^^&&^^

Edit File: extending.html

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

<html xmlns="http://www.w3.org/1999/xhtml">
 <head>
 <meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
 
 <title>Extending/Embedding FAQ &mdash; Python 2.7.5 documentation</title>
 
 <link rel="stylesheet" href="../_static/default.css" type="text/css" />
 <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
 
 <script type="text/javascript">
 var DOCUMENTATION_OPTIONS = {
 URL_ROOT: '../',
 VERSION: '2.7.5',
 COLLAPSE_INDEX: false,
 FILE_SUFFIX: '.html',
 HAS_SOURCE: true
 };
 </script>
 <script type="text/javascript" src="../_static/jquery.js"></script>
 <script type="text/javascript" src="../_static/underscore.js"></script>
 <script type="text/javascript" src="../_static/doctools.js"></script>
 <script type="text/javascript" src="../_static/sidebar.js"></script>
 <link rel="search" type="application/opensearchdescription+xml"
 title="Search within Python 2.7.5 documentation"
 href="../_static/opensearch.xml"/>
 <link rel="author" title="About these documents" href="../about.html" />
 <link rel="copyright" title="Copyright" href="../copyright.html" />
 <link rel="top" title="Python 2.7.5 documentation" href="../index.html" />
 <link rel="up" title="Python Frequently Asked Questions" href="index.html" />
 <link rel="next" title="Python on Windows FAQ" href="windows.html" />
 <link rel="prev" title="Library and Extension FAQ" href="library.html" />
 <link rel="shortcut icon" type="image/png" href="../_static/py.png" />
 <script type="text/javascript" src="../_static/copybutton.js"></script>

</head>
 <body>
 <div class="related">
 <h3>Navigation</h3>
 <ul>
 <li class="right" style="margin-right: 10px">
 <a href="../genindex.html" title="General Index"
 accesskey="I">index</a></li>
 <li class="right" >
 <a href="../py-modindex.html" title="Python Module Index"
 >modules</a> |</li>
 <li class="right" >
 <a href="windows.html" title="Python on Windows FAQ"
 accesskey="N">next</a> |</li>
 <li class="right" >
 <a href="library.html" title="Library and Extension FAQ"
 accesskey="P">previous</a> |</li>
 <li><img src="../_static/py.png" alt=""
 style="vertical-align: middle; margin-top: -1px"/></li>
 <li><a href="http://www.python.org/">Python</a> &raquo;</li>
 <li>
 <a href="../index.html">Python 2.7.5 documentation</a> &raquo;
 </li>

<li><a href="index.html" accesskey="U">Python Frequently Asked Questions</a> &raquo;</li> 
 </ul>
 </div>

<div class="document">
 <div class="documentwrapper">
 <div class="bodywrapper">
 <div class="body">
 
 <div class="section" id="extending-embedding-faq">
<h1><a class="toc-backref" href="#id2">Extending/Embedding FAQ</a><a class="headerlink" href="#extending-embedding-faq" title="Permalink to this headline">¶</a></h1>
<div class="contents topic" id="contents">
Contents
<ul class="simple">
<li><a class="reference internal" href="#extending-embedding-faq" id="id2">Extending/Embedding FAQ</a><ul>
<li><a class="reference internal" href="#can-i-create-my-own-functions-in-c" id="id3">Can I create my own functions in C?</a></li>
<li><a class="reference internal" href="#id1" id="id4">Can I create my own functions in C++?</a></li>
<li><a class="reference internal" href="#writing-c-is-hard-are-there-any-alternatives" id="id5">Writing C is hard; are there any alternatives?</a></li>
<li><a class="reference internal" href="#how-can-i-execute-arbitrary-python-statements-from-c" id="id6">How can I execute arbitrary Python statements from C?</a></li>
<li><a class="reference internal" href="#how-can-i-evaluate-an-arbitrary-python-expression-from-c" id="id7">How can I evaluate an arbitrary Python expression from C?</a></li>
<li><a class="reference internal" href="#how-do-i-extract-c-values-from-a-python-object" id="id8">How do I extract C values from a Python object?</a></li>
<li><a class="reference internal" href="#how-do-i-use-py-buildvalue-to-create-a-tuple-of-arbitrary-length" id="id9">How do I use Py_BuildValue() to create a tuple of arbitrary length?</a></li>
<li><a class="reference internal" href="#how-do-i-call-an-object-s-method-from-c" id="id10">How do I call an object&#8217;s method from C?</a></li>
<li><a class="reference internal" href="#how-do-i-catch-the-output-from-pyerr-print-or-anything-that-prints-to-stdout-stderr" id="id11">How do I catch the output from PyErr_Print() (or anything that prints to stdout/stderr)?</a></li>
<li><a class="reference internal" href="#how-do-i-access-a-module-written-in-python-from-c" id="id12">How do I access a module written in Python from C?</a></li>
<li><a class="reference internal" href="#how-do-i-interface-to-c-objects-from-python" id="id13">How do I interface to C++ objects from Python?</a></li>
<li><a class="reference internal" href="#i-added-a-module-using-the-setup-file-and-the-make-fails-why" id="id14">I added a module using the Setup file and the make fails; why?</a></li>
<li><a class="reference internal" href="#how-do-i-debug-an-extension" id="id15">How do I debug an extension?</a></li>
<li><a class="reference internal" href="#i-want-to-compile-a-python-module-on-my-linux-system-but-some-files-are-missing-why" id="id16">I want to compile a Python module on my Linux system, but some files are missing. Why?</a></li>
<li><a class="reference internal" href="#what-does-systemerror-pyimport-fixupextension-module-yourmodule-not-loaded-mean" id="id17">What does &#8220;SystemError: _PyImport_FixupExtension: module yourmodule not loaded&#8221; mean?</a></li>
<li><a class="reference internal" href="#how-do-i-tell-incomplete-input-from-invalid-input" id="id18">How do I tell &#8220;incomplete input&#8221; from &#8220;invalid input&#8221;?</a></li>
<li><a class="reference internal" href="#how-do-i-find-undefined-g-symbols-builtin-new-or-pure-virtual" id="id19">How do I find undefined g++ symbols __builtin_new or __pure_virtual?</a></li>
<li><a class="reference internal" href="#can-i-create-an-object-class-with-some-methods-implemented-in-c-and-others-in-python-e-g-through-inheritance" id="id20">Can I create an object class with some methods implemented in C and others in Python (e.g. through inheritance)?</a></li>
<li><a class="reference internal" href="#when-importing-module-x-why-do-i-get-undefined-symbol-pyunicodeucs2" id="id21">When importing module X, why do I get &#8220;undefined symbol: PyUnicodeUCS2*&#8221;?</a></li>
</ul>
</li>
</ul>
</div>
<div class="section" id="can-i-create-my-own-functions-in-c">
<h2><a class="toc-backref" href="#id3">Can I create my own functions in C?</a><a class="headerlink" href="#can-i-create-my-own-functions-in-c" title="Permalink to this headline">¶</a></h2>
Yes, you can create built-in modules containing functions, variables, exceptions
and even new types in C. This is explained in the document
<a class="reference internal" href="../extending/index.html#extending-index">Extending and Embedding the Python Interpreter</a>.
Most intermediate or advanced Python books will also cover this topic.
</div>
<div class="section" id="id1">
<h2><a class="toc-backref" href="#id4">Can I create my own functions in C++?</a><a class="headerlink" href="#id1" title="Permalink to this headline">¶</a></h2>
Yes, using the C compatibility features found in C++. Place <tt class="docutils literal">extern &quot;C&quot; {
... }</tt> around the Python include files and put <tt class="docutils literal">extern &quot;C&quot;</tt> before each
function that is going to be called by the Python interpreter. Global or static
C++ objects with constructors are probably not a good idea.
</div>
<div class="section" id="writing-c-is-hard-are-there-any-alternatives">
<h2><a class="toc-backref" href="#id5">Writing C is hard; are there any alternatives?</a><a class="headerlink" href="#writing-c-is-hard-are-there-any-alternatives" title="Permalink to this headline">¶</a></h2>
There are a number of alternatives to writing your own C extensions, depending
on what you&#8217;re trying to do.
If you need more speed, <a class="reference external" href="http://psyco.sourceforge.net/">Psyco</a> generates x86
assembly code from Python bytecode. You can use Psyco to compile the most
time-critical functions in your code, and gain a significant improvement with
very little effort, as long as you&#8217;re running on a machine with an
x86-compatible processor.
<a class="reference external" href="http://www.cosc.canterbury.ac.nz/~greg/python/Pyrex/">Pyrex</a> is a compiler
that accepts a slightly modified form of Python and generates the corresponding
C code. Pyrex makes it possible to write an extension without having to learn
Python&#8217;s C API.
If you need to interface to some C or C++ library for which no Python extension
currently exists, you can try wrapping the library&#8217;s data types and functions
with a tool such as <a class="reference external" href="http://www.swig.org">SWIG</a>. <a class="reference external" href="http://www.riverbankcomputing.co.uk/software/sip/">SIP</a>, <a class="reference external" href="http://cxx.sourceforge.net/">CXX</a> <a class="reference external" href="http://www.boost.org/libs/python/doc/index.html">Boost</a>, or <a class="reference external" href="http://www.scipy.org/Weave">Weave</a> are also alternatives for wrapping
C++ libraries.
</div>
<div class="section" id="how-can-i-execute-arbitrary-python-statements-from-c">
<h2><a class="toc-backref" href="#id6">How can I execute arbitrary Python statements from C?</a><a class="headerlink" href="#how-can-i-execute-arbitrary-python-statements-from-c" title="Permalink to this headline">¶</a></h2>
The highest-level function to do this is <a class="reference internal" href="../c-api/veryhigh.html#PyRun_SimpleString" title="PyRun_SimpleString"><tt class="xref c c-func docutils literal">PyRun_SimpleString()</tt></a> which takes
a single string argument to be executed in the context of the module
<tt class="docutils literal">__main__</tt> and returns 0 for success and -1 when an exception occurred
(including <tt class="docutils literal">SyntaxError</tt>). If you want more control, use
<a class="reference internal" href="../c-api/veryhigh.html#PyRun_String" title="PyRun_String"><tt class="xref c c-func docutils literal">PyRun_String()</tt></a>; see the source for <a class="reference internal" href="../c-api/veryhigh.html#PyRun_SimpleString" title="PyRun_SimpleString"><tt class="xref c c-func docutils literal">PyRun_SimpleString()</tt></a> in
<tt class="docutils literal">Python/pythonrun.c</tt>.
</div>
<div class="section" id="how-can-i-evaluate-an-arbitrary-python-expression-from-c">
<h2><a class="toc-backref" href="#id7">How can I evaluate an arbitrary Python expression from C?</a><a class="headerlink" href="#how-can-i-evaluate-an-arbitrary-python-expression-from-c" title="Permalink to this headline">¶</a></h2>
Call the function <a class="reference internal" href="../c-api/veryhigh.html#PyRun_String" title="PyRun_String"><tt class="xref c c-func docutils literal">PyRun_String()</tt></a> from the previous question with the
start symbol <a class="reference internal" href="../c-api/veryhigh.html#Py_eval_input" title="Py_eval_input"><tt class="xref c c-data docutils literal">Py_eval_input</tt></a>; it parses an expression, evaluates it and
returns its value.
</div>
<div class="section" id="how-do-i-extract-c-values-from-a-python-object">
<h2><a class="toc-backref" href="#id8">How do I extract C values from a Python object?</a><a class="headerlink" href="#how-do-i-extract-c-values-from-a-python-object" title="Permalink to this headline">¶</a></h2>
That depends on the object&#8217;s type. If it&#8217;s a tuple, <a class="reference internal" href="../c-api/tuple.html#PyTuple_Size" title="PyTuple_Size"><tt class="xref c c-func docutils literal">PyTuple_Size()</tt></a>
returns its length and <a class="reference internal" href="../c-api/tuple.html#PyTuple_GetItem" title="PyTuple_GetItem"><tt class="xref c c-func docutils literal">PyTuple_GetItem()</tt></a> returns the item at a specified
index. Lists have similar functions, <tt class="xref c c-func docutils literal">PyListSize()</tt> and
<a class="reference internal" href="../c-api/list.html#PyList_GetItem" title="PyList_GetItem"><tt class="xref c c-func docutils literal">PyList_GetItem()</tt></a>.
For strings, <a class="reference internal" href="../c-api/string.html#PyString_Size" title="PyString_Size"><tt class="xref c c-func docutils literal">PyString_Size()</tt></a> returns its length and
<a class="reference internal" href="../c-api/string.html#PyString_AsString" title="PyString_AsString"><tt class="xref c c-func docutils literal">PyString_AsString()</tt></a> a pointer to its value. Note that Python strings may
contain null bytes so C&#8217;s <tt class="xref c c-func docutils literal">strlen()</tt> should not be used.
To test the type of an object, first make sure it isn&#8217;t NULL, and then use
<a class="reference internal" href="../c-api/string.html#PyString_Check" title="PyString_Check"><tt class="xref c c-func docutils literal">PyString_Check()</tt></a>, <a class="reference internal" href="../c-api/tuple.html#PyTuple_Check" title="PyTuple_Check"><tt class="xref c c-func docutils literal">PyTuple_Check()</tt></a>, <a class="reference internal" href="../c-api/list.html#PyList_Check" title="PyList_Check"><tt class="xref c c-func docutils literal">PyList_Check()</tt></a>, etc.
There is also a high-level API to Python objects which is provided by the
so-called &#8216;abstract&#8217; interface &#8211; read <tt class="docutils literal">Include/abstract.h</tt> for further
details. It allows interfacing with any kind of Python sequence using calls
like <a class="reference internal" href="../c-api/sequence.html#PySequence_Length" title="PySequence_Length"><tt class="xref c c-func docutils literal">PySequence_Length()</tt></a>, <a class="reference internal" href="../c-api/sequence.html#PySequence_GetItem" title="PySequence_GetItem"><tt class="xref c c-func docutils literal">PySequence_GetItem()</tt></a>, etc.) as well as
many other useful protocols.
</div>
<div class="section" id="how-do-i-use-py-buildvalue-to-create-a-tuple-of-arbitrary-length">
<h2><a class="toc-backref" href="#id9">How do I use Py_BuildValue() to create a tuple of arbitrary length?</a><a class="headerlink" href="#how-do-i-use-py-buildvalue-to-create-a-tuple-of-arbitrary-length" title="Permalink to this headline">¶</a></h2>
You can&#8217;t. Use <tt class="docutils literal">t = PyTuple_New(n)</tt> instead, and fill it with objects using
<tt class="docutils literal">PyTuple_SetItem(t, i, o)</tt> &#8211; note that this &#8220;eats&#8221; a reference count of
<tt class="docutils literal">o</tt>, so you have to <a class="reference internal" href="../c-api/refcounting.html#Py_INCREF" title="Py_INCREF"><tt class="xref c c-func docutils literal">Py_INCREF()</tt></a> it. Lists have similar functions
<tt class="docutils literal">PyList_New(n)</tt> and <tt class="docutils literal">PyList_SetItem(l, i, o)</tt>. Note that you must set all
the tuple items to some value before you pass the tuple to Python code &#8211;
<tt class="docutils literal">PyTuple_New(n)</tt> initializes them to NULL, which isn&#8217;t a valid Python value.
</div>
<div class="section" id="how-do-i-call-an-object-s-method-from-c">
<h2><a class="toc-backref" href="#id10">How do I call an object&#8217;s method from C?</a><a class="headerlink" href="#how-do-i-call-an-object-s-method-from-c" title="Permalink to this headline">¶</a></h2>
The <a class="reference internal" href="../c-api/object.html#PyObject_CallMethod" title="PyObject_CallMethod"><tt class="xref c c-func docutils literal">PyObject_CallMethod()</tt></a> function can be used to call an arbitrary
method of an object. The parameters are the object, the name of the method to
call, a format string like that used with <a class="reference internal" href="../c-api/arg.html#Py_BuildValue" title="Py_BuildValue"><tt class="xref c c-func docutils literal">Py_BuildValue()</tt></a>, and the
argument values:
<div class="highlight-c"><div class="highlight"><pre>PyObject *
PyObject_CallMethod(PyObject *object, char *method_name,
 char *arg_format, ...);
</pre></div>
</div>
This works for any object that has methods &#8211; whether built-in or user-defined.
You are responsible for eventually <a class="reference internal" href="../c-api/refcounting.html#Py_DECREF" title="Py_DECREF"><tt class="xref c c-func docutils literal">Py_DECREF()</tt></a>&#8216;ing the return value.
To call, e.g., a file object&#8217;s &#8220;seek&#8221; method with arguments 10, 0 (assuming the
file object pointer is &#8220;f&#8221;):
<div class="highlight-c"><div class="highlight"><pre>res = PyObject_CallMethod(f, &quot;seek&quot;, &quot;(ii)&quot;, 10, 0);
if (res == NULL) {
 ... an exception occurred ...
}
else {
 Py_DECREF(res);
}
</pre></div>
</div>
Note that since <a class="reference internal" href="../c-api/object.html#PyObject_CallObject" title="PyObject_CallObject"><tt class="xref c c-func docutils literal">PyObject_CallObject()</tt></a> always wants a tuple for the
argument list, to call a function without arguments, pass &#8220;()&#8221; for the format,
and to call a function with one argument, surround the argument in parentheses,
e.g. &#8220;(i)&#8221;.
</div>
<div class="section" id="how-do-i-catch-the-output-from-pyerr-print-or-anything-that-prints-to-stdout-stderr">
<h2><a class="toc-backref" href="#id11">How do I catch the output from PyErr_Print() (or anything that prints to stdout/stderr)?</a><a class="headerlink" href="#how-do-i-catch-the-output-from-pyerr-print-or-anything-that-prints-to-stdout-stderr" title="Permalink to this headline">¶</a></h2>
In Python code, define an object that supports the <tt class="docutils literal">write()</tt> method. Assign
this object to <a class="reference internal" href="../library/sys.html#sys.stdout" title="sys.stdout"><tt class="xref py py-data docutils literal">sys.stdout</tt></a> and <a class="reference internal" href="../library/sys.html#sys.stderr" title="sys.stderr"><tt class="xref py py-data docutils literal">sys.stderr</tt></a>. Call print_error, or
just allow the standard traceback mechanism to work. Then, the output will go
wherever your <tt class="docutils literal">write()</tt> method sends it.
The easiest way to do this is to use the StringIO class in the standard library.
Sample code and use for catching stdout:
<div class="highlight-c"><pre>&gt;&gt;&gt; class StdoutCatcher:
... def __init__(self):
... self.data = ''
... def write(self, stuff):
... self.data = self.data + stuff
...
&gt;&gt;&gt; import sys
&gt;&gt;&gt; sys.stdout = StdoutCatcher()
&gt;&gt;&gt; print 'foo'
&gt;&gt;&gt; print 'hello world!'
&gt;&gt;&gt; sys.stderr.write(sys.stdout.data)
foo
hello world!</pre>
</div>
</div>
<div class="section" id="how-do-i-access-a-module-written-in-python-from-c">
<h2><a class="toc-backref" href="#id12">How do I access a module written in Python from C?</a><a class="headerlink" href="#how-do-i-access-a-module-written-in-python-from-c" title="Permalink to this headline">¶</a></h2>
You can get a pointer to the module object as follows:
<div class="highlight-c"><div class="highlight"><pre>module = PyImport_ImportModule(&quot;&lt;modulename&gt;&quot;);
</pre></div>
</div>
If the module hasn&#8217;t been imported yet (i.e. it is not yet present in
<a class="reference internal" href="../library/sys.html#sys.modules" title="sys.modules"><tt class="xref py py-data docutils literal">sys.modules</tt></a>), this initializes the module; otherwise it simply returns
the value of <tt class="docutils literal">sys.modules[&quot;&lt;modulename&gt;&quot;]</tt>. Note that it doesn&#8217;t enter the
module into any namespace &#8211; it only ensures it has been initialized and is
stored in <a class="reference internal" href="../library/sys.html#sys.modules" title="sys.modules"><tt class="xref py py-data docutils literal">sys.modules</tt></a>.
You can then access the module&#8217;s attributes (i.e. any name defined in the
module) as follows:
<div class="highlight-c"><div class="highlight"><pre>attr = PyObject_GetAttrString(module, &quot;&lt;attrname&gt;&quot;);
</pre></div>
</div>
Calling <a class="reference internal" href="../c-api/object.html#PyObject_SetAttrString" title="PyObject_SetAttrString"><tt class="xref c c-func docutils literal">PyObject_SetAttrString()</tt></a> to assign to variables in the module
also works.
</div>
<div class="section" id="how-do-i-interface-to-c-objects-from-python">
<h2><a class="toc-backref" href="#id13">How do I interface to C++ objects from Python?</a><a class="headerlink" href="#how-do-i-interface-to-c-objects-from-python" title="Permalink to this headline">¶</a></h2>
Depending on your requirements, there are many approaches. To do this manually,
begin by reading <a class="reference internal" href="../extending/index.html#extending-index">the &#8220;Extending and Embedding&#8221; document</a>. Realize that for the Python run-time system, there isn&#8217;t a
whole lot of difference between C and C++ &#8211; so the strategy of building a new
Python type around a C structure (pointer) type will also work for C++ objects.
For C++ libraries, see <a class="reference internal" href="#c-wrapper-software">Writing C is hard; are there any alternatives?</a>.
</div>
<div class="section" id="i-added-a-module-using-the-setup-file-and-the-make-fails-why">
<h2><a class="toc-backref" href="#id14">I added a module using the Setup file and the make fails; why?</a><a class="headerlink" href="#i-added-a-module-using-the-setup-file-and-the-make-fails-why" title="Permalink to this headline">¶</a></h2>
Setup must end in a newline, if there is no newline there, the build process
fails. (Fixing this requires some ugly shell script hackery, and this bug is so
minor that it doesn&#8217;t seem worth the effort.)
</div>
<div class="section" id="how-do-i-debug-an-extension">
<h2><a class="toc-backref" href="#id15">How do I debug an extension?</a><a class="headerlink" href="#how-do-i-debug-an-extension" title="Permalink to this headline">¶</a></h2>
When using GDB with dynamically loaded extensions, you can&#8217;t set a breakpoint in
your extension until your extension is loaded.
In your <tt class="docutils literal">.gdbinit</tt> file (or interactively), add the command:
<div class="highlight-c"><div class="highlight"><pre>br _PyImport_LoadDynamicModule
</pre></div>
</div>
Then, when you run GDB:
<div class="highlight-c"><pre>$ gdb /local/bin/python
gdb) run myscript.py
gdb) continue # repeat until your extension is loaded
gdb) finish # so that your extension is loaded
gdb) br myfunction.c:50
gdb) continue</pre>
</div>
</div>
<div class="section" id="i-want-to-compile-a-python-module-on-my-linux-system-but-some-files-are-missing-why">
<h2><a class="toc-backref" href="#id16">I want to compile a Python module on my Linux system, but some files are missing. Why?</a><a class="headerlink" href="#i-want-to-compile-a-python-module-on-my-linux-system-but-some-files-are-missing-why" title="Permalink to this headline">¶</a></h2>
Most packaged versions of Python don&#8217;t include the
<tt class="file docutils literal">/usr/lib/python2.x/config/</tt> directory, which contains various files
required for compiling Python extensions.
For Red Hat, install the python-devel RPM to get the necessary files.
For Debian, run <tt class="docutils literal">apt-get install python-dev</tt>.
</div>
<div class="section" id="what-does-systemerror-pyimport-fixupextension-module-yourmodule-not-loaded-mean">
<h2><a class="toc-backref" href="#id17">What does &#8220;SystemError: _PyImport_FixupExtension: module yourmodule not loaded&#8221; mean?</a><a class="headerlink" href="#what-does-systemerror-pyimport-fixupextension-module-yourmodule-not-loaded-mean" title="Permalink to this headline">¶</a></h2>
This means that you have created an extension module named &#8220;yourmodule&#8221;, but
your module init function does not initialize with that name.
Every module init function will have a line similar to:
<div class="highlight-c"><div class="highlight"><pre>module = Py_InitModule(&quot;yourmodule&quot;, yourmodule_functions);
</pre></div>
</div>
If the string passed to this function is not the same name as your extension
module, the <a class="reference internal" href="../library/exceptions.html#exceptions.SystemError" title="exceptions.SystemError"><tt class="xref py py-exc docutils literal">SystemError</tt></a> exception will be raised.
</div>
<div class="section" id="how-do-i-tell-incomplete-input-from-invalid-input">
<h2><a class="toc-backref" href="#id18">How do I tell &#8220;incomplete input&#8221; from &#8220;invalid input&#8221;?</a><a class="headerlink" href="#how-do-i-tell-incomplete-input-from-invalid-input" title="Permalink to this headline">¶</a></h2>
Sometimes you want to emulate the Python interactive interpreter&#8217;s behavior,
where it gives you a continuation prompt when the input is incomplete (e.g. you
typed the start of an &#8220;if&#8221; statement or you didn&#8217;t close your parentheses or
triple string quotes), but it gives you a syntax error message immediately when
the input is invalid.
In Python you can use the <a class="reference internal" href="../library/codeop.html#module-codeop" title="codeop: Compile (possibly incomplete) Python code."><tt class="xref py py-mod docutils literal">codeop</tt></a> module, which approximates the parser&#8217;s
behavior sufficiently. IDLE uses this, for example.
The easiest way to do it in C is to call <a class="reference internal" href="../c-api/veryhigh.html#PyRun_InteractiveLoop" title="PyRun_InteractiveLoop"><tt class="xref c c-func docutils literal">PyRun_InteractiveLoop()</tt></a> (perhaps
in a separate thread) and let the Python interpreter handle the input for
you. You can also set the <tt class="xref c c-func docutils literal">PyOS_ReadlineFunctionPointer()</tt> to point at your
custom input function. See <tt class="docutils literal">Modules/readline.c</tt> and <tt class="docutils literal">Parser/myreadline.c</tt>
for more hints.
However sometimes you have to run the embedded Python interpreter in the same
thread as your rest application and you can&#8217;t allow the
<a class="reference internal" href="../c-api/veryhigh.html#PyRun_InteractiveLoop" title="PyRun_InteractiveLoop"><tt class="xref c c-func docutils literal">PyRun_InteractiveLoop()</tt></a> to stop while waiting for user input. The one
solution then is to call <tt class="xref c c-func docutils literal">PyParser_ParseString()</tt> and test for <tt class="docutils literal">e.error</tt>
equal to <tt class="docutils literal">E_EOF</tt>, which means the input is incomplete). Here&#8217;s a sample code
fragment, untested, inspired by code from Alex Farber:
<div class="highlight-c"><div class="highlight"><pre>#include &lt;Python.h&gt;
#include &lt;node.h&gt;
#include &lt;errcode.h&gt;
#include &lt;grammar.h&gt;
#include &lt;parsetok.h&gt;
#include &lt;compile.h&gt;

int testcomplete(char *code)
 /* code should end in \n */
 /* return -1 for error, 0 for incomplete, 1 for complete */
{
 node *n;
 perrdetail e;

n = PyParser_ParseString(code, &amp;_PyParser_Grammar,
 Py_file_input, &amp;e);
 if (n == NULL) {
 if (e.error == E_EOF)
 return 0;
 return -1;
 }

PyNode_Free(n);
 return 1;
}
</pre></div>
</div>
Another solution is trying to compile the received string with
<a class="reference internal" href="../c-api/veryhigh.html#Py_CompileString" title="Py_CompileString"><tt class="xref c c-func docutils literal">Py_CompileString()</tt></a>. If it compiles without errors, try to execute the
returned code object by calling <a class="reference internal" href="../c-api/veryhigh.html#PyEval_EvalCode" title="PyEval_EvalCode"><tt class="xref c c-func docutils literal">PyEval_EvalCode()</tt></a>. Otherwise save the
input for later. If the compilation fails, find out if it&#8217;s an error or just
more input is required - by extracting the message string from the exception
tuple and comparing it to the string &#8220;unexpected EOF while parsing&#8221;. Here is a
complete example using the GNU readline library (you may want to ignore
SIGINT while calling readline()):
<div class="highlight-c"><div class="highlight"><pre>#include &lt;stdio.h&gt;
#include &lt;readline.h&gt;

#include &lt;Python.h&gt;
#include &lt;object.h&gt;
#include &lt;compile.h&gt;
#include &lt;eval.h&gt;

int main (int argc, char* argv[])
{
 int i, j, done = 0; /* lengths of line, code */
 char ps1[] = &quot;&gt;&gt;&gt; &quot;;
 char ps2[] = &quot;... &quot;;
 char *prompt = ps1;
 char *msg, *line, *code = NULL;
 PyObject *src, *glb, *loc;
 PyObject *exc, *val, *trb, *obj, *dum;

Py_Initialize ();
 loc = PyDict_New ();
 glb = PyDict_New ();
 PyDict_SetItemString (glb, &quot;__builtins__&quot;, PyEval_GetBuiltins ());

while (!done)
 {
 line = readline (prompt);

if (NULL == line) /* CTRL-D pressed */
 {
 done = 1;
 }
 else
 {
 i = strlen (line);

if (i &gt; 0)
 add_history (line); /* save non-empty lines */

if (NULL == code) /* nothing in code yet */
 j = 0;
 else
 j = strlen (code);

code = realloc (code, i + j + 2);
 if (NULL == code) /* out of memory */
 exit (1);

if (0 == j) /* code was empty, so */
 code[0] = &#39;\0&#39;; /* keep strncat happy */

strncat (code, line, i); /* append line to code */
 code[i + j] = &#39;\n&#39;; /* append &#39;\n&#39; to code */
 code[i + j + 1] = &#39;\0&#39;;

src = Py_CompileString (code, &quot;&lt;stdin&gt;&quot;, Py_single_input);

if (NULL != src) /* compiled just fine - */
 {
 if (ps1 == prompt || /* &quot;&gt;&gt;&gt; &quot; or */
 &#39;\n&#39; == code[i + j - 1]) /* &quot;... &quot; and double &#39;\n&#39; */
 { /* so execute it */
 dum = PyEval_EvalCode ((PyCodeObject *)src, glb, loc);
 Py_XDECREF (dum);
 Py_XDECREF (src);
 free (code);
 code = NULL;
 if (PyErr_Occurred ())
 PyErr_Print ();
 prompt = ps1;
 }
 } /* syntax error or E_EOF? */
 else if (PyErr_ExceptionMatches (PyExc_SyntaxError))
 {
 PyErr_Fetch (&amp;exc, &amp;val, &amp;trb); /* clears exception! */

if (PyArg_ParseTuple (val, &quot;sO&quot;, &amp;msg, &amp;obj) &amp;&amp;
 !strcmp (msg, &quot;unexpected EOF while parsing&quot;)) /* E_EOF */
 {
 Py_XDECREF (exc);
 Py_XDECREF (val);
 Py_XDECREF (trb);
 prompt = ps2;
 }
 else /* some other syntax error */
 {
 PyErr_Restore (exc, val, trb);
 PyErr_Print ();
 free (code);
 code = NULL;
 prompt = ps1;
 }
 }
 else /* some non-syntax error */
 {
 PyErr_Print ();
 free (code);
 code = NULL;
 prompt = ps1;
 }

free (line);
 }
 }

Py_XDECREF(glb);
 Py_XDECREF(loc);
 Py_Finalize();
 exit(0);
}
</pre></div>
</div>
</div>
<div class="section" id="how-do-i-find-undefined-g-symbols-builtin-new-or-pure-virtual">
<h2><a class="toc-backref" href="#id19">How do I find undefined g++ symbols __builtin_new or __pure_virtual?</a><a class="headerlink" href="#how-do-i-find-undefined-g-symbols-builtin-new-or-pure-virtual" title="Permalink to this headline">¶</a></h2>
To dynamically load g++ extension modules, you must recompile Python, relink it
using g++ (change LINKCC in the Python Modules Makefile), and link your
extension module using g++ (e.g., <tt class="docutils literal">g++ -shared -o mymodule.so mymodule.o</tt>).
</div>
<div class="section" id="can-i-create-an-object-class-with-some-methods-implemented-in-c-and-others-in-python-e-g-through-inheritance">
<h2><a class="toc-backref" href="#id20">Can I create an object class with some methods implemented in C and others in Python (e.g. through inheritance)?</a><a class="headerlink" href="#can-i-create-an-object-class-with-some-methods-implemented-in-c-and-others-in-python-e-g-through-inheritance" title="Permalink to this headline">¶</a></h2>
In Python 2.2, you can inherit from built-in classes such as <a class="reference internal" href="../library/functions.html#int" title="int"><tt class="xref py py-class docutils literal">int</tt></a>,
<a class="reference internal" href="../library/functions.html#list" title="list"><tt class="xref py py-class docutils literal">list</tt></a>, <a class="reference internal" href="../library/stdtypes.html#dict" title="dict"><tt class="xref py py-class docutils literal">dict</tt></a>, etc.
The Boost Python Library (BPL, <a class="reference external" href="http://www.boost.org/libs/python/doc/index.html">http://www.boost.org/libs/python/doc/index.html</a>)
provides a way of doing this from C++ (i.e. you can inherit from an extension
class written in C++ using the BPL).
</div>
<div class="section" id="when-importing-module-x-why-do-i-get-undefined-symbol-pyunicodeucs2">
<h2><a class="toc-backref" href="#id21">When importing module X, why do I get &#8220;undefined symbol: PyUnicodeUCS2*&#8221;?</a><a class="headerlink" href="#when-importing-module-x-why-do-i-get-undefined-symbol-pyunicodeucs2" title="Permalink to this headline">¶</a></h2>
You are using a version of Python that uses a 4-byte representation for Unicode
characters, but some C extension module you are importing was compiled using a
Python that uses a 2-byte representation for Unicode characters (the default).
If instead the name of the undefined symbol starts with <tt class="docutils literal">PyUnicodeUCS4</tt>, the
problem is the reverse: Python was built using 2-byte Unicode characters, and
the extension module was compiled using a Python with 4-byte Unicode characters.
This can easily occur when using pre-built extension packages. RedHat Linux
7.x, in particular, provided a &#8220;python2&#8221; binary that is compiled with 4-byte
Unicode. This only causes the link failure if the extension uses any of the
<tt class="docutils literal">PyUnicode_*()</tt> functions. It is also a problem if an extension uses any of
the Unicode-related format specifiers for <a class="reference internal" href="../c-api/arg.html#Py_BuildValue" title="Py_BuildValue"><tt class="xref c c-func docutils literal">Py_BuildValue()</tt></a> (or similar) or
parameter specifications for <a class="reference internal" href="../c-api/arg.html#PyArg_ParseTuple" title="PyArg_ParseTuple"><tt class="xref c c-func docutils literal">PyArg_ParseTuple()</tt></a>.
You can check the size of the Unicode character a Python interpreter is using by
checking the value of sys.maxunicode:
<div class="highlight-c"><pre>&gt;&gt;&gt; import sys
&gt;&gt;&gt; if sys.maxunicode &gt; 65535:
... print 'UCS4 build'
... else:
... print 'UCS2 build'</pre>
</div>
The only way to solve this problem is to use extension modules compiled with a
Python binary built using the same size for Unicode characters.
</div>
</div>

</div>
 </div>
 </div>
 <div class="sphinxsidebar">
 <div class="sphinxsidebarwrapper">
 <h3><a href="../contents.html">Table Of Contents</a></h3>
 <ul>
<li><a class="reference internal" href="#">Extending/Embedding FAQ</a><ul>
<li><a class="reference internal" href="#can-i-create-my-own-functions-in-c">Can I create my own functions in C?</a></li>
<li><a class="reference internal" href="#id1">Can I create my own functions in C++?</a></li>
<li><a class="reference internal" href="#writing-c-is-hard-are-there-any-alternatives">Writing C is hard; are there any alternatives?</a></li>
<li><a class="reference internal" href="#how-can-i-execute-arbitrary-python-statements-from-c">How can I execute arbitrary Python statements from C?</a></li>
<li><a class="reference internal" href="#how-can-i-evaluate-an-arbitrary-python-expression-from-c">How can I evaluate an arbitrary Python expression from C?</a></li>
<li><a class="reference internal" href="#how-do-i-extract-c-values-from-a-python-object">How do I extract C values from a Python object?</a></li>
<li><a class="reference internal" href="#how-do-i-use-py-buildvalue-to-create-a-tuple-of-arbitrary-length">How do I use Py_BuildValue() to create a tuple of arbitrary length?</a></li>
<li><a class="reference internal" href="#how-do-i-call-an-object-s-method-from-c">How do I call an object&#8217;s method from C?</a></li>
<li><a class="reference internal" href="#how-do-i-catch-the-output-from-pyerr-print-or-anything-that-prints-to-stdout-stderr">How do I catch the output from PyErr_Print() (or anything that prints to stdout/stderr)?</a></li>
<li><a class="reference internal" href="#how-do-i-access-a-module-written-in-python-from-c">How do I access a module written in Python from C?</a></li>
<li><a class="reference internal" href="#how-do-i-interface-to-c-objects-from-python">How do I interface to C++ objects from Python?</a></li>
<li><a class="reference internal" href="#i-added-a-module-using-the-setup-file-and-the-make-fails-why">I added a module using the Setup file and the make fails; why?</a></li>
<li><a class="reference internal" href="#how-do-i-debug-an-extension">How do I debug an extension?</a></li>
<li><a class="reference internal" href="#i-want-to-compile-a-python-module-on-my-linux-system-but-some-files-are-missing-why">I want to compile a Python module on my Linux system, but some files are missing. Why?</a></li>
<li><a class="reference internal" href="#what-does-systemerror-pyimport-fixupextension-module-yourmodule-not-loaded-mean">What does &#8220;SystemError: _PyImport_FixupExtension: module yourmodule not loaded&#8221; mean?</a></li>
<li><a class="reference internal" href="#how-do-i-tell-incomplete-input-from-invalid-input">How do I tell &#8220;incomplete input&#8221; from &#8220;invalid input&#8221;?</a></li>
<li><a class="reference internal" href="#how-do-i-find-undefined-g-symbols-builtin-new-or-pure-virtual">How do I find undefined g++ symbols __builtin_new or __pure_virtual?</a></li>
<li><a class="reference internal" href="#can-i-create-an-object-class-with-some-methods-implemented-in-c-and-others-in-python-e-g-through-inheritance">Can I create an object class with some methods implemented in C and others in Python (e.g. through inheritance)?</a></li>
<li><a class="reference internal" href="#when-importing-module-x-why-do-i-get-undefined-symbol-pyunicodeucs2">When importing module X, why do I get &#8220;undefined symbol: PyUnicodeUCS2*&#8221;?</a></li>
</ul>
</li>
</ul>

<h4>Previous topic</h4>
 <a href="library.html"
 title="previous chapter">Library and Extension FAQ</a>
 <h4>Next topic</h4>
 <a href="windows.html"
 title="next chapter">Python on Windows FAQ</a>
<h3>This Page</h3>
<ul class="this-page-menu">
 <li><a href="../bugs.html">Report a Bug</a></li>
 <li><a href="../_sources/faq/extending.txt"
 rel="nofollow">Show Source</a></li>
</ul>

<div id="searchbox" style="display: none">
 <h3>Quick search</h3>
 <form class="search" action="../search.html" method="get">
 <input type="text" name="q" />
 <input type="submit" value="Go" />
 <input type="hidden" name="check_keywords" value="yes" />
 <input type="hidden" name="area" value="default" />
 </form>
 
 Enter search terms or a module, class or function name.
 
</div>
<script type="text/javascript">$('#searchbox').show(0);</script>
 </div>
 </div>
 <div class="clearer"></div>
 </div>
 <div class="related">
 <h3>Navigation</h3>
 <ul>
 <li class="right" style="margin-right: 10px">
 <a href="../genindex.html" title="General Index"
 >index</a></li>
 <li class="right" >
 <a href="../py-modindex.html" title="Python Module Index"
 >modules</a> |</li>
 <li class="right" >
 <a href="windows.html" title="Python on Windows FAQ"
 >next</a> |</li>
 <li class="right" >
 <a href="library.html" title="Library and Extension FAQ"
 >previous</a> |</li>
 <li><img src="../_static/py.png" alt=""
 style="vertical-align: middle; margin-top: -1px"/></li>
 <li><a href="http://www.python.org/">Python</a> &raquo;</li>
 <li>
 <a href="../index.html">Python 2.7.5 documentation</a> &raquo;
 </li>

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

</body>
</html>