Permalink to this headline">露</a></h3> The <tt class="docutils literal">.plist</tt> format is commonly used on Mac OS X to store basic data types (numbers, strings, lists, and dictionaries) by serializing them into an XML-based format. It resembles the XML-RPC serialization of data types. Despite being primarily used on Mac OS X, the format has nothing Mac-specific about it and the Python implementation works on any platform that Python supports, so the <a title="Generate and parse Mac OS X plist files." class="reference external" href="../library/plistlib.html#module-plistlib"><tt class="xref docutils literal">plistlib</tt></a> module has been promoted to the standard library. Using the module is simple: <div class="highlight-python3"><div class="highlight"><pre>import sys import plistlib import datetime # Create data structure data_struct = dict(lastAccessed=datetime.datetime.now(), version=1, categories=('Personal','Shared','Private')) # Create string containing XML. plist_str = plistlib.writePlistToString(data_struct) new_struct = plistlib.readPlistFromString(plist_str) print data_struct print new_struct # Write data structure to a file and read it back. plistlib.writePlist(data_struct, '/tmp/customizations.plist') new_struct = plistlib.readPlist('/tmp/customizations.plist') # read/writePlist accepts file-like objects as well as paths. plistlib.writePlist(data_struct, sys.stdout) </pre></div> </div> </div> <div class="section" id="ctypes-enhancements"> <h3>ctypes Enhancements<a class="headerlink" href="#ctypes-enhancements" title="Permalink to this headline">露</a></h3> Thomas Heller continued to maintain and enhance the <a title="A foreign function library for Python." class="reference external" href="../library/ctypes.html#module-ctypes"><tt class="xref docutils literal">ctypes</tt></a> module. <a title="A foreign function library for Python." class="reference external" href="../library/ctypes.html#module-ctypes"><tt class="xref docutils literal">ctypes</tt></a> now supports a <tt class="xref docutils literal">c_bool</tt> datatype that represents the C99 <tt class="docutils literal">bool</tt> type. (Contributed by David Remahl; <a class="reference external" href="http://bugs.python.org/issue1649190">issue 1649190</a>.) The <a title="A foreign function library for Python." class="reference external" href="../library/ctypes.html#module-ctypes"><tt class="xref docutils literal">ctypes</tt></a> string, buffer and array types have improved support for extended slicing syntax, where various combinations of <tt class="docutils literal">(start, stop, step)</tt> are supplied. (Implemented by Thomas Wouters.) All <a title="A foreign function library for Python." class="reference external" href="../library/ctypes.html#module-ctypes"><tt class="xref docutils literal">ctypes</tt></a> data types now support <tt class="xref docutils literal">from_buffer()</tt> and <tt class="xref docutils literal">from_buffer_copy()</tt> methods that create a ctypes instance based on a provided buffer object. <tt class="xref docutils literal">from_buffer_copy()</tt> copies the contents of the object, while <tt class="xref docutils literal">from_buffer()</tt> will share the same memory area. A new calling convention tells <a title="A foreign function library for Python." class="reference external" href="../library/ctypes.html#module-ctypes"><tt class="xref docutils literal">ctypes</tt></a> to clear the <tt class="docutils literal">errno</tt> or Win32 LastError variables at the outset of each wrapped call. (Implemented by Thomas Heller; <a class="reference external" href="http://bugs.python.org/issue1798">issue 1798</a>.) You can now retrieve the Unix <tt class="docutils literal">errno</tt> variable after a function call. When creating a wrapped function, you can supply <tt class="docutils literal">use_errno=True</tt> as a keyword parameter to the <tt class="xref docutils literal">DLL()</tt> function and then call the module-level methods <tt class="xref docutils literal">set_errno()</tt> and <tt class="xref docutils literal">get_errno()</tt> to set and retrieve the error value. The Win32 LastError variable is similarly supported by the <tt class="xref docutils literal">DLL()</tt>, <tt class="xref docutils literal">OleDLL()</tt>, and <tt class="xref docutils literal">WinDLL()</tt> functions. You supply <tt class="docutils literal">use_last_error=True</tt> as a keyword parameter and then call the module-level methods <tt class="xref docutils literal">set_last_error()</tt> and <tt class="xref docutils literal">get_last_error()</tt>. The <tt class="xref docutils literal">byref()</tt> function, used to retrieve a pointer to a ctypes instance, now has an optional offset parameter that is a byte count that will be added to the returned pointer. </div> <div class="section" id="improved-ssl-support"> <h3>Improved SSL Support<a class="headerlink" href="#improved-ssl-support" title="Permalink to this headline">露</a></h3> Bill Janssen made extensive improvements to Python 2.6’s support for the Secure Sockets Layer by adding a new module, <a title="TLS/SSL wrapper for socket objects" class="reference external" href="../library/ssl.html#module-ssl"><tt class="xref docutils literal">ssl</tt></a>, that’s built atop the <a class="reference external" href="http://www.openssl.org/">OpenSSL</a> library. This new module provides more control over the protocol negotiated, the X.509 certificates used, and has better support for writing SSL servers (as opposed to clients) in Python. The existing SSL support in the <a title="Low-level networking interface." class="reference external" href="../library/socket.html#module-socket"><tt class="xref docutils literal">socket</tt></a> module hasn’t been removed and continues to work, though it will be removed in Python 3.0. To use the new module, you must first create a TCP connection in the usual way and then pass it to the <a title="ssl.wrap_socket" class="reference external" href="../library/ssl.html#ssl.wrap_socket"><tt class="xref docutils literal">ssl.wrap_socket()</tt></a> function. It’s possible to specify whether a certificate is required, and to obtain certificate info by calling the <tt class="xref docutils literal">getpeercert()</tt> method. <div class="admonition-see-also admonition seealso"> See also The documentation for the <a title="TLS/SSL wrapper for socket objects" class="reference external" href="../library/ssl.html#module-ssl"><tt class="xref docutils literal">ssl</tt></a> module. </div> </div> </div> <div class="section" id="deprecations-and-removals"> <h2>Deprecations and Removals<a class="headerlink" href="#deprecations-and-removals" title="Permalink to this headline">露</a></h2> <ul> <li>String exceptions have been removed. Attempting to use them raises a <a title="TypeError" class="reference external" href="../library/exceptions.html#TypeError"><tt class="xref docutils literal">TypeError</tt></a>. </li> <li>Changes to the <a title="Exception" class="reference external" href="../library/exceptions.html#Exception"><tt class="xref docutils literal">Exception</tt></a> interface as dictated by <a class="reference external" href="http://www.python.org/dev/peps/pep-0352">PEP 352</a> continue to be made. For 2.6, the <tt class="xref docutils literal">message</tt> attribute is being deprecated in favor of the <tt class="xref docutils literal">args</tt> attribute. </li> <li>(3.0-warning mode) Python 3.0 will feature a reorganized standard library that will drop many outdated modules and rename others. Python 2.6 running in 3.0-warning mode will warn about these modules when they are imported. The list of deprecated modules is: <tt class="xref docutils literal">audiodev</tt>, <tt class="xref docutils literal">bgenlocations</tt>, <tt class="xref docutils literal">buildtools</tt>, <tt class="xref docutils literal">bundlebuilder</tt>, <tt class="xref docutils literal">Canvas</tt>, <tt class="xref docutils literal">compiler</tt>, <tt class="xref docutils literal">dircache</tt>, <tt class="xref docutils literal">dl</tt>, <tt class="xref docutils literal">fpformat</tt>, <tt class="xref docutils literal">gensuitemodule</tt>, <tt class="xref docutils literal">ihooks</tt>, <tt class="xref docutils literal">imageop</tt>, <tt class="xref docutils literal">imgfile</tt>, <tt class="xref docutils literal">linuxaudiodev</tt>, <tt class="xref docutils literal">mhlib</tt>, <tt class="xref docutils literal">mimetools</tt>, <tt class="xref docutils literal">multifile</tt>, <tt class="xref docutils literal">new</tt>, <tt class="xref docutils literal">pure</tt>, <tt class="xref docutils literal">statvfs</tt>, <tt class="xref docutils literal">sunaudiodev</tt>, <tt class="xref docutils literal">test.testall</tt>, and <tt class="xref docutils literal">toaiff</tt>. </li> <li>The <tt class="xref docutils literal">gopherlib</tt> module has been removed. </li> <li>The <tt class="xref docutils literal">MimeWriter</tt> module and <tt class="xref docutils literal">mimify</tt> module have been deprecated; use the <a title="Package supporting the parsing, manipulating, and generating email messages, including MIME documents." class="reference external" href="../library/email.html#module-email"><tt class="xref docutils literal">email</tt></a> package instead. </li> <li>The <tt class="xref docutils literal">md5</tt> module has been deprecated; use the <a title="Secure hash and message digest algorithms." class="reference external" href="../library/hashlib.html#module-hashlib"><tt class="xref docutils literal">hashlib</tt></a> module instead. </li> <li>The <tt class="xref docutils literal">posixfile</tt> module has been deprecated; <a title="fcntl.lockf" class="reference external" href="../library/fcntl.html#fcntl.lockf"><tt class="xref docutils literal">fcntl.lockf()</tt></a> provides better locking. </li> <li>The <tt class="xref docutils literal">popen2</tt> module has been deprecated; use the <a title="Subprocess management." class="reference external" href="../library/subprocess.html#module-subprocess"><tt class="xref docutils literal">subprocess</tt></a> module. </li> <li>The <tt class="xref docutils literal">rgbimg</tt> module has been removed. </li> <li>The <tt class="xref docutils literal">sets</tt> module has been deprecated; it’s better to use the built-in <a title="set" class="reference external" href="../library/stdtypes.html#set"><tt class="xref docutils literal">set</tt></a> and <a title="frozenset" class="reference external" href="../library/stdtypes.html#frozenset"><tt class="xref docutils literal">frozenset</tt></a> types. </li> <li>The <tt class="xref docutils literal">sha</tt> module has been deprecated; use the <a title="Secure hash and message digest algorithms." class="reference external" href="../library/hashlib.html#module-hashlib"><tt class="xref docutils literal">hashlib</tt></a> module instead. </li> </ul> </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">露</a></h2> Changes to Python’s build process and to the C API include: <ul> <li>Python now must be compiled with C89 compilers (after 19 years!). This means that the Python source tree has dropped its own implementations of <tt class="xref docutils literal">memmove()</tt> and <tt class="xref docutils literal">strerror()</tt>, which are in the C89 standard library. </li> <li>Python 2.6 can be built with Microsoft Visual Studio 2008 (version 9.0), and this is the new default compiler. See the <tt class="docutils literal">PCbuild</tt> directory for the build files. (Implemented by Christian Heimes.) </li> <li>On Mac OS X, Python 2.6 can be compiled as a 4-way universal build. The configure script can take a --with-universal-archs=[32-bit|64-bit|all] switch, controlling whether the binaries are built for 32-bit architectures (x86, PowerPC), 64-bit (x86-64 and PPC-64), or both. (Contributed by Ronald Oussoren.) </li> <li>The BerkeleyDB module now has a C API object, available as <tt class="docutils literal">bsddb.db.api</tt>. This object can be used by other C extensions that wish to use the <tt class="xref docutils literal">bsddb</tt> module for their own purposes. (Contributed by Duncan Grisby.) </li> <li>The new buffer interface, previously described in <a class="reference external" href="#pep-3118-revised-buffer-protocol">the PEP 3118 section</a>, adds <a title="PyObject_GetBuffer" class="reference external" href="../c-api/buffer.html#PyObject_GetBuffer"><tt class="xref docutils literal">PyObject_GetBuffer()</tt></a> and <a title="PyBuffer_Release" class="reference external" href="../c-api/buffer.html#PyBuffer_Release"><tt class="xref docutils literal">PyBuffer_Release()</tt></a>, as well as a few other functions. </li> <li>Python’s use of the C stdio library is now thread-safe, or at least as thread-safe as the underlying library is. A long-standing potential bug occurred if one thread closed a file object while another thread was reading from or writing to the object. In 2.6 file objects have a reference count, manipulated by the <tt class="xref docutils literal">PyFile_IncUseCount()</tt> and <tt class="xref docutils literal">PyFile_DecUseCount()</tt> functions. File objects can’t be closed unless the reference count is zero. <tt class="xref docutils literal">PyFile_IncUseCount()</tt> should be called while the GIL is still held, before carrying out an I/O operation using the <tt class="docutils literal">FILE *</tt> pointer, and <tt class="xref docutils literal">PyFile_DecUseCount()</tt> should be called immediately after the GIL is re-acquired. (Contributed by Antoine Pitrou and Gregory P. Smith.) </li> <li>Importing modules simultaneously in two different threads no longer deadlocks; it will now raise an <a title="ImportError" class="reference external" href="../library/exceptions.html#ImportError"><tt class="xref docutils literal">ImportError</tt></a>. A new API function, <a title="PyImport_ImportModuleNoBlock" class="reference external" href="../c-api/import.html#PyImport_ImportModuleNoBlock"><tt class="xref docutils literal">PyImport_ImportModuleNoBlock()</tt></a>, will look for a module in <tt class="docutils literal">sys.modules</tt> first, then try to import it after acquiring an import lock. If the import lock is held by another thread, an <a title="ImportError" class="reference external" href="../library/exceptions.html#ImportError"><tt class="xref docutils literal">ImportError</tt></a> is raised. (Contributed by Christian Heimes.) </li> <li>Several functions return information about the platform’s floating-point support. <a title="PyFloat_GetMax" class="reference external" href="../c-api/float.html#PyFloat_GetMax"><tt class="xref docutils literal">PyFloat_GetMax()</tt></a> returns the maximum representable floating point value, and <a title="PyFloat_GetMin" class="reference external" href="../c-api/float.html#PyFloat_GetMin"><tt class="xref docutils literal">PyFloat_GetMin()</tt></a> returns the minimum positive value. <a title="PyFloat_GetInfo" class="reference external" href="../c-api/float.html#PyFloat_GetInfo"><tt class="xref docutils literal">PyFloat_GetInfo()</tt></a> returns an object containing more information from the <tt class="docutils literal">float.h</tt> file, such as <tt class="docutils literal">"mant_dig"</tt> (number of digits in the mantissa), <tt class="docutils literal">"epsilon"</tt> (smallest difference between 1.0 and the next largest value representable), and several others. (Contributed by Christian Heimes; <a class="reference external" href="http://bugs.python.org/issue1534">issue 1534</a>.) </li> <li>C functions and methods that use <a title="PyComplex_AsCComplex" class="reference external" href="../c-api/complex.html#PyComplex_AsCComplex"><tt class="xref docutils literal">PyComplex_AsCComplex()</tt></a> will now accept arguments that have a <a title="object.__complex__" class="reference external" href="../reference/datamodel.html#object.__complex__"><tt class="xref docutils literal">__complex__()</tt></a> method. In particular, the functions in the <a title="Mathematical functions for complex numbers." class="reference external" href="../library/cmath.html#module-cmath"><tt class="xref docutils literal">cmath</tt></a> module will now accept objects with this method. This is a backport of a Python 3.0 change. (Contributed by Mark Dickinson; <a class="reference external" href="http://bugs.python.org/issue1675423">issue 1675423</a>.) </li> <li>Python’s C API now includes two functions for case-insensitive string comparisons, <tt class="docutils literal">PyOS_stricmp(char*, char*)</tt> and <tt class="docutils literal">PyOS_strnicmp(char*, char*, Py_ssize_t)</tt>. (Contributed by Christian Heimes; <a class="reference external" href="http://bugs.python.org/issue1635">issue 1635</a>.) </li> <li>Many C extensions define their own little macro for adding integers and strings to the module’s dictionary in the <tt class="docutils literal">init*</tt> function. Python 2.6 finally defines standard macros for adding values to a module, <a title="PyModule_AddStringMacro" class="reference external" href="../c-api/module.html#PyModule_AddStringMacro"><tt class="xref docutils literal">PyModule_AddStringMacro</tt></a> and <a title="PyModule_AddIntMacro" class="reference external" href="../c-api/module.html#PyModule_AddIntMacro"><tt class="xref docutils literal">PyModule_AddIntMacro()</tt></a>. (Contributed by Christian Heimes.) </li> <li>Some macros were renamed in both 3.0 and 2.6 to make it clearer that they are macros, not functions. <tt class="xref docutils literal">Py_Size()</tt> became <tt class="xref docutils literal">Py_SIZE()</tt>, <tt class="xref docutils literal">Py_Type()</tt> became <tt class="xref docutils literal">Py_TYPE()</tt>, and <tt class="xref docutils literal">Py_Refcnt()</tt> became <tt class="xref docutils literal">Py_REFCNT()</tt>. The mixed-case macros are still available in Python 2.6 for backward compatibility. (<a class="reference external" href="http://bugs.python.org/issue1629">issue 1629</a>) </li> <li>Distutils now places C extensions it builds in a different directory when running on a debug version of Python. (Contributed by Collin Winter; <a class="reference external" href="http://bugs.python.org/issue1530959">issue 1530959</a>.) </li> <li>Several basic data types, such as integers and strings, maintain internal free lists of objects that can be re-used. The data structures for these free lists now follow a naming convention: the variable is always named <tt class="docutils literal">free_list</tt>, the counter is always named <tt class="docutils literal">numfree</tt>, and a macro <tt class="docutils literal">Py<typename>_MAXFREELIST</tt> is always defined. </li> <li>A new Makefile target, “make patchcheck”, prepares the Python source tree for making a patch: it fixes trailing whitespace in all modified <tt class="docutils literal">.py</tt> files, checks whether the documentation has been changed, and reports whether the <tt class="docutils literal">Misc/ACKS</tt> and <tt class="docutils literal">Misc/NEWS</tt> files have been updated. (Contributed by Brett Cannon.) Another new target, “make profile-opt”, compiles a Python binary using GCC’s profile-guided optimization. It compiles Python with profiling enabled, runs the test suite to obtain a set of profiling results, and then compiles using these results for optimization. (Contributed by Gregory P. Smith.) </li> </ul> <div class="section" id="port-specific-changes-windows"> <h3>Port-Specific Changes: Windows<a class="headerlink" href="#port-specific-changes-windows" title="Permalink to this headline">露</a></h3> <ul> <li>The support for Windows 95, 98, ME and NT4 has been dropped. Python 2.6 requires at least Windows 2000 SP4. </li> <li>The new default compiler on Windows is Visual Studio 2008 (version 9.0). The build directories for Visual Studio 2003 (version 7.1) and 2005 (version 8.0) were moved into the PC/ directory. The new <tt class="docutils literal">PCbuild</tt> directory supports cross compilation for X64, debug builds and Profile Guided Optimization (PGO). PGO builds are roughly 10% faster than normal builds. (Contributed by Christian Heimes with help from Amaury Forgeot d’Arc and Martin von Loewis.) </li> <li>The <a title="(Windows) Miscellaneous useful routines from the MS VC++ runtime." class="reference external" href="../library/msvcrt.html#module-msvcrt"><tt class="xref docutils literal">msvcrt</tt></a> module now supports both the normal and wide char variants of the console I/O API. The <tt class="xref docutils literal">getwch()</tt> function reads a keypress and returns a Unicode value, as does the <tt class="xref docutils literal">getwche()</tt> function. The <tt class="xref docutils literal">putwch()</tt> function takes a Unicode character and writes it to the console. (Contributed by Christian Heimes.) </li> <li><a title="os.path.expandvars" class="reference external" href="../library/os.path.html#os.path.expandvars"><tt class="xref docutils literal">os.path.expandvars()</tt></a> will now expand environment variables in the form “%var%”, and “~user” will be expanded into the user’s home directory path. (Contributed by Josiah Carlson; <a class="reference external" href="http://bugs.python.org/issue957650">issue 957650</a>.) </li> <li>The <a title="Low-level networking interface." class="reference external" href="../library/socket.html#module-socket"><tt class="xref docutils literal">socket</tt></a> module’s socket objects now have an <tt class="xref docutils literal">ioctl()</tt> method that provides a limited interface to the <tt class="xref docutils literal">WSAIoctl()</tt> system interface. </li> <li>The <tt class="xref docutils literal">_winreg</tt> module now has a function, <tt class="xref docutils literal">ExpandEnvironmentStrings()</tt>, that expands environment variable references such as <tt class="docutils literal">%NAME%</tt> in an input string. The handle objects provided by this module now support the context protocol, so they can be used in <a class="reference external" href="../reference/compound_stmts.html#with"><tt class="xref docutils literal">with</tt></a> statements. (Contributed by Christian Heimes.) <tt class="xref docutils literal">_winreg</tt> also has better support for x64 systems, exposing the <tt class="xref docutils literal">DisableReflectionKey()</tt>, <tt class="xref docutils literal">EnableReflectionKey()</tt>, and <tt class="xref docutils literal">QueryReflectionKey()</tt> functions, which enable and disable registry reflection for 32-bit processes running on 64-bit systems. (<a class="reference external" href="http://bugs.python.org/issue1753245">issue 1753245</a>) </li> <li>The <a title="(Windows) Creation of Microsoft Installer files, and CAB files." class="reference external" href="../library/msilib.html#module-msilib"><tt class="xref docutils literal">msilib</tt></a> module’s <tt class="xref docutils literal">Record</tt> object gained <tt class="xref docutils literal">GetInteger()</tt> and <tt class="xref docutils literal">GetString()</tt> methods that return field values as an integer or a string. (Contributed by Floris Bruynooghe; <a class="reference external" href="http://bugs.python.org/issue2125">issue 2125</a>.) </li> </ul> </div> <div class="section" id="port-specific-changes-mac-os-x"> <h3>Port-Specific Changes: Mac OS X<a class="headerlink" href="#port-specific-changes-mac-os-x" title="Permalink to this headline">露</a></h3> <ul class="simple"> <li>When compiling a framework build of Python, you can now specify the framework name to be used by providing the --with-framework-name= option to the configure script.</li> <li>The <tt class="xref docutils literal">macfs</tt> module has been removed. This in turn required the <tt class="xref docutils literal">macostools.touched()</tt> function to be removed because it depended on the <tt class="xref docutils literal">macfs</tt> module. (<a class="reference external" href="http://bugs.python.org/issue1490190">issue 1490190</a>)</li> <li>Many other Mac OS modules have been deprecated and will removed in Python 3.0: <tt class="xref docutils literal">_builtinSuites</tt>, <tt class="xref docutils literal">aepack</tt>, <tt class="xref docutils literal">aetools</tt>, <tt class="xref docutils literal">aetypes</tt>, <tt class="xref docutils literal">applesingle</tt>, <tt class="xref docutils literal">appletrawmain</tt>, <tt class="xref docutils literal">appletrunner</tt>, <tt class="xref docutils literal">argvemulator</tt>, <tt class="xref docutils literal">Audio_mac</tt>, <tt class="xref docutils literal">autoGIL</tt>, <tt class="xref docutils literal">Carbon</tt>, <tt class="xref docutils literal">cfmfile</tt>, <tt class="xref docutils literal">CodeWarrior</tt>, <tt class="xref docutils literal">ColorPicker</tt>, <tt class="xref docutils literal">EasyDialogs</tt>, <tt class="xref docutils literal">Explorer</tt>, <tt class="xref docutils literal">Finder</tt>, <tt class="xref docutils literal">FrameWork</tt>, <tt class="xref docutils literal">findertools</tt>, <tt class="xref docutils literal">ic</tt>, <tt class="xref docutils literal">icglue</tt>, <tt class="xref docutils literal">icopen</tt>, <tt class="xref docutils literal">macerrors</tt>, <tt class="xref docutils literal">MacOS</tt>, <tt class="xref docutils literal">macfs</tt>, <tt class="xref docutils literal">macostools</tt>, <tt class="xref docutils literal">macresource</tt>, <tt class="xref docutils literal">MiniAEFrame</tt>, <tt class="xref docutils literal">Nav</tt>, <tt class="xref docutils literal">Netscape</tt>, <tt class="xref docutils literal">OSATerminology</tt>, <tt class="xref docutils literal">pimp</tt>, <tt class="xref docutils literal">PixMapWrapper</tt>, <tt class="xref docutils literal">StdSuites</tt>, <tt class="xref docutils literal">SystemEvents</tt>, <tt class="xref docutils literal">Terminal</tt>, and <tt class="xref docutils literal">terminalcommand</tt>.</li> </ul> </div> <div class="section" id="port-specific-changes-irix"> <h3>Port-Specific Changes: IRIX<a class="headerlink" href="#port-specific-changes-irix" title="Permalink to this headline">露</a></h3> A number of old IRIX-specific modules were deprecated and will be removed in Python 3.0: <tt class="xref docutils literal">al</tt> and <tt class="xref docutils literal">AL</tt>, <tt class="xref docutils literal">cd</tt>, <tt class="xref docutils literal">cddb</tt>, <tt class="xref docutils literal">cdplayer</tt>, <tt class="xref docutils literal">CL</tt> and <tt class="xref docutils literal">cl</tt>, <tt class="xref docutils literal">DEVICE</tt>, <tt class="xref docutils literal">ERRNO</tt>, <tt class="xref docutils literal">FILE</tt>, <tt class="xref docutils literal">FL</tt> and <tt class="xref docutils literal">fl</tt>, <tt class="xref docutils literal">flp</tt>, <tt class="xref docutils literal">fm</tt>, <tt class="xref docutils literal">GET</tt>, <tt class="xref docutils literal">GLWS</tt>, <tt class="xref docutils literal">GL</tt> and <tt class="xref docutils literal">gl</tt>, <tt class="xref docutils literal">IN</tt>, <tt class="xref docutils literal">IOCTL</tt>, <tt class="xref docutils literal">jpeg</tt>, <tt class="xref docutils literal">panelparser</tt>, <tt class="xref docutils literal">readcd</tt>, <tt class="xref docutils literal">SV</tt> and <tt class="xref docutils literal">sv</tt>, <tt class="xref docutils literal">torgb</tt>, <tt class="xref docutils literal">videoreader</tt>, and <tt class="xref docutils literal">WAIT</tt>. </div> </div> <div class="section" id="porting-to-python-2-6"> <h2>Porting to Python 2.6<a class="headerlink" href="#porting-to-python-2-6" title="Permalink to this headline">露</a></h2> This section lists previously described changes and other bugfixes that may require changes to your code: <ul> <li>Classes that aren’t supposed to be hashable should set <tt class="docutils literal">__hash__ = None</tt> in their definitions to indicate the fact. </li> <li>String exceptions have been removed. Attempting to use them raises a <a title="TypeError" class="reference external" href="../library/exceptions.html#TypeError"><tt class="xref docutils literal">TypeError</tt></a>. </li> <li>The <a title="object.__init__" class="reference external" href="../reference/datamodel.html#object.__init__"><tt class="xref docutils literal">__init__()</tt></a> method of <a title="collections.deque" class="reference external" href="../library/collections.html#collections.deque"><tt class="xref docutils literal">collections.deque</tt></a> now clears any existing contents of the deque before adding elements from the iterable. This change makes the behavior match <tt class="docutils literal">list.__init__()</tt>. </li> <li><a title="object.__init__" class="reference external" href="../reference/datamodel.html#object.__init__"><tt class="xref docutils literal">object.__init__()</tt></a> previously accepted arbitrary arguments and keyword arguments, ignoring them. In Python 2.6, this is no longer allowed and will result in a <a title="TypeError" class="reference external" href="../library/exceptions.html#TypeError"><tt class="xref docutils literal">TypeError</tt></a>. This will affect <a title="object.__init__" class="reference external" href="../reference/datamodel.html#object.__init__"><tt class="xref docutils literal">__init__()</tt></a> methods that end up calling the corresponding method on <a title="object" class="reference external" href="../library/functions.html#object"><tt class="xref docutils literal">object</tt></a> (perhaps through using <a title="super" class="reference external" href="../library/functions.html#super"><tt class="xref docutils literal">super()</tt></a>). See <a class="reference external" href="http://bugs.python.org/issue1683368">issue 1683368</a> for discussion. </li> <li>The <tt class="xref docutils literal">Decimal</tt> constructor now accepts leading and trailing whitespace when passed a string. Previously it would raise an <tt class="xref docutils literal">InvalidOperation</tt> exception. On the other hand, the <tt class="xref docutils literal">create_decimal()</tt> method of <tt class="xref docutils literal">Context</tt> objects now explicitly disallows extra whitespace, raising a <tt class="xref docutils literal">ConversionSyntax</tt> exception. </li> <li>Due to an implementation accident, if you passed a file path to the built-in <a title="__import__" class="reference external" href="../library/functions.html#__import__"><tt class="xref docutils literal">__import__()</tt></a> function, it would actually import the specified file. This was never intended to work, however, and the implementation now explicitly checks for this case and raises an <a title="ImportError" class="reference external" href="../library/exceptions.html#ImportError"><tt class="xref docutils literal">ImportError</tt></a>. </li> <li>C API: the <a title="PyImport_Import" class="reference external" href="../c-api/import.html#PyImport_Import"><tt class="xref docutils literal">PyImport_Import()</tt></a> and <a title="PyImport_ImportModule" class="reference external" href="../c-api/import.html#PyImport_ImportModule"><tt class="xref docutils literal">PyImport_ImportModule()</tt></a> functions now default to absolute imports, not relative imports. This will affect C extensions that import other modules. </li> <li>C API: extension data types that shouldn’t be hashable should define their <tt class="docutils literal">tp_hash</tt> slot to <a title="PyObject_HashNotImplemented" class="reference external" href="../c-api/object.html#PyObject_HashNotImplemented"><tt class="xref docutils literal">PyObject_HashNotImplemented()</tt></a>. </li> <li>The <a title="Low-level networking interface." class="reference external" href="../library/socket.html#module-socket"><tt class="xref docutils literal">socket</tt></a> module exception <a title="socket.error" class="reference external" href="../library/socket.html#socket.error"><tt class="xref docutils literal">socket.error</tt></a> now inherits from <a title="IOError" class="reference external" href="../library/exceptions.html#IOError"><tt class="xref docutils literal">IOError</tt></a>. Previously it wasn’t a subclass of <tt class="xref docutils literal">StandardError</tt> but now it is, through <a title="IOError" class="reference external" href="../library/exceptions.html#IOError"><tt class="xref docutils literal">IOError</tt></a>. (Implemented by Gregory P. Smith; <a class="reference external" href="http://bugs.python.org/issue1706815">issue 1706815</a>.) </li> <li>The <tt class="xref docutils literal">xmlrpclib</tt> module no longer automatically converts <a title="datetime.date" class="reference external" href="../library/datetime.html#datetime.date"><tt class="xref docutils literal">datetime.date</tt></a> and <a title="datetime.time" class="reference external" href="../library/datetime.html#datetime.time"><tt class="xref docutils literal">datetime.time</tt></a> to the <tt class="xref docutils literal">xmlrpclib.DateTime</tt> type; the conversion semantics were not necessarily correct for all applications. Code using <tt class="xref docutils literal">xmlrpclib</tt> should convert <tt class="xref docutils literal">date</tt> and <tt class="xref docutils literal">time</tt> instances. (<a class="reference external" href="http://bugs.python.org/issue1330538">issue 1330538</a>) </li> <li>(3.0-warning mode) The <a title="Exception" class="reference external" href="../library/exceptions.html#Exception"><tt class="xref docutils literal">Exception</tt></a> class now warns when accessed using slicing or index access; having <a title="Exception" class="reference external" href="../library/exceptions.html#Exception"><tt class="xref docutils literal">Exception</tt></a> behave like a tuple is being phased out. </li> <li>(3.0-warning mode) inequality comparisons between two dictionaries or two objects that don’t implement comparison methods are reported as warnings. <tt class="docutils literal">dict1 == dict2</tt> still works, but <tt class="docutils literal">dict1 < dict2</tt> is being phased out. Comparisons between cells, which are an implementation detail of Python’s scoping rules, also cause warnings because such comparisons are forbidden entirely in 3.0. </li> </ul> </div> <div class="section" id="acknowledgements"> <h2>Acknowledgements<a class="headerlink" href="#acknowledgements" title="Permalink to this headline">露</a></h2> The author would like to thank the following people for offering suggestions, corrections and assistance with various drafts of this article: Georg Brandl, Steve Brown, Nick Coghlan, Ralph Corderoy, Jim Jewett, Kent Johnson, Chris Lambacher, Martin Michlmayr, Antoine Pitrou, Brian Warner. </div> </div> </div> </div> </div> <div class="sphinxsidebar"> <div class="sphinxsidebarwrapper"> <h3><a href="../contents.html">Table Of Contents</a></h3> <ul> <li><a class="reference external" href="#">What’s New in Python 2.6</a><ul> <li><a class="reference external" href="#python-3-0">Python 3.0</a></li> <li><a class="reference external" href="#changes-to-the-development-process">Changes to the Development Process</a><ul> <li><a class="reference external" href="#new-issue-tracker-roundup">New Issue Tracker: Roundup</a></li> <li><a class="reference external" href="#new-documentation-format-restructuredtext-using-sphinx">New Documentation Format: reStructuredText Using Sphinx</a></li> </ul> </li> <li><a class="reference external" href="#pep-343-the-with-statement">PEP 343: The ‘with’ statement</a><ul> <li><a class="reference external" href="#writing-context-managers">Writing Context Managers</a></li> <li><a class="reference external" href="#the-contextlib-module">The contextlib module</a></li> </ul> </li> <li><a class="reference external" href="#pep-366-explicit-relative-imports-from-a-main-module">PEP 366: Explicit Relative Imports From a Main Module</a></li> <li><a class="reference external" href="#pep-370-per-user-site-packages-directory">PEP 370: Per-user <tt class="docutils literal">site-packages</tt> Directory</a></li> <li><a class="reference external" href="#pep-371-the-multiprocessing-package">PEP 371: The <tt class="docutils literal">multiprocessing</tt> Package</a></li> <li><a class="reference external" href="#pep-3101-advanced-string-formatting">PEP 3101: Advanced String Formatting</a></li> <li><a class="reference external" href="#pep-3105-print-as-a-function">PEP 3105: <tt class="docutils literal">print</tt> As a Function</a></li> <li><a class="reference external" href="#pep-3110-exception-handling-changes">PEP 3110: Exception-Handling Changes</a></li> <li><a class="reference external" href="#pep-3112-byte-literals">PEP 3112: Byte Literals</a></li> <li><a class="reference external" href="#pep-3116-new-i-o-library">PEP 3116: New I/O Library</a></li> <li><a class="reference external" href="#pep-3118-revised-buffer-protocol">PEP 3118: Revised Buffer Protocol</a></li> <li><a class="reference external" href="#pep-3119-abstract-base-classes">PEP 3119: Abstract Base Classes</a></li> <li><a class="reference external" href="#pep-3127-integer-literal-support-and-syntax">PEP 3127: Integer Literal Support and Syntax</a></li> <li><a class="reference external" href="#pep-3129-class-decorators">PEP 3129: Class Decorators</a></li> <li><a class="reference external" href="#pep-3141-a-type-hierarchy-for-numbers">PEP 3141: A Type Hierarchy for Numbers</a><ul> <li><a class="reference external" href="#the-fractions-module">The <tt class="docutils literal">fractions</tt> Module</a></li> </ul> </li> <li><a class="reference external" href="#other-language-changes">Other Language Changes</a><ul> <li><a class="reference external" href="#optimizations">Optimizations</a></li> <li><a class="reference external" href="#interpreter-changes">Interpreter Changes</a></li> </ul> </li> <li><a class="reference external" href="#new-and-improved-modules">New and Improved Modules</a><ul> <li><a class="reference external" href="#the-ast-module">The <tt class="docutils literal">ast</tt> module</a></li> <li><a class="reference external" href="#the-future-builtins-module">The <tt class="docutils literal">future_builtins</tt> module</a></li> <li><a class="reference external" href="#the-json-module-javascript-object-notation">The <tt class="docutils literal">json</tt> module: JavaScript Object Notation</a></li> <li><a class="reference external" href="#the-plistlib-module-a-property-list-parser">The <tt class="docutils literal">plistlib</tt> module: A Property-List Parser</a></li> <li><a class="reference external" href="#ctypes-enhancements">ctypes Enhancements</a></li> <li><a class="reference external" href="#improved-ssl-support">Improved SSL Support</a></li> </ul> </li> <li><a class="reference external" href="#deprecations-and-removals">Deprecations and Removals</a></li> <li><a class="reference external" href="#build-and-c-api-changes">Build and C API Changes</a><ul> <li><a class="reference external" href="#port-specific-changes-windows">Port-Specific Changes: Windows</a></li> <li><a class="reference external" href="#port-specific-changes-mac-os-x">Port-Specific Changes: Mac OS X</a></li> <li><a class="reference external" href="#port-specific-changes-irix">Port-Specific Changes: IRIX</a></li> </ul> </li> <li><a class="reference external" href="#porting-to-python-2-6">Porting to Python 2.6</a></li> <li><a class="reference external" href="#acknowledgements">Acknowledgements</a></li> </ul> </li> </ul> <h4>Previous topic</h4> <a href="3.0.html" title="previous chapter">What’s New In Python 3.0</a> <h4>Next topic</h4> <a href="2.5.html" title="next chapter">What’s New in Python 2.5</a> <h3>This Page</h3> <ul class="this-page-menu"> <li><a href="../bugs.html">Report a Bug</a></li> <li><a href="../_sources/whatsnew/2.6.txt" rel="nofollow">Show Source</a></li> </ul> <div id="searchbox" style="display: none"> <h3>Quick search</h3> <form class="search" action="../search.html" method="get"> <input type="text" name="q" size="18" /> <input type="submit" value="Go" /> <input type="hidden" name="check_keywords" value="yes" /> <input type="hidden" name="area" value="default" /> </form> Enter search terms or a module, class or function name. </div> <script type="text/javascript">$('#searchbox').show(0);</script> </div> </div> <div class="clearer"></div> </div> <div class="related"> <h3>Navigation</h3> <ul> <li class="right" style="margin-right: 10px"> <a href="../genindex.html" title="General Index" >index</a></li> <li class="right" > <a href="../modindex.html" title="Global Module Index" >modules</a> |</li> <li class="right" > <a href="2.5.html" title="What鈥檚 New in Python 2.5" >next</a> |</li> <li class="right" > <a href="3.0.html" title="What鈥檚 New In Python 3.0" >previous</a> |</li> <li><img src="../_static/py.png" alt="" style="vertical-align: middle; margin-top: -1px"/></li> <li><a href="../index.html">Python v3.1.5 documentation</a> »</li> <li><a href="index.html" >What’s New in Python</a> »</li> </ul> </div> <div class="footer"> © <a href="../copyright.html">Copyright</a> 1990-2012, Python Software Foundation. The Python Software Foundation is a non-profit corporation. <a href="http://www.python.org/psf/donations/">Please donate.</a> Last updated on Apr 09, 2012. <a href="../bugs.html">Found a bug</a>? Created using <a href="http://sphinx.pocoo.org/">Sphinx</a> 0.6.5. </div> </body> </html>