22

✏️ Editing: clinic.html

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

<html xmlns="http://www.w3.org/1999/xhtml">
 <head>
 <meta http-equiv="X-UA-Compatible" content="IE=Edge" />
 <meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
 <title>Argument Clinic How-To &#8212; Python 3.6.7 documentation</title>
 <link rel="stylesheet" href="../_static/pydoctheme.css" type="text/css" />
 <link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
 <script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
 <script type="text/javascript" src="../_static/jquery.js"></script>
 <script type="text/javascript" src="../_static/underscore.js"></script>
 <script type="text/javascript" src="../_static/doctools.js"></script>
 <script type="text/javascript" src="../_static/sidebar.js"></script>
 <link rel="search" type="application/opensearchdescription+xml"
 title="Search within Python 3.6.7 documentation"
 href="../_static/opensearch.xml"/>
 <link rel="author" title="About these documents" href="../about.html" />
 <link rel="index" title="Index" href="../genindex.html" />
 <link rel="search" title="Search" href="../search.html" />
 <link rel="copyright" title="Copyright" href="../copyright.html" />
 <link rel="next" title="Instrumenting CPython with DTrace and SystemTap" href="instrumentation.html" />
 <link rel="prev" title="An introduction to the ipaddress module" href="ipaddress.html" />
 <link rel="shortcut icon" type="image/png" href="../_static/py.png" />
 <link rel="canonical" href="https://docs.python.org/3/howto/clinic.html" />
 
 <script type="text/javascript" src="../_static/copybutton.js"></script>

</head><body> 
 <div class="related" role="navigation" aria-label="related navigation">
 <h3>Navigation</h3>
 <ul>
 <li class="right" style="margin-right: 10px">
 <a href="../genindex.html" title="General Index"
 accesskey="I">index</a></li>
 <li class="right" >
 <a href="../py-modindex.html" title="Python Module Index"
 >modules</a> |</li>
 <li class="right" >
 <a href="instrumentation.html" title="Instrumenting CPython with DTrace and SystemTap"
 accesskey="N">next</a> |</li>
 <li class="right" >
 <a href="ipaddress.html" title="An introduction to the ipaddress module"
 accesskey="P">previous</a> |</li>
 <li><img src="../_static/py.png" alt=""
 style="vertical-align: middle; margin-top: -1px"/></li>
 <li><a href="https://www.python.org/">Python</a> &#187;</li>
 <li>
 <a href="../index.html">3.6.7 Documentation</a> &#187;
 </li>

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

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

</ul>
 </div>

<div class="document">
 <div class="documentwrapper">
 <div class="bodywrapper">
 <div class="body" role="main">
 
 <div class="section" id="argument-clinic-how-to">
<h1>Argument Clinic How-To<a class="headerlink" href="#argument-clinic-how-to" title="Permalink to this headline">¶</a></h1>
<table class="docutils field-list" frame="void" rules="none">
<col class="field-name" />
<col class="field-body" />
<tbody valign="top">
<tr class="field-odd field"><th class="field-name">author:</th><td class="field-body">Larry Hastings</td>
</tr>
</tbody>
</table>
<div class="topic">
Abstract
Argument Clinic is a preprocessor for CPython C files.
Its purpose is to automate all the boilerplate involved
with writing argument parsing code for “builtins”.
This document shows you how to convert your first C
function to work with Argument Clinic, and then introduces
some advanced topics on Argument Clinic usage.
Currently Argument Clinic is considered internal-only
for CPython. Its use is not supported for files outside
CPython, and no guarantees are made regarding backwards
compatibility for future versions. In other words: if you
maintain an external C extension for CPython, you’re welcome
to experiment with Argument Clinic in your own code. But the
version of Argument Clinic that ships with the next version
of CPython could be totally incompatible and break all your code.
</div>
<div class="section" id="the-goals-of-argument-clinic">
<h2>The Goals Of Argument Clinic<a class="headerlink" href="#the-goals-of-argument-clinic" title="Permalink to this headline">¶</a></h2>
Argument Clinic’s primary goal
is to take over responsibility for all argument parsing code
inside CPython. This means that, when you convert a function
to work with Argument Clinic, that function should no longer
do any of its own argument parsing—the code generated by
Argument Clinic should be a “black box” to you, where CPython
calls in at the top, and your code gets called at the bottom,
with <code class="docutils literal notranslate">PyObject *args</code> (and maybe <code class="docutils literal notranslate">PyObject *kwargs</code>)
magically converted into the C variables and types you need.
In order for Argument Clinic to accomplish its primary goal,
it must be easy to use. Currently, working with CPython’s
argument parsing library is a chore, requiring maintaining
redundant information in a surprising number of places.
When you use Argument Clinic, you don’t have to repeat yourself.
Obviously, no one would want to use Argument Clinic unless
it’s solving their problem—and without creating new problems of
its own.
So it’s paramount that Argument Clinic generate correct code.
It’d be nice if the code was faster, too, but at the very least
it should not introduce a major speed regression. (Eventually Argument
Clinic should make a major speedup possible—we could
rewrite its code generator to produce tailor-made argument
parsing code, rather than calling the general-purpose CPython
argument parsing library. That would make for the fastest
argument parsing possible!)
Additionally, Argument Clinic must be flexible enough to
work with any approach to argument parsing. Python has
some functions with some very strange parsing behaviors;
Argument Clinic’s goal is to support all of them.
Finally, the original motivation for Argument Clinic was
to provide introspection “signatures” for CPython builtins.
It used to be, the introspection query functions would throw
an exception if you passed in a builtin. With Argument
Clinic, that’s a thing of the past!
One idea you should keep in mind, as you work with
Argument Clinic: the more information you give it, the
better job it’ll be able to do.
Argument Clinic is admittedly relatively simple right
now. But as it evolves it will get more sophisticated,
and it should be able to do many interesting and smart
things with all the information you give it.
</div>
<div class="section" id="basic-concepts-and-usage">
<h2>Basic Concepts And Usage<a class="headerlink" href="#basic-concepts-and-usage" title="Permalink to this headline">¶</a></h2>
Argument Clinic ships with CPython; you’ll find it in <code class="docutils literal notranslate">Tools/clinic/clinic.py</code>.
If you run that script, specifying a C file as an argument:
<div class="highlight-shell-session notranslate"><div class="highlight"><pre>$ python3 Tools/clinic/clinic.py foo.c
</pre></div>
</div>
Argument Clinic will scan over the file looking for lines that
look exactly like this:
<div class="highlight-none notranslate"><div class="highlight"><pre>/*[clinic input]
</pre></div>
</div>
When it finds one, it reads everything up to a line that looks
exactly like this:
<div class="highlight-none notranslate"><div class="highlight"><pre>[clinic start generated code]*/
</pre></div>
</div>
Everything in between these two lines is input for Argument Clinic.
All of these lines, including the beginning and ending comment
lines, are collectively called an Argument Clinic “block”.
When Argument Clinic parses one of these blocks, it
generates output. This output is rewritten into the C file
immediately after the block, followed by a comment containing a checksum.
The Argument Clinic block now looks like this:
<div class="highlight-none notranslate"><div class="highlight"><pre>/*[clinic input]
... clinic input goes here ...
[clinic start generated code]*/
... clinic output goes here ...
/*[clinic end generated code: checksum=...]*/
</pre></div>
</div>
If you run Argument Clinic on the same file a second time, Argument Clinic
will discard the old output and write out the new output with a fresh checksum
line. However, if the input hasn’t changed, the output won’t change either.
You should never modify the output portion of an Argument Clinic block. Instead,
change the input until it produces the output you want. (That’s the purpose of the
checksum—to detect if someone changed the output, as these edits would be lost
the next time Argument Clinic writes out fresh output.)
For the sake of clarity, here’s the terminology we’ll use with Argument Clinic:
<ul class="simple">
<li>The first line of the comment (<code class="docutils literal notranslate">/*[clinic input]</code>) is the start line.</li>
<li>The last line of the initial comment (<code class="docutils literal notranslate">[clinic start generated code]*/</code>) is the end line.</li>
<li>The last line (<code class="docutils literal notranslate">/*[clinic end generated code: checksum=...]*/</code>) is the checksum line.</li>
<li>In between the start line and the end line is the input.</li>
<li>In between the end line and the checksum line is the output.</li>
<li>All the text collectively, from the start line to the checksum line inclusively,
is the block. (A block that hasn’t been successfully processed by Argument
Clinic yet doesn’t have output or a checksum line, but it’s still considered
a block.)</li>
</ul>
</div>
<div class="section" id="converting-your-first-function">
<h2>Converting Your First Function<a class="headerlink" href="#converting-your-first-function" title="Permalink to this headline">¶</a></h2>
The best way to get a sense of how Argument Clinic works is to
convert a function to work with it. Here, then, are the bare
minimum steps you’d need to follow to convert a function to
work with Argument Clinic. Note that for code you plan to
check in to CPython, you really should take the conversion farther,
using some of the advanced concepts you’ll see later on in
the document (like “return converters” and “self converters”).
But we’ll keep it simple for this walkthrough so you can learn.
Let’s dive in!
<ol class="arabic" start="0">
<li>Make sure you’re working with a freshly updated checkout
of the CPython trunk.
</li>
<li>Find a Python builtin that calls either <a class="reference internal" href="../c-api/arg.html#c.PyArg_ParseTuple" title="PyArg_ParseTuple"><code class="xref c c-func docutils literal notranslate">PyArg_ParseTuple()</code></a>
or <a class="reference internal" href="../c-api/arg.html#c.PyArg_ParseTupleAndKeywords" title="PyArg_ParseTupleAndKeywords"><code class="xref c c-func docutils literal notranslate">PyArg_ParseTupleAndKeywords()</code></a>, and hasn’t been converted
to work with Argument Clinic yet.
For my example I’m using <code class="docutils literal notranslate">_pickle.Pickler.dump()</code>.
</li>
<li>If the call to the <code class="docutils literal notranslate">PyArg_Parse</code> function uses any of the
following format units:
<div class="highlight-none notranslate"><div class="highlight"><pre>O&amp;
O!
es
es#
et
et#
</pre></div>
</div>
or if it has multiple calls to <a class="reference internal" href="../c-api/arg.html#c.PyArg_ParseTuple" title="PyArg_ParseTuple"><code class="xref c c-func docutils literal notranslate">PyArg_ParseTuple()</code></a>,
you should choose a different function. Argument Clinic does
support all of these scenarios. But these are advanced
topics—let’s do something simpler for your first function.
Also, if the function has multiple calls to <a class="reference internal" href="../c-api/arg.html#c.PyArg_ParseTuple" title="PyArg_ParseTuple"><code class="xref c c-func docutils literal notranslate">PyArg_ParseTuple()</code></a>
or <a class="reference internal" href="../c-api/arg.html#c.PyArg_ParseTupleAndKeywords" title="PyArg_ParseTupleAndKeywords"><code class="xref c c-func docutils literal notranslate">PyArg_ParseTupleAndKeywords()</code></a> where it supports different
types for the same argument, or if the function uses something besides
PyArg_Parse functions to parse its arguments, it probably
isn’t suitable for conversion to Argument Clinic. Argument Clinic
doesn’t support generic functions or polymorphic parameters.
</li>
<li>Add the following boilerplate above the function, creating our block:
<div class="highlight-c notranslate"><div class="highlight"><pre>/*[clinic input]
[clinic start generated code]*/
</pre></div>
</div>
</li>
<li>Cut the docstring and paste it in between the <code class="docutils literal notranslate">[clinic]</code> lines,
removing all the junk that makes it a properly quoted C string.
When you’re done you should have just the text, based at the left
margin, with no line wider than 80 characters.
(Argument Clinic will preserve indents inside the docstring.)
If the old docstring had a first line that looked like a function
signature, throw that line away. (The docstring doesn’t need it
anymore—when you use <code class="docutils literal notranslate">help()</code> on your builtin in the future,
the first line will be built automatically based on the function’s
signature.)
Sample:
<div class="highlight-c notranslate"><div class="highlight"><pre>/*[clinic input]
Write a pickled representation of obj to the open file.
[clinic start generated code]*/
</pre></div>
</div>
</li>
<li>If your docstring doesn’t have a “summary” line, Argument Clinic will
complain. So let’s make sure it has one. The “summary” line should
be a paragraph consisting of a single 80-column line
at the beginning of the docstring.
(Our example docstring consists solely of a summary line, so the sample
code doesn’t have to change for this step.)
</li>
<li>Above the docstring, enter the name of the function, followed
by a blank line. This should be the Python name of the function,
and should be the full dotted path
to the function—it should start with the name of the module,
include any sub-modules, and if the function is a method on
a class it should include the class name too.
Sample:
<div class="highlight-c notranslate"><div class="highlight"><pre>/*[clinic input]
_pickle.Pickler.dump

Write a pickled representation of obj to the open file.
[clinic start generated code]*/
</pre></div>
</div>
</li>
<li>If this is the first time that module or class has been used with Argument
Clinic in this C file,
you must declare the module and/or class. Proper Argument Clinic hygiene
prefers declaring these in a separate block somewhere near the
top of the C file, in the same way that include files and statics go at
the top. (In our sample code we’ll just show the two blocks next to
each other.)
The name of the class and module should be the same as the one
seen by Python. Check the name defined in the <a class="reference internal" href="../c-api/module.html#c.PyModuleDef" title="PyModuleDef"><code class="xref c c-type docutils literal notranslate">PyModuleDef</code></a>
or <a class="reference internal" href="../c-api/type.html#c.PyTypeObject" title="PyTypeObject"><code class="xref c c-type docutils literal notranslate">PyTypeObject</code></a> as appropriate.
When you declare a class, you must also specify two aspects of its type
in C: the type declaration you’d use for a pointer to an instance of
this class, and a pointer to the <a class="reference internal" href="../c-api/type.html#c.PyTypeObject" title="PyTypeObject"><code class="xref c c-type docutils literal notranslate">PyTypeObject</code></a> for this class.
Sample:
<div class="highlight-c notranslate"><div class="highlight"><pre>/*[clinic input]
module _pickle
class _pickle.Pickler &quot;PicklerObject *&quot; &quot;&amp;Pickler_Type&quot;
[clinic start generated code]*/

/*[clinic input]
_pickle.Pickler.dump

Write a pickled representation of obj to the open file.
[clinic start generated code]*/
</pre></div>
</div>
</li>
<li>Declare each of the parameters to the function. Each parameter
should get its own line. All the parameter lines should be
indented from the function name and the docstring.
The general form of these parameter lines is as follows:
<div class="highlight-none notranslate"><div class="highlight"><pre>name_of_parameter: converter
</pre></div>
</div>
If the parameter has a default value, add that after the
converter:
<div class="highlight-none notranslate"><div class="highlight"><pre>name_of_parameter: converter = default_value
</pre></div>
</div>
Argument Clinic’s support for “default values” is quite sophisticated;
please see <a class="reference internal" href="#default-values">the section below on default values</a>
for more information.
Add a blank line below the parameters.
What’s a “converter”? It establishes both the type
of the variable used in C, and the method to convert the Python
value into a C value at runtime.
For now you’re going to use what’s called a “legacy converter”—a
convenience syntax intended to make porting old code into Argument
Clinic easier.
For each parameter, copy the “format unit” for that
parameter from the <code class="docutils literal notranslate">PyArg_Parse()</code> format argument and
specify that as its converter, as a quoted
string. (“format unit” is the formal name for the one-to-three
character substring of the <code class="docutils literal notranslate">format</code> parameter that tells
the argument parsing function what the type of the variable
is and how to convert it. For more on format units please
see <a class="reference internal" href="../c-api/arg.html#arg-parsing">Parsing arguments and building values</a>.)
For multicharacter format units like <code class="docutils literal notranslate">z#</code>, use the
entire two-or-three character string.
Sample:
<div class="highlight-c notranslate"><div class="highlight"><pre> /*[clinic input]
 module _pickle
 class _pickle.Pickler &quot;PicklerObject *&quot; &quot;&amp;Pickler_Type&quot;
 [clinic start generated code]*/

/*[clinic input]
 _pickle.Pickler.dump

obj: &#39;O&#39;

Write a pickled representation of obj to the open file.
[clinic start generated code]*/
</pre></div>
</div>
</li>
<li>If your function has <code class="docutils literal notranslate">|</code> in the format string, meaning some
parameters have default values, you can ignore it. Argument
Clinic infers which parameters are optional based on whether
or not they have default values.
If your function has <code class="docutils literal notranslate">$</code> in the format string, meaning it
takes keyword-only arguments, specify <code class="docutils literal notranslate">*</code> on a line by
itself before the first keyword-only argument, indented the
same as the parameter lines.
(<code class="docutils literal notranslate">_pickle.Pickler.dump</code> has neither, so our sample is unchanged.)
</li>
<li>If the existing C function calls <a class="reference internal" href="../c-api/arg.html#c.PyArg_ParseTuple" title="PyArg_ParseTuple"><code class="xref c c-func docutils literal notranslate">PyArg_ParseTuple()</code></a>
(as opposed to <a class="reference internal" href="../c-api/arg.html#c.PyArg_ParseTupleAndKeywords" title="PyArg_ParseTupleAndKeywords"><code class="xref c c-func docutils literal notranslate">PyArg_ParseTupleAndKeywords()</code></a>), then all its
arguments are positional-only.
To mark all parameters as positional-only in Argument Clinic,
add a <code class="docutils literal notranslate">/</code> on a line by itself after the last parameter,
indented the same as the parameter lines.
Currently this is all-or-nothing; either all parameters are
positional-only, or none of them are. (In the future Argument
Clinic may relax this restriction.)
Sample:
<div class="highlight-c notranslate"><div class="highlight"><pre>/*[clinic input]
module _pickle
class _pickle.Pickler &quot;PicklerObject *&quot; &quot;&amp;Pickler_Type&quot;
[clinic start generated code]*/

/*[clinic input]
_pickle.Pickler.dump

obj: &#39;O&#39;
 /

Write a pickled representation of obj to the open file.
[clinic start generated code]*/
</pre></div>
</div>
</li>
<li>It’s helpful to write a per-parameter docstring for each parameter.
But per-parameter docstrings are optional; you can skip this step
if you prefer.
Here’s how to add a per-parameter docstring. The first line
of the per-parameter docstring must be indented further than the
parameter definition. The left margin of this first line establishes
the left margin for the whole per-parameter docstring; all the text
you write will be outdented by this amount. You can write as much
text as you like, across multiple lines if you wish.
Sample:
<div class="highlight-c notranslate"><div class="highlight"><pre>/*[clinic input]
module _pickle
class _pickle.Pickler &quot;PicklerObject *&quot; &quot;&amp;Pickler_Type&quot;
[clinic start generated code]*/

/*[clinic input]
_pickle.Pickler.dump

obj: &#39;O&#39;
 The object to be pickled.
 /

Write a pickled representation of obj to the open file.
[clinic start generated code]*/
</pre></div>
</div>
</li>
<li>Save and close the file, then run <code class="docutils literal notranslate">Tools/clinic/clinic.py</code> on
it. With luck everything worked—your block now has output, and
a <code class="docutils literal notranslate">.c.h</code> file has been generated! Reopen the file in your
text editor to see:
<div class="highlight-c notranslate"><div class="highlight"><pre>/*[clinic input]
_pickle.Pickler.dump

obj: &#39;O&#39;
 The object to be pickled.
 /

Write a pickled representation of obj to the open file.
[clinic start generated code]*/

static PyObject *
_pickle_Pickler_dump(PicklerObject *self, PyObject *obj)
/*[clinic end generated code: output=87ecad1261e02ac7 input=552eb1c0f52260d9]*/
</pre></div>
</div>
Obviously, if Argument Clinic didn’t produce any output, it’s because
it found an error in your input. Keep fixing your errors and retrying
until Argument Clinic processes your file without complaint.
For readability, most of the glue code has been generated to a <code class="docutils literal notranslate">.c.h</code>
file. You’ll need to include that in your original <code class="docutils literal notranslate">.c</code> file,
typically right after the clinic module block:
<div class="highlight-c notranslate"><div class="highlight"><pre>#include &quot;clinic/_pickle.c.h&quot;
</pre></div>
</div>
</li>
<li>Double-check that the argument-parsing code Argument Clinic generated
looks basically the same as the existing code.
First, ensure both places use the same argument-parsing function.
The existing code must call either
<a class="reference internal" href="../c-api/arg.html#c.PyArg_ParseTuple" title="PyArg_ParseTuple"><code class="xref c c-func docutils literal notranslate">PyArg_ParseTuple()</code></a> or <a class="reference internal" href="../c-api/arg.html#c.PyArg_ParseTupleAndKeywords" title="PyArg_ParseTupleAndKeywords"><code class="xref c c-func docutils literal notranslate">PyArg_ParseTupleAndKeywords()</code></a>;
ensure that the code generated by Argument Clinic calls the
exact same function.
Second, the format string passed in to <a class="reference internal" href="../c-api/arg.html#c.PyArg_ParseTuple" title="PyArg_ParseTuple"><code class="xref c c-func docutils literal notranslate">PyArg_ParseTuple()</code></a> or
<a class="reference internal" href="../c-api/arg.html#c.PyArg_ParseTupleAndKeywords" title="PyArg_ParseTupleAndKeywords"><code class="xref c c-func docutils literal notranslate">PyArg_ParseTupleAndKeywords()</code></a> should be exactly the same
as the hand-written one in the existing function, up to the colon
or semi-colon.
(Argument Clinic always generates its format strings
with a <code class="docutils literal notranslate">:</code> followed by the name of the function. If the
existing code’s format string ends with <code class="docutils literal notranslate">;</code>, to provide
usage help, this change is harmless—don’t worry about it.)
Third, for parameters whose format units require two arguments
(like a length variable, or an encoding string, or a pointer
to a conversion function), ensure that the second argument is
exactly the same between the two invocations.
Fourth, inside the output portion of the block you’ll find a preprocessor
macro defining the appropriate static <a class="reference internal" href="../c-api/structures.html#c.PyMethodDef" title="PyMethodDef"><code class="xref c c-type docutils literal notranslate">PyMethodDef</code></a> structure for
this builtin:
<div class="highlight-c notranslate"><div class="highlight"><pre>#define __PICKLE_PICKLER_DUMP_METHODDEF \
{&quot;dump&quot;, (PyCFunction)__pickle_Pickler_dump, METH_O, __pickle_Pickler_dump__doc__},
</pre></div>
</div>
This static structure should be exactly the same as the existing static
<a class="reference internal" href="../c-api/structures.html#c.PyMethodDef" title="PyMethodDef"><code class="xref c c-type docutils literal notranslate">PyMethodDef</code></a> structure for this builtin.
If any of these items differ in any way,
adjust your Argument Clinic function specification and rerun
<code class="docutils literal notranslate">Tools/clinic/clinic.py</code> until they are the same.
</li>
<li>Notice that the last line of its output is the declaration
of your “impl” function. This is where the builtin’s implementation goes.
Delete the existing prototype of the function you’re modifying, but leave
the opening curly brace. Now delete its argument parsing code and the
declarations of all the variables it dumps the arguments into.
Notice how the Python arguments are now arguments to this impl function;
if the implementation used different names for these variables, fix it.
Let’s reiterate, just because it’s kind of weird. Your code should now
look like this:
<div class="highlight-c notranslate"><div class="highlight"><pre>static return_type
your_function_impl(...)
/*[clinic end generated code: checksum=...]*/
{
...
</pre></div>
</div>
Argument Clinic generated the checksum line and the function prototype just
above it. You should write the opening (and closing) curly braces for the
function, and the implementation inside.
Sample:
<div class="highlight-c notranslate"><div class="highlight"><pre>/*[clinic input]
module _pickle
class _pickle.Pickler &quot;PicklerObject *&quot; &quot;&amp;Pickler_Type&quot;
[clinic start generated code]*/
/*[clinic end generated code: checksum=da39a3ee5e6b4b0d3255bfef95601890afd80709]*/

/*[clinic input]
_pickle.Pickler.dump

obj: &#39;O&#39;
 The object to be pickled.
 /

Write a pickled representation of obj to the open file.
[clinic start generated code]*/

PyDoc_STRVAR(__pickle_Pickler_dump__doc__,
&quot;Write a pickled representation of obj to the open file.\n&quot;
&quot;\n&quot;
...
static PyObject *
_pickle_Pickler_dump_impl(PicklerObject *self, PyObject *obj)
/*[clinic end generated code: checksum=3bd30745bf206a48f8b576a1da3d90f55a0a4187]*/
{
 /* Check whether the Pickler was initialized correctly (issue3664).
 Developers often forget to call __init__() in their subclasses, which
 would trigger a segfault without this check. */
 if (self-&gt;write == NULL) {
 PyErr_Format(PicklingError,
 &quot;Pickler.__init__() was not called by %s.__init__()&quot;,
 Py_TYPE(self)-&gt;tp_name);
 return NULL;
 }

if (_Pickler_ClearBuffer(self) &lt; 0)
 return NULL;

...
</pre></div>
</div>
</li>
<li>Remember the macro with the <a class="reference internal" href="../c-api/structures.html#c.PyMethodDef" title="PyMethodDef"><code class="xref c c-type docutils literal notranslate">PyMethodDef</code></a> structure for this
function? Find the existing <a class="reference internal" href="../c-api/structures.html#c.PyMethodDef" title="PyMethodDef"><code class="xref c c-type docutils literal notranslate">PyMethodDef</code></a> structure for this
function and replace it with a reference to the macro. (If the builtin
is at module scope, this will probably be very near the end of the file;
if the builtin is a class method, this will probably be below but relatively
near to the implementation.)
Note that the body of the macro contains a trailing comma. So when you
replace the existing static <a class="reference internal" href="../c-api/structures.html#c.PyMethodDef" title="PyMethodDef"><code class="xref c c-type docutils literal notranslate">PyMethodDef</code></a> structure with the macro,
don’t add a comma to the end.
Sample:
<div class="highlight-c notranslate"><div class="highlight"><pre>static struct PyMethodDef Pickler_methods[] = {
 __PICKLE_PICKLER_DUMP_METHODDEF
 __PICKLE_PICKLER_CLEAR_MEMO_METHODDEF
 {NULL, NULL} /* sentinel */
};
</pre></div>
</div>
</li>
<li>Compile, then run the relevant portions of the regression-test suite.
This change should not introduce any new compile-time warnings or errors,
and there should be no externally-visible change to Python’s behavior.
Well, except for one difference: <code class="docutils literal notranslate">inspect.signature()</code> run on your function
should now provide a valid signature!
Congratulations, you’ve ported your first function to work with Argument Clinic!
</li>
</ol>
</div>
<div class="section" id="advanced-topics">
<h2>Advanced Topics<a class="headerlink" href="#advanced-topics" title="Permalink to this headline">¶</a></h2>
Now that you’ve had some experience working with Argument Clinic, it’s time
for some advanced topics.
<div class="section" id="symbolic-default-values">
<h3>Symbolic default values<a class="headerlink" href="#symbolic-default-values" title="Permalink to this headline">¶</a></h3>
The default value you provide for a parameter can’t be any arbitrary
expression. Currently the following are explicitly supported:
<ul class="simple">
<li>Numeric constants (integer and float)</li>
<li>String constants</li>
<li><code class="docutils literal notranslate">True</code>, <code class="docutils literal notranslate">False</code>, and <code class="docutils literal notranslate">None</code></li>
<li>Simple symbolic constants like <code class="docutils literal notranslate">sys.maxsize</code>, which must
start with the name of the module</li>
</ul>
In case you’re curious, this is implemented in <code class="docutils literal notranslate">from_builtin()</code>
in <code class="docutils literal notranslate">Lib/inspect.py</code>.
(In the future, this may need to get even more elaborate,
to allow full expressions like <code class="docutils literal notranslate">CONSTANT - 1</code>.)
</div>
<div class="section" id="renaming-the-c-functions-and-variables-generated-by-argument-clinic">
<h3>Renaming the C functions and variables generated by Argument Clinic<a class="headerlink" href="#renaming-the-c-functions-and-variables-generated-by-argument-clinic" title="Permalink to this headline">¶</a></h3>
Argument Clinic automatically names the functions it generates for you.
Occasionally this may cause a problem, if the generated name collides with
the name of an existing C function. There’s an easy solution: override the names
used for the C functions. Just add the keyword <code class="docutils literal notranslate">&quot;as&quot;</code>
to your function declaration line, followed by the function name you wish to use.
Argument Clinic will use that function name for the base (generated) function,
then add <code class="docutils literal notranslate">&quot;_impl&quot;</code> to the end and use that for the name of the impl function.
For example, if we wanted to rename the C function names generated for
<code class="docutils literal notranslate">pickle.Pickler.dump</code>, it’d look like this:
<div class="highlight-c notranslate"><div class="highlight"><pre>/*[clinic input]
pickle.Pickler.dump as pickler_dumper

...
</pre></div>
</div>
The base function would now be named <code class="docutils literal notranslate">pickler_dumper()</code>,
and the impl function would now be named <code class="docutils literal notranslate">pickler_dumper_impl()</code>.
Similarly, you may have a problem where you want to give a parameter
a specific Python name, but that name may be inconvenient in C. Argument
Clinic allows you to give a parameter different names in Python and in C,
using the same <code class="docutils literal notranslate">&quot;as&quot;</code> syntax:
<div class="highlight-c notranslate"><div class="highlight"><pre>/*[clinic input]
pickle.Pickler.dump

obj: object
 file as file_obj: object
 protocol: object = NULL
 *
 fix_imports: bool = True
</pre></div>
</div>
Here, the name used in Python (in the signature and the <code class="docutils literal notranslate">keywords</code>
array) would be <code class="docutils literal notranslate">file</code>, but the C variable would be named <code class="docutils literal notranslate">file_obj</code>.
You can use this to rename the <code class="docutils literal notranslate">self</code> parameter too!
</div>
<div class="section" id="converting-functions-using-pyarg-unpacktuple">
<h3>Converting functions using PyArg_UnpackTuple<a class="headerlink" href="#converting-functions-using-pyarg-unpacktuple" title="Permalink to this headline">¶</a></h3>
To convert a function parsing its arguments with <a class="reference internal" href="../c-api/arg.html#c.PyArg_UnpackTuple" title="PyArg_UnpackTuple"><code class="xref c c-func docutils literal notranslate">PyArg_UnpackTuple()</code></a>,
simply write out all the arguments, specifying each as an <code class="docutils literal notranslate">object</code>. You
may specify the <code class="docutils literal notranslate">type</code> argument to cast the type as appropriate. All
arguments should be marked positional-only (add a <code class="docutils literal notranslate">/</code> on a line by itself
after the last argument).
Currently the generated code will use <a class="reference internal" href="../c-api/arg.html#c.PyArg_ParseTuple" title="PyArg_ParseTuple"><code class="xref c c-func docutils literal notranslate">PyArg_ParseTuple()</code></a>, but this
will change soon.
</div>
<div class="section" id="optional-groups">
<h3>Optional Groups<a class="headerlink" href="#optional-groups" title="Permalink to this headline">¶</a></h3>
Some legacy functions have a tricky approach to parsing their arguments:
they count the number of positional arguments, then use a <code class="docutils literal notranslate">switch</code> statement
to call one of several different <a class="reference internal" href="../c-api/arg.html#c.PyArg_ParseTuple" title="PyArg_ParseTuple"><code class="xref c c-func docutils literal notranslate">PyArg_ParseTuple()</code></a> calls depending on
how many positional arguments there are. (These functions cannot accept
keyword-only arguments.) This approach was used to simulate optional
arguments back before <a class="reference internal" href="../c-api/arg.html#c.PyArg_ParseTupleAndKeywords" title="PyArg_ParseTupleAndKeywords"><code class="xref c c-func docutils literal notranslate">PyArg_ParseTupleAndKeywords()</code></a> was created.
While functions using this approach can often be converted to
use <a class="reference internal" href="../c-api/arg.html#c.PyArg_ParseTupleAndKeywords" title="PyArg_ParseTupleAndKeywords"><code class="xref c c-func docutils literal notranslate">PyArg_ParseTupleAndKeywords()</code></a>, optional arguments, and default values,
it’s not always possible. Some of these legacy functions have
behaviors <a class="reference internal" href="../c-api/arg.html#c.PyArg_ParseTupleAndKeywords" title="PyArg_ParseTupleAndKeywords"><code class="xref c c-func docutils literal notranslate">PyArg_ParseTupleAndKeywords()</code></a> doesn’t directly support.
The most obvious example is the builtin function <code class="docutils literal notranslate">range()</code>, which has
an optional argument on the left side of its required argument!
Another example is <code class="docutils literal notranslate">curses.window.addch()</code>, which has a group of two
arguments that must always be specified together. (The arguments are
called <code class="docutils literal notranslate">x</code> and <code class="docutils literal notranslate">y</code>; if you call the function passing in <code class="docutils literal notranslate">x</code>,
you must also pass in <code class="docutils literal notranslate">y</code>—and if you don’t pass in <code class="docutils literal notranslate">x</code> you may not
pass in <code class="docutils literal notranslate">y</code> either.)
In any case, the goal of Argument Clinic is to support argument parsing
for all existing CPython builtins without changing their semantics.
Therefore Argument Clinic supports
this alternate approach to parsing, using what are called optional groups.
Optional groups are groups of arguments that must all be passed in together.
They can be to the left or the right of the required arguments. They
can only be used with positional-only parameters.
<div class="admonition note">
Note
Optional groups are only intended for use when converting
functions that make multiple calls to <a class="reference internal" href="../c-api/arg.html#c.PyArg_ParseTuple" title="PyArg_ParseTuple"><code class="xref c c-func docutils literal notranslate">PyArg_ParseTuple()</code></a>!
Functions that use any other approach for parsing arguments
should almost never be converted to Argument Clinic using
optional groups. Functions using optional groups currently
cannot have accurate signatures in Python, because Python just
doesn’t understand the concept. Please avoid using optional
groups wherever possible.
</div>
To specify an optional group, add a <code class="docutils literal notranslate">[</code> on a line by itself before
the parameters you wish to group together, and a <code class="docutils literal notranslate">]</code> on a line by itself
after these parameters. As an example, here’s how <code class="docutils literal notranslate">curses.window.addch</code>
uses optional groups to make the first two parameters and the last
parameter optional:
<div class="highlight-c notranslate"><div class="highlight"><pre>/*[clinic input]

curses.window.addch

[
 x: int
 X-coordinate.
 y: int
 Y-coordinate.
 ]

ch: object
 Character to add.

[
 attr: long
 Attributes for the character.
 ]
 /

...
</pre></div>
</div>
Notes:
<ul class="simple">
<li>For every optional group, one additional parameter will be passed into the
impl function representing the group. The parameter will be an int named
<code class="docutils literal notranslate">group_{direction}_{number}</code>,
where <code class="docutils literal notranslate">{direction}</code> is either <code class="docutils literal notranslate">right</code> or <code class="docutils literal notranslate">left</code> depending on whether the group
is before or after the required parameters, and <code class="docutils literal notranslate">{number}</code> is a monotonically
increasing number (starting at 1) indicating how far away the group is from
the required parameters. When the impl is called, this parameter will be set
to zero if this group was unused, and set to non-zero if this group was used.
(By used or unused, I mean whether or not the parameters received arguments
in this invocation.)</li>
<li>If there are no required arguments, the optional groups will behave
as if they’re to the right of the required arguments.</li>
<li>In the case of ambiguity, the argument parsing code
favors parameters on the left (before the required parameters).</li>
<li>Optional groups can only contain positional-only parameters.</li>
<li>Optional groups are only intended for legacy code. Please do not
use optional groups for new code.</li>
</ul>
</div>
<div class="section" id="using-real-argument-clinic-converters-instead-of-legacy-converters">
<h3>Using real Argument Clinic converters, instead of “legacy converters”<a class="headerlink" href="#using-real-argument-clinic-converters-instead-of-legacy-converters" title="Permalink to this headline">¶</a></h3>
To save time, and to minimize how much you need to learn
to achieve your first port to Argument Clinic, the walkthrough above tells
you to use “legacy converters”. “Legacy converters” are a convenience,
designed explicitly to make porting existing code to Argument Clinic
easier. And to be clear, their use is acceptable when porting code for
Python 3.4.
However, in the long term we probably want all our blocks to
use Argument Clinic’s real syntax for converters. Why? A couple
reasons:
<ul class="simple">
<li>The proper converters are far easier to read and clearer in their intent.</li>
<li>There are some format units that are unsupported as “legacy converters”,
because they require arguments, and the legacy converter syntax doesn’t
support specifying arguments.</li>
<li>In the future we may have a new argument parsing library that isn’t
restricted to what <a class="reference internal" href="../c-api/arg.html#c.PyArg_ParseTuple" title="PyArg_ParseTuple"><code class="xref c c-func docutils literal notranslate">PyArg_ParseTuple()</code></a> supports; this flexibility
won’t be available to parameters using legacy converters.</li>
</ul>
Therefore, if you don’t mind a little extra effort, please use the normal
converters instead of legacy converters.
In a nutshell, the syntax for Argument Clinic (non-legacy) converters
looks like a Python function call. However, if there are no explicit
arguments to the function (all functions take their default values),
you may omit the parentheses. Thus <code class="docutils literal notranslate">bool</code> and <code class="docutils literal notranslate">bool()</code> are exactly
the same converters.
All arguments to Argument Clinic converters are keyword-only.
All Argument Clinic converters accept the following arguments:
<blockquote>
<div><dl class="docutils">
<dt><code class="docutils literal notranslate">c_default</code></dt>
<dd>The default value for this parameter when defined in C.
Specifically, this will be the initializer for the variable declared
in the “parse function”. See <a class="reference internal" href="#default-values">the section on default values</a>
for how to use this.
Specified as a string.</dd>
<dt><code class="docutils literal notranslate">annotation</code></dt>
<dd>The annotation value for this parameter. Not currently supported,
because PEP 8 mandates that the Python library may not use
annotations.</dd>
</dl>
</div></blockquote>
In addition, some converters accept additional arguments. Here is a list
of these arguments, along with their meanings:
<blockquote>
<div><dl class="docutils">
<dt><code class="docutils literal notranslate">accept</code></dt>
<dd>A set of Python types (and possibly pseudo-types);
this restricts the allowable Python argument to values of these types.
(This is not a general-purpose facility; as a rule it only supports
specific lists of types as shown in the legacy converter table.)
To accept <code class="docutils literal notranslate">None</code>, add <code class="docutils literal notranslate">NoneType</code> to this set.
</dd>
<dt><code class="docutils literal notranslate">bitwise</code></dt>
<dd>Only supported for unsigned integers. The native integer value of this
Python argument will be written to the parameter without any range checking,
even for negative values.</dd>
<dt><code class="docutils literal notranslate">converter</code></dt>
<dd>Only supported by the <code class="docutils literal notranslate">object</code> converter. Specifies the name of a
<a class="reference internal" href="../c-api/arg.html#o-ampersand">C “converter function”</a>
to use to convert this object to a native type.</dd>
<dt><code class="docutils literal notranslate">encoding</code></dt>
<dd>Only supported for strings. Specifies the encoding to use when converting
this string from a Python str (Unicode) value into a C <code class="docutils literal notranslate">char *</code> value.</dd>
<dt><code class="docutils literal notranslate">subclass_of</code></dt>
<dd>Only supported for the <code class="docutils literal notranslate">object</code> converter. Requires that the Python
value be a subclass of a Python type, as expressed in C.</dd>
<dt><code class="docutils literal notranslate">type</code></dt>
<dd>Only supported for the <code class="docutils literal notranslate">object</code> and <code class="docutils literal notranslate">self</code> converters. Specifies
the C type that will be used to declare the variable. Default value is
<code class="docutils literal notranslate">&quot;PyObject *&quot;</code>.</dd>
<dt><code class="docutils literal notranslate">zeroes</code></dt>
<dd>Only supported for strings. If true, embedded NUL bytes (<code class="docutils literal notranslate">'\\0'</code>) are
permitted inside the value. The length of the string will be passed in
to the impl function, just after the string parameter, as a parameter named
<code class="docutils literal notranslate">&lt;parameter_name&gt;_length</code>.</dd>
</dl>
</div></blockquote>
Please note, not every possible combination of arguments will work.
Usually these arguments are implemented by specific <code class="docutils literal notranslate">PyArg_ParseTuple</code>
format units, with specific behavior. For example, currently you cannot
call <code class="docutils literal notranslate">unsigned_short</code> without also specifying <code class="docutils literal notranslate">bitwise=True</code>.
Although it’s perfectly reasonable to think this would work, these semantics don’t
map to any existing format unit. So Argument Clinic doesn’t support it. (Or, at
least, not yet.)
Below is a table showing the mapping of legacy converters into real
Argument Clinic converters. On the left is the legacy converter,
on the right is the text you’d replace it with.
<table border="1" class="docutils">
<colgroup>
<col width="10%" />
<col width="90%" />
</colgroup>
<tbody valign="top">
<tr class="row-odd"><td><code class="docutils literal notranslate">'B'</code></td>
<td><code class="docutils literal notranslate">unsigned_char(bitwise=True)</code></td>
</tr>
<tr class="row-even"><td><code class="docutils literal notranslate">'b'</code></td>
<td><code class="docutils literal notranslate">unsigned_char</code></td>
</tr>
<tr class="row-odd"><td><code class="docutils literal notranslate">'c'</code></td>
<td><code class="docutils literal notranslate">char</code></td>
</tr>
<tr class="row-even"><td><code class="docutils literal notranslate">'C'</code></td>
<td><code class="docutils literal notranslate">int(accept={str})</code></td>
</tr>
<tr class="row-odd"><td><code class="docutils literal notranslate">'d'</code></td>
<td><code class="docutils literal notranslate">double</code></td>
</tr>
<tr class="row-even"><td><code class="docutils literal notranslate">'D'</code></td>
<td><code class="docutils literal notranslate">Py_complex</code></td>
</tr>
<tr class="row-odd"><td><code class="docutils literal notranslate">'es'</code></td>
<td><code class="docutils literal notranslate">str(encoding='name_of_encoding')</code></td>
</tr>
<tr class="row-even"><td><code class="docutils literal notranslate">'es#'</code></td>
<td><code class="docutils literal notranslate">str(encoding='name_of_encoding', zeroes=True)</code></td>
</tr>
<tr class="row-odd"><td><code class="docutils literal notranslate">'et'</code></td>
<td><code class="docutils literal notranslate">str(encoding='name_of_encoding', accept={bytes, bytearray, str})</code></td>
</tr>
<tr class="row-even"><td><code class="docutils literal notranslate">'et#'</code></td>
<td><code class="docutils literal notranslate">str(encoding='name_of_encoding', accept={bytes, bytearray, str}, zeroes=True)</code></td>
</tr>
<tr class="row-odd"><td><code class="docutils literal notranslate">'f'</code></td>
<td><code class="docutils literal notranslate">float</code></td>
</tr>
<tr class="row-even"><td><code class="docutils literal notranslate">'h'</code></td>
<td><code class="docutils literal notranslate">short</code></td>
</tr>
<tr class="row-odd"><td><code class="docutils literal notranslate">'H'</code></td>
<td><code class="docutils literal notranslate">unsigned_short(bitwise=True)</code></td>
</tr>
<tr class="row-even"><td><code class="docutils literal notranslate">'i'</code></td>
<td><code class="docutils literal notranslate">int</code></td>
</tr>
<tr class="row-odd"><td><code class="docutils literal notranslate">'I'</code></td>
<td><code class="docutils literal notranslate">unsigned_int(bitwise=True)</code></td>
</tr>
<tr class="row-even"><td><code class="docutils literal notranslate">'k'</code></td>
<td><code class="docutils literal notranslate">unsigned_long(bitwise=True)</code></td>
</tr>
<tr class="row-odd"><td><code class="docutils literal notranslate">'K'</code></td>
<td><code class="docutils literal notranslate">unsigned_long_long(bitwise=True)</code></td>
</tr>
<tr class="row-even"><td><code class="docutils literal notranslate">'l'</code></td>
<td><code class="docutils literal notranslate">long</code></td>
</tr>
<tr class="row-odd"><td><code class="docutils literal notranslate">'L'</code></td>
<td><code class="docutils literal notranslate">long long</code></td>
</tr>
<tr class="row-even"><td><code class="docutils literal notranslate">'n'</code></td>
<td><code class="docutils literal notranslate">Py_ssize_t</code></td>
</tr>
<tr class="row-odd"><td><code class="docutils literal notranslate">'O'</code></td>
<td><code class="docutils literal notranslate">object</code></td>
</tr>
<tr class="row-even"><td><code class="docutils literal notranslate">'O!'</code></td>
<td><code class="docutils literal notranslate">object(subclass_of='&amp;PySomething_Type')</code></td>
</tr>
<tr class="row-odd"><td><code class="docutils literal notranslate">'O&amp;'</code></td>
<td><code class="docutils literal notranslate">object(converter='name_of_c_function')</code></td>
</tr>
<tr class="row-even"><td><code class="docutils literal notranslate">'p'</code></td>
<td><code class="docutils literal notranslate">bool</code></td>
</tr>
<tr class="row-odd"><td><code class="docutils literal notranslate">'S'</code></td>
<td><code class="docutils literal notranslate">PyBytesObject</code></td>
</tr>
<tr class="row-even"><td><code class="docutils literal notranslate">'s'</code></td>
<td><code class="docutils literal notranslate">str</code></td>
</tr>
<tr class="row-odd"><td><code class="docutils literal notranslate">'s#'</code></td>
<td><code class="docutils literal notranslate">str(zeroes=True)</code></td>
</tr>
<tr class="row-even"><td><code class="docutils literal notranslate">'s*'</code></td>
<td><code class="docutils literal notranslate">Py_buffer(accept={buffer, str})</code></td>
</tr>
<tr class="row-odd"><td><code class="docutils literal notranslate">'U'</code></td>
<td><code class="docutils literal notranslate">unicode</code></td>
</tr>
<tr class="row-even"><td><code class="docutils literal notranslate">'u'</code></td>
<td><code class="docutils literal notranslate">Py_UNICODE</code></td>
</tr>
<tr class="row-odd"><td><code class="docutils literal notranslate">'u#'</code></td>
<td><code class="docutils literal notranslate">Py_UNICODE(zeroes=True)</code></td>
</tr>
<tr class="row-even"><td><code class="docutils literal notranslate">'w*'</code></td>
<td><code class="docutils literal notranslate">Py_buffer(accept={rwbuffer})</code></td>
</tr>
<tr class="row-odd"><td><code class="docutils literal notranslate">'Y'</code></td>
<td><code class="docutils literal notranslate">PyByteArrayObject</code></td>
</tr>
<tr class="row-even"><td><code class="docutils literal notranslate">'y'</code></td>
<td><code class="docutils literal notranslate">str(accept={bytes})</code></td>
</tr>
<tr class="row-odd"><td><code class="docutils literal notranslate">'y#'</code></td>
<td><code class="docutils literal notranslate">str(accept={robuffer}, zeroes=True)</code></td>
</tr>
<tr class="row-even"><td><code class="docutils literal notranslate">'y*'</code></td>
<td><code class="docutils literal notranslate">Py_buffer</code></td>
</tr>
<tr class="row-odd"><td><code class="docutils literal notranslate">'Z'</code></td>
<td><code class="docutils literal notranslate">Py_UNICODE(accept={str, NoneType})</code></td>
</tr>
<tr class="row-even"><td><code class="docutils literal notranslate">'Z#'</code></td>
<td><code class="docutils literal notranslate">Py_UNICODE(accept={str, NoneType}, zeroes=True)</code></td>
</tr>
<tr class="row-odd"><td><code class="docutils literal notranslate">'z'</code></td>
<td><code class="docutils literal notranslate">str(accept={str, NoneType})</code></td>
</tr>
<tr class="row-even"><td><code class="docutils literal notranslate">'z#'</code></td>
<td><code class="docutils literal notranslate">str(accept={str, NoneType}, zeroes=True)</code></td>
</tr>
<tr class="row-odd"><td><code class="docutils literal notranslate">'z*'</code></td>
<td><code class="docutils literal notranslate">Py_buffer(accept={buffer, str, NoneType})</code></td>
</tr>
</tbody>
</table>
As an example, here’s our sample <code class="docutils literal notranslate">pickle.Pickler.dump</code> using the proper
converter:
<div class="highlight-c notranslate"><div class="highlight"><pre>/*[clinic input]
pickle.Pickler.dump

obj: object
 The object to be pickled.
 /

Write a pickled representation of obj to the open file.
[clinic start generated code]*/
</pre></div>
</div>
Argument Clinic will show you all the converters it has
available. For each converter it’ll show you all the parameters
it accepts, along with the default value for each parameter.
Just run <code class="docutils literal notranslate">Tools/clinic/clinic.py --converters</code> to see the full list.
</div>
<div class="section" id="py-buffer">
<h3>Py_buffer<a class="headerlink" href="#py-buffer" title="Permalink to this headline">¶</a></h3>
When using the <code class="docutils literal notranslate">Py_buffer</code> converter
(or the <code class="docutils literal notranslate">'s*'</code>, <code class="docutils literal notranslate">'w*'</code>, <code class="docutils literal notranslate">'*y'</code>, or <code class="docutils literal notranslate">'z*'</code> legacy converters),
you must not call <a class="reference internal" href="../c-api/buffer.html#c.PyBuffer_Release" title="PyBuffer_Release"><code class="xref c c-func docutils literal notranslate">PyBuffer_Release()</code></a> on the provided buffer.
Argument Clinic generates code that does it for you (in the parsing function).
</div>
<div class="section" id="advanced-converters">
<h3>Advanced converters<a class="headerlink" href="#advanced-converters" title="Permalink to this headline">¶</a></h3>
Remember those format units you skipped for your first
time because they were advanced? Here’s how to handle those too.
The trick is, all those format units take arguments—either
conversion functions, or types, or strings specifying an encoding.
(But “legacy converters” don’t support arguments. That’s why we
skipped them for your first function.) The argument you specified
to the format unit is now an argument to the converter; this
argument is either <code class="docutils literal notranslate">converter</code> (for <code class="docutils literal notranslate">O&amp;</code>), <code class="docutils literal notranslate">subclass_of</code> (for <code class="docutils literal notranslate">O!</code>),
or <code class="docutils literal notranslate">encoding</code> (for all the format units that start with <code class="docutils literal notranslate">e</code>).
When using <code class="docutils literal notranslate">subclass_of</code>, you may also want to use the other
custom argument for <code class="docutils literal notranslate">object()</code>: <code class="docutils literal notranslate">type</code>, which lets you set the type
actually used for the parameter. For example, if you want to ensure
that the object is a subclass of <code class="docutils literal notranslate">PyUnicode_Type</code>, you probably want
to use the converter <code class="docutils literal notranslate">object(type='PyUnicodeObject *', subclass_of='&amp;PyUnicode_Type')</code>.
One possible problem with using Argument Clinic: it takes away some possible
flexibility for the format units starting with <code class="docutils literal notranslate">e</code>. When writing a
<code class="docutils literal notranslate">PyArg_Parse</code> call by hand, you could theoretically decide at runtime what
encoding string to pass in to <a class="reference internal" href="../c-api/arg.html#c.PyArg_ParseTuple" title="PyArg_ParseTuple"><code class="xref c c-func docutils literal notranslate">PyArg_ParseTuple()</code></a>. But now this string must
be hard-coded at Argument-Clinic-preprocessing-time. This limitation is deliberate;
it made supporting this format unit much easier, and may allow for future optimizations.
This restriction doesn’t seem unreasonable; CPython itself always passes in static
hard-coded encoding strings for parameters whose format units start with <code class="docutils literal notranslate">e</code>.
</div>
<div class="section" id="parameter-default-values">
<h3>Parameter default values<a class="headerlink" href="#parameter-default-values" title="Permalink to this headline">¶</a></h3>
Default values for parameters can be any of a number of values.
At their simplest, they can be string, int, or float literals:
<div class="highlight-none notranslate"><div class="highlight"><pre>foo: str = &quot;abc&quot;
bar: int = 123
bat: float = 45.6
</pre></div>
</div>
They can also use any of Python’s built-in constants:
<div class="highlight-none notranslate"><div class="highlight"><pre>yep: bool = True
nope: bool = False
nada: object = None
</pre></div>
</div>
There’s also special support for a default value of <code class="docutils literal notranslate">NULL</code>, and
for simple expressions, documented in the following sections.
</div>
<div class="section" id="the-null-default-value">
<h3>The <code class="docutils literal notranslate">NULL</code> default value<a class="headerlink" href="#the-null-default-value" title="Permalink to this headline">¶</a></h3>
For string and object parameters, you can set them to <code class="docutils literal notranslate">None</code> to indicate
that there’s no default. However, that means the C variable will be
initialized to <code class="docutils literal notranslate">Py_None</code>. For convenience’s sakes, there’s a special
value called <code class="docutils literal notranslate">NULL</code> for just this reason: from Python’s perspective it
behaves like a default value of <code class="docutils literal notranslate">None</code>, but the C variable is initialized
with <code class="docutils literal notranslate">NULL</code>.
</div>
<div class="section" id="expressions-specified-as-default-values">
<h3>Expressions specified as default values<a class="headerlink" href="#expressions-specified-as-default-values" title="Permalink to this headline">¶</a></h3>
The default value for a parameter can be more than just a literal value.
It can be an entire expression, using math operators and looking up attributes
on objects. However, this support isn’t exactly simple, because of some
non-obvious semantics.
Consider the following example:
<div class="highlight-none notranslate"><div class="highlight"><pre>foo: Py_ssize_t = sys.maxsize - 1
</pre></div>
</div>
<code class="docutils literal notranslate">sys.maxsize</code> can have different values on different platforms. Therefore
Argument Clinic can’t simply evaluate that expression locally and hard-code it
in C. So it stores the default in such a way that it will get evaluated at
runtime, when the user asks for the function’s signature.
What namespace is available when the expression is evaluated? It’s evaluated
in the context of the module the builtin came from. So, if your module has an
attribute called “<code class="docutils literal notranslate">max_widgets</code>”, you may simply use it:
<div class="highlight-none notranslate"><div class="highlight"><pre>foo: Py_ssize_t = max_widgets
</pre></div>
</div>
If the symbol isn’t found in the current module, it fails over to looking in
<code class="docutils literal notranslate">sys.modules</code>. That’s how it can find <code class="docutils literal notranslate">sys.maxsize</code> for example. (Since you
don’t know in advance what modules the user will load into their interpreter,
it’s best to restrict yourself to modules that are preloaded by Python itself.)
Evaluating default values only at runtime means Argument Clinic can’t compute
the correct equivalent C default value. So you need to tell it explicitly.
When you use an expression, you must also specify the equivalent expression
in C, using the <code class="docutils literal notranslate">c_default</code> parameter to the converter:
<div class="highlight-none notranslate"><div class="highlight"><pre>foo: Py_ssize_t(c_default=&quot;PY_SSIZE_T_MAX - 1&quot;) = sys.maxsize - 1
</pre></div>
</div>
Another complication: Argument Clinic can’t know in advance whether or not the
expression you supply is valid. It parses it to make sure it looks legal, but
it can’t actually know. You must be very careful when using expressions to
specify values that are guaranteed to be valid at runtime!
Finally, because expressions must be representable as static C values, there
are many restrictions on legal expressions. Here’s a list of Python features
you’re not permitted to use:
<ul class="simple">
<li>Function calls.</li>
<li>Inline if statements (<code class="docutils literal notranslate">3 if foo else 5</code>).</li>
<li>Automatic sequence unpacking (<code class="docutils literal notranslate">*[1, 2, 3]</code>).</li>
<li>List/set/dict comprehensions and generator expressions.</li>
<li>Tuple/list/set/dict literals.</li>
</ul>
</div>
<div class="section" id="using-a-return-converter">
<h3>Using a return converter<a class="headerlink" href="#using-a-return-converter" title="Permalink to this headline">¶</a></h3>
By default the impl function Argument Clinic generates for you returns <code class="docutils literal notranslate">PyObject *</code>.
But your C function often computes some C type, then converts it into the <code class="docutils literal notranslate">PyObject *</code>
at the last moment. Argument Clinic handles converting your inputs from Python types
into native C types—why not have it convert your return value from a native C type
into a Python type too?
That’s what a “return converter” does. It changes your impl function to return
some C type, then adds code to the generated (non-impl) function to handle converting
that value into the appropriate <code class="docutils literal notranslate">PyObject *</code>.
The syntax for return converters is similar to that of parameter converters.
You specify the return converter like it was a return annotation on the
function itself. Return converters behave much the same as parameter converters;
they take arguments, the arguments are all keyword-only, and if you’re not changing
any of the default arguments you can omit the parentheses.
(If you use both <code class="docutils literal notranslate">&quot;as&quot;</code> and a return converter for your function,
the <code class="docutils literal notranslate">&quot;as&quot;</code> should come before the return converter.)
There’s one additional complication when using return converters: how do you
indicate an error has occurred? Normally, a function returns a valid (non-<code class="docutils literal notranslate">NULL</code>)
pointer for success, and <code class="docutils literal notranslate">NULL</code> for failure. But if you use an integer return converter,
all integers are valid. How can Argument Clinic detect an error? Its solution: each return
converter implicitly looks for a special value that indicates an error. If you return
that value, and an error has been set (<code class="docutils literal notranslate">PyErr_Occurred()</code> returns a true
value), then the generated code will propagate the error. Otherwise it will
encode the value you return like normal.
Currently Argument Clinic supports only a few return converters:
<div class="highlight-none notranslate"><div class="highlight"><pre>bool
int
unsigned int
long
unsigned int
size_t
Py_ssize_t
float
double
DecodeFSDefault
</pre></div>
</div>
None of these take parameters. For the first three, return -1 to indicate
error. For <code class="docutils literal notranslate">DecodeFSDefault</code>, the return type is <code class="docutils literal notranslate">char *</code>; return a NULL
pointer to indicate an error.
(There’s also an experimental <code class="docutils literal notranslate">NoneType</code> converter, which lets you
return <code class="docutils literal notranslate">Py_None</code> on success or <code class="docutils literal notranslate">NULL</code> on failure, without having
to increment the reference count on <code class="docutils literal notranslate">Py_None</code>. I’m not sure it adds
enough clarity to be worth using.)
To see all the return converters Argument Clinic supports, along with
their parameters (if any),
just run <code class="docutils literal notranslate">Tools/clinic/clinic.py --converters</code> for the full list.
</div>
<div class="section" id="cloning-existing-functions">
<h3>Cloning existing functions<a class="headerlink" href="#cloning-existing-functions" title="Permalink to this headline">¶</a></h3>
If you have a number of functions that look similar, you may be able to
use Clinic’s “clone” feature. When you clone an existing function,
you reuse:
<ul class="simple">
<li>its parameters, including<ul>
<li>their names,</li>
<li>their converters, with all parameters,</li>
<li>their default values,</li>
<li>their per-parameter docstrings,</li>
<li>their kind (whether they’re positional only,
positional or keyword, or keyword only), and</li>
</ul>
</li>
<li>its return converter.</li>
</ul>
The only thing not copied from the original function is its docstring;
the syntax allows you to specify a new docstring.
Here’s the syntax for cloning a function:
<div class="highlight-c notranslate"><div class="highlight"><pre>/*[clinic input]
module.class.new_function [as c_basename] = module.class.existing_function

Docstring for new_function goes here.
[clinic start generated code]*/
</pre></div>
</div>
(The functions can be in different modules or classes. I wrote
<code class="docutils literal notranslate">module.class</code> in the sample just to illustrate that you must
use the full path to both functions.)
Sorry, there’s no syntax for partially-cloning a function, or cloning a function
then modifying it. Cloning is an all-or nothing proposition.
Also, the function you are cloning from must have been previously defined
in the current file.
</div>
<div class="section" id="calling-python-code">
<h3>Calling Python code<a class="headerlink" href="#calling-python-code" title="Permalink to this headline">¶</a></h3>
The rest of the advanced topics require you to write Python code
which lives inside your C file and modifies Argument Clinic’s
runtime state. This is simple: you simply define a Python block.
A Python block uses different delimiter lines than an Argument
Clinic function block. It looks like this:
<div class="highlight-c notranslate"><div class="highlight"><pre>/*[python input]
# python code goes here
[python start generated code]*/
</pre></div>
</div>
All the code inside the Python block is executed at the
time it’s parsed. All text written to stdout inside the block
is redirected into the “output” after the block.
As an example, here’s a Python block that adds a static integer
variable to the C code:
<div class="highlight-c notranslate"><div class="highlight"><pre>/*[python input]
print(&#39;static int __ignored_unused_variable__ = 0;&#39;)
[python start generated code]*/
static int __ignored_unused_variable__ = 0;
/*[python checksum:...]*/
</pre></div>
</div>
</div>
<div class="section" id="using-a-self-converter">
<h3>Using a “self converter”<a class="headerlink" href="#using-a-self-converter" title="Permalink to this headline">¶</a></h3>
Argument Clinic automatically adds a “self” parameter for you
using a default converter. It automatically sets the <code class="docutils literal notranslate">type</code>
of this parameter to the “pointer to an instance” you specified
when you declared the type. However, you can override
Argument Clinic’s converter and specify one yourself.
Just add your own <code class="docutils literal notranslate">self</code> parameter as the first parameter in a
block, and ensure that its converter is an instance of
<code class="docutils literal notranslate">self_converter</code> or a subclass thereof.
What’s the point? This lets you override the type of <code class="docutils literal notranslate">self</code>,
or give it a different default name.
How do you specify the custom type you want to cast <code class="docutils literal notranslate">self</code> to?
If you only have one or two functions with the same type for <code class="docutils literal notranslate">self</code>,
you can directly use Argument Clinic’s existing <code class="docutils literal notranslate">self</code> converter,
passing in the type you want to use as the <code class="docutils literal notranslate">type</code> parameter:
<div class="highlight-c notranslate"><div class="highlight"><pre>/*[clinic input]

_pickle.Pickler.dump

self: self(type=&quot;PicklerObject *&quot;)
 obj: object
 /

Write a pickled representation of the given object to the open file.
[clinic start generated code]*/
</pre></div>
</div>
On the other hand, if you have a lot of functions that will use the same
type for <code class="docutils literal notranslate">self</code>, it’s best to create your own converter, subclassing
<code class="docutils literal notranslate">self_converter</code> but overwriting the <code class="docutils literal notranslate">type</code> member:
<div class="highlight-c notranslate"><div class="highlight"><pre>/*[python input]
class PicklerObject_converter(self_converter):
 type = &quot;PicklerObject *&quot;
[python start generated code]*/

/*[clinic input]

_pickle.Pickler.dump

self: PicklerObject
 obj: object
 /

Write a pickled representation of the given object to the open file.
[clinic start generated code]*/
</pre></div>
</div>
</div>
<div class="section" id="writing-a-custom-converter">
<h3>Writing a custom converter<a class="headerlink" href="#writing-a-custom-converter" title="Permalink to this headline">¶</a></h3>
As we hinted at in the previous section… you can write your own converters!
A converter is simply a Python class that inherits from <code class="docutils literal notranslate">CConverter</code>.
The main purpose of a custom converter is if you have a parameter using
the <code class="docutils literal notranslate">O&amp;</code> format unit—parsing this parameter means calling
a <a class="reference internal" href="../c-api/arg.html#c.PyArg_ParseTuple" title="PyArg_ParseTuple"><code class="xref c c-func docutils literal notranslate">PyArg_ParseTuple()</code></a> “converter function”.
Your converter class should be named <code class="docutils literal notranslate">*something*_converter</code>.
If the name follows this convention, then your converter class
will be automatically registered with Argument Clinic; its name
will be the name of your class with the <code class="docutils literal notranslate">_converter</code> suffix
stripped off. (This is accomplished with a metaclass.)
You shouldn’t subclass <code class="docutils literal notranslate">CConverter.__init__</code>. Instead, you should
write a <code class="docutils literal notranslate">converter_init()</code> function. <code class="docutils literal notranslate">converter_init()</code>
always accepts a <code class="docutils literal notranslate">self</code> parameter; after that, all additional
parameters must be keyword-only. Any arguments passed in to
the converter in Argument Clinic will be passed along to your
<code class="docutils literal notranslate">converter_init()</code>.
There are some additional members of <code class="docutils literal notranslate">CConverter</code> you may wish
to specify in your subclass. Here’s the current list:
<dl class="docutils">
<dt><code class="docutils literal notranslate">type</code></dt>
<dd>The C type to use for this variable.
<code class="docutils literal notranslate">type</code> should be a Python string specifying the type, e.g. <code class="docutils literal notranslate">int</code>.
If this is a pointer type, the type string should end with <code class="docutils literal notranslate">' *'</code>.</dd>
<dt><code class="docutils literal notranslate">default</code></dt>
<dd>The Python default value for this parameter, as a Python value.
Or the magic value <code class="docutils literal notranslate">unspecified</code> if there is no default.</dd>
<dt><code class="docutils literal notranslate">py_default</code></dt>
<dd><code class="docutils literal notranslate">default</code> as it should appear in Python code,
as a string.
Or <code class="docutils literal notranslate">None</code> if there is no default.</dd>
<dt><code class="docutils literal notranslate">c_default</code></dt>
<dd><code class="docutils literal notranslate">default</code> as it should appear in C code,
as a string.
Or <code class="docutils literal notranslate">None</code> if there is no default.</dd>
<dt><code class="docutils literal notranslate">c_ignored_default</code></dt>
<dd>The default value used to initialize the C variable when
there is no default, but not specifying a default may
result in an “uninitialized variable” warning. This can
easily happen when using option groups—although
properly-written code will never actually use this value,
the variable does get passed in to the impl, and the
C compiler will complain about the “use” of the
uninitialized value. This value should always be a
non-empty string.</dd>
<dt><code class="docutils literal notranslate">converter</code></dt>
<dd>The name of the C converter function, as a string.</dd>
<dt><code class="docutils literal notranslate">impl_by_reference</code></dt>
<dd>A boolean value. If true,
Argument Clinic will add a <code class="docutils literal notranslate">&amp;</code> in front of the name of
the variable when passing it into the impl function.</dd>
<dt><code class="docutils literal notranslate">parse_by_reference</code></dt>
<dd>A boolean value. If true,
Argument Clinic will add a <code class="docutils literal notranslate">&amp;</code> in front of the name of
the variable when passing it into <a class="reference internal" href="../c-api/arg.html#c.PyArg_ParseTuple" title="PyArg_ParseTuple"><code class="xref c c-func docutils literal notranslate">PyArg_ParseTuple()</code></a>.</dd>
</dl>
Here’s the simplest example of a custom converter, from <code class="docutils literal notranslate">Modules/zlibmodule.c</code>:
<div class="highlight-c notranslate"><div class="highlight"><pre>/*[python input]

class ssize_t_converter(CConverter):
 type = &#39;Py_ssize_t&#39;
 converter = &#39;ssize_t_converter&#39;

[python start generated code]*/
/*[python end generated code: output=da39a3ee5e6b4b0d input=35521e4e733823c7]*/
</pre></div>
</div>
This block adds a converter to Argument Clinic named <code class="docutils literal notranslate">ssize_t</code>. Parameters
declared as <code class="docutils literal notranslate">ssize_t</code> will be declared as type <code class="docutils literal notranslate">Py_ssize_t</code>, and will
be parsed by the <code class="docutils literal notranslate">'O&amp;'</code> format unit, which will call the
<code class="docutils literal notranslate">ssize_t_converter</code> converter function. <code class="docutils literal notranslate">ssize_t</code> variables
automatically support default values.
More sophisticated custom converters can insert custom C code to
handle initialization and cleanup.
You can see more examples of custom converters in the CPython
source tree; grep the C files for the string <code class="docutils literal notranslate">CConverter</code>.
</div>
<div class="section" id="writing-a-custom-return-converter">
<h3>Writing a custom return converter<a class="headerlink" href="#writing-a-custom-return-converter" title="Permalink to this headline">¶</a></h3>
Writing a custom return converter is much like writing
a custom converter. Except it’s somewhat simpler, because return
converters are themselves much simpler.
Return converters must subclass <code class="docutils literal notranslate">CReturnConverter</code>.
There are no examples yet of custom return converters,
because they are not widely used yet. If you wish to
write your own return converter, please read <code class="docutils literal notranslate">Tools/clinic/clinic.py</code>,
specifically the implementation of <code class="docutils literal notranslate">CReturnConverter</code> and
all its subclasses.
</div>
<div class="section" id="meth-o-and-meth-noargs">
<h3>METH_O and METH_NOARGS<a class="headerlink" href="#meth-o-and-meth-noargs" title="Permalink to this headline">¶</a></h3>
To convert a function using <code class="docutils literal notranslate">METH_O</code>, make sure the function’s
single argument is using the <code class="docutils literal notranslate">object</code> converter, and mark the
arguments as positional-only:
<div class="highlight-c notranslate"><div class="highlight"><pre>/*[clinic input]
meth_o_sample

argument: object
 /
[clinic start generated code]*/
</pre></div>
</div>
To convert a function using <code class="docutils literal notranslate">METH_NOARGS</code>, just don’t specify
any arguments.
You can still use a self converter, a return converter, and specify
a <code class="docutils literal notranslate">type</code> argument to the object converter for <code class="docutils literal notranslate">METH_O</code>.
</div>
<div class="section" id="tp-new-and-tp-init-functions">
<h3>tp_new and tp_init functions<a class="headerlink" href="#tp-new-and-tp-init-functions" title="Permalink to this headline">¶</a></h3>
You can convert <code class="docutils literal notranslate">tp_new</code> and <code class="docutils literal notranslate">tp_init</code> functions. Just name
them <code class="docutils literal notranslate">__new__</code> or <code class="docutils literal notranslate">__init__</code> as appropriate. Notes:
<ul class="simple">
<li>The function name generated for <code class="docutils literal notranslate">__new__</code> doesn’t end in <code class="docutils literal notranslate">__new__</code>
like it would by default. It’s just the name of the class, converted
into a valid C identifier.</li>
<li>No <code class="docutils literal notranslate">PyMethodDef</code> <code class="docutils literal notranslate">#define</code> is generated for these functions.</li>
<li><code class="docutils literal notranslate">__init__</code> functions return <code class="docutils literal notranslate">int</code>, not <code class="docutils literal notranslate">PyObject *</code>.</li>
<li>Use the docstring as the class docstring.</li>
<li>Although <code class="docutils literal notranslate">__new__</code> and <code class="docutils literal notranslate">__init__</code> functions must always
accept both the <code class="docutils literal notranslate">args</code> and <code class="docutils literal notranslate">kwargs</code> objects, when converting
you may specify any signature for these functions that you like.
(If your function doesn’t support keywords, the parsing function
generated will throw an exception if it receives any.)</li>
</ul>
</div>
<div class="section" id="changing-and-redirecting-clinic-s-output">
<h3>Changing and redirecting Clinic’s output<a class="headerlink" href="#changing-and-redirecting-clinic-s-output" title="Permalink to this headline">¶</a></h3>
It can be inconvenient to have Clinic’s output interspersed with
your conventional hand-edited C code. Luckily, Clinic is configurable:
you can buffer up its output for printing later (or earlier!), or write
its output to a separate file. You can also add a prefix or suffix to
every line of Clinic’s generated output.
While changing Clinic’s output in this manner can be a boon to readability,
it may result in Clinic code using types before they are defined, or
your code attempting to use Clinic-generated code before it is defined.
These problems can be easily solved by rearranging the declarations in your file,
or moving where Clinic’s generated code goes. (This is why the default behavior
of Clinic is to output everything into the current block; while many people
consider this hampers readability, it will never require rearranging your
code to fix definition-before-use problems.)
Let’s start with defining some terminology:
<dl class="docutils">
<dt>field</dt>
<dd>A field, in this context, is a subsection of Clinic’s output.
For example, the <code class="docutils literal notranslate">#define</code> for the <code class="docutils literal notranslate">PyMethodDef</code> structure
is a field, called <code class="docutils literal notranslate">methoddef_define</code>. Clinic has seven
different fields it can output per function definition:
<div class="highlight-none notranslate"><div class="highlight"><pre>docstring_prototype
docstring_definition
methoddef_define
impl_prototype
parser_prototype
parser_definition
impl_definition
</pre></div>
</div>
All the names are of the form <code class="docutils literal notranslate">&quot;&lt;a&gt;_&lt;b&gt;&quot;</code>,
where <code class="docutils literal notranslate">&quot;&lt;a&gt;&quot;</code> is the semantic object represented (the parsing function,
the impl function, the docstring, or the methoddef structure) and <code class="docutils literal notranslate">&quot;&lt;b&gt;&quot;</code>
represents what kind of statement the field is. Field names that end in
<code class="docutils literal notranslate">&quot;_prototype&quot;</code>
represent forward declarations of that thing, without the actual body/data
of the thing; field names that end in <code class="docutils literal notranslate">&quot;_definition&quot;</code> represent the actual
definition of the thing, with the body/data of the thing. (<code class="docutils literal notranslate">&quot;methoddef&quot;</code>
is special, it’s the only one that ends with <code class="docutils literal notranslate">&quot;_define&quot;</code>, representing that
it’s a preprocessor #define.)
</dd>
<dt>destination</dt>
<dd>A destination is a place Clinic can write output to. There are
five built-in destinations:
<dl class="last docutils">
<dt><code class="docutils literal notranslate">block</code></dt>
<dd>The default destination: printed in the output section of
the current Clinic block.</dd>
<dt><code class="docutils literal notranslate">buffer</code></dt>
<dd>A text buffer where you can save text for later. Text sent
here is appended to the end of any existing text. It’s an
error to have any text left in the buffer when Clinic finishes
processing a file.</dd>
<dt><code class="docutils literal notranslate">file</code></dt>
<dd>A separate “clinic file” that will be created automatically by Clinic.
The filename chosen for the file is <code class="docutils literal notranslate">{basename}.clinic{extension}</code>,
where <code class="docutils literal notranslate">basename</code> and <code class="docutils literal notranslate">extension</code> were assigned the output
from <code class="docutils literal notranslate">os.path.splitext()</code> run on the current file. (Example:
the <code class="docutils literal notranslate">file</code> destination for <code class="docutils literal notranslate">_pickle.c</code> would be written to
<code class="docutils literal notranslate">_pickle.clinic.c</code>.)
Important: When using a <code class="docutils literal notranslate">file</code> destination, you
must check in the generated file!
</dd>
<dt><code class="docutils literal notranslate">two-pass</code></dt>
<dd>A buffer like <code class="docutils literal notranslate">buffer</code>. However, a two-pass buffer can only
be dumped once, and it prints out all text sent to it during
all processing, even from Clinic blocks after the dumping point.</dd>
<dt><code class="docutils literal notranslate">suppress</code></dt>
<dd>The text is suppressed—thrown away.</dd>
</dl>
</dd>
</dl>
Clinic defines five new directives that let you reconfigure its output.
The first new directive is <code class="docutils literal notranslate">dump</code>:
<div class="highlight-none notranslate"><div class="highlight"><pre>dump &lt;destination&gt;
</pre></div>
</div>
This dumps the current contents of the named destination into the output of
the current block, and empties it. This only works with <code class="docutils literal notranslate">buffer</code> and
<code class="docutils literal notranslate">two-pass</code> destinations.
The second new directive is <code class="docutils literal notranslate">output</code>. The most basic form of <code class="docutils literal notranslate">output</code>
is like this:
<div class="highlight-none notranslate"><div class="highlight"><pre>output &lt;field&gt; &lt;destination&gt;
</pre></div>
</div>
This tells Clinic to output field to destination. <code class="docutils literal notranslate">output</code> also
supports a special meta-destination, called <code class="docutils literal notranslate">everything</code>, which tells
Clinic to output all fields to that destination.
<code class="docutils literal notranslate">output</code> has a number of other functions:
<div class="highlight-none notranslate"><div class="highlight"><pre>output push
output pop
output preset &lt;preset&gt;
</pre></div>
</div>
<code class="docutils literal notranslate">output push</code> and <code class="docutils literal notranslate">output pop</code> allow you to push and pop
configurations on an internal configuration stack, so that you
can temporarily modify the output configuration, then easily restore
the previous configuration. Simply push before your change to save
the current configuration, then pop when you wish to restore the
previous configuration.
<code class="docutils literal notranslate">output preset</code> sets Clinic’s output to one of several built-in
preset configurations, as follows:
<blockquote>
<div><dl class="docutils">
<dt><code class="docutils literal notranslate">block</code></dt>
<dd>Clinic’s original starting configuration. Writes everything
immediately after the input block.
Suppress the <code class="docutils literal notranslate">parser_prototype</code>
and <code class="docutils literal notranslate">docstring_prototype</code>, write everything else to <code class="docutils literal notranslate">block</code>.
</dd>
<dt><code class="docutils literal notranslate">file</code></dt>
<dd>Designed to write everything to the “clinic file” that it can.
You then <code class="docutils literal notranslate">#include</code> this file near the top of your file.
You may need to rearrange your file to make this work, though
usually this just means creating forward declarations for various
<code class="docutils literal notranslate">typedef</code> and <code class="docutils literal notranslate">PyTypeObject</code> definitions.
Suppress the <code class="docutils literal notranslate">parser_prototype</code>
and <code class="docutils literal notranslate">docstring_prototype</code>, write the <code class="docutils literal notranslate">impl_definition</code> to
<code class="docutils literal notranslate">block</code>, and write everything else to <code class="docutils literal notranslate">file</code>.
The default filename is <code class="docutils literal notranslate">&quot;{dirname}/clinic/{basename}.h&quot;</code>.
</dd>
<dt><code class="docutils literal notranslate">buffer</code></dt>
<dd>Save up most of the output from Clinic, to be written into
your file near the end. For Python files implementing modules
or builtin types, it’s recommended that you dump the buffer
just above the static structures for your module or
builtin type; these are normally very near the end. Using
<code class="docutils literal notranslate">buffer</code> may require even more editing than <code class="docutils literal notranslate">file</code>, if
your file has static <code class="docutils literal notranslate">PyMethodDef</code> arrays defined in the
middle of the file.
Suppress the <code class="docutils literal notranslate">parser_prototype</code>, <code class="docutils literal notranslate">impl_prototype</code>,
and <code class="docutils literal notranslate">docstring_prototype</code>, write the <code class="docutils literal notranslate">impl_definition</code> to
<code class="docutils literal notranslate">block</code>, and write everything else to <code class="docutils literal notranslate">file</code>.
</dd>
<dt><code class="docutils literal notranslate">two-pass</code></dt>
<dd>Similar to the <code class="docutils literal notranslate">buffer</code> preset, but writes forward declarations to
the <code class="docutils literal notranslate">two-pass</code> buffer, and definitions to the <code class="docutils literal notranslate">buffer</code>.
This is similar to the <code class="docutils literal notranslate">buffer</code> preset, but may require
less editing than <code class="docutils literal notranslate">buffer</code>. Dump the <code class="docutils literal notranslate">two-pass</code> buffer
near the top of your file, and dump the <code class="docutils literal notranslate">buffer</code> near
the end just like you would when using the <code class="docutils literal notranslate">buffer</code> preset.
Suppresses the <code class="docutils literal notranslate">impl_prototype</code>, write the <code class="docutils literal notranslate">impl_definition</code>
to <code class="docutils literal notranslate">block</code>, write <code class="docutils literal notranslate">docstring_prototype</code>, <code class="docutils literal notranslate">methoddef_define</code>,
and <code class="docutils literal notranslate">parser_prototype</code> to <code class="docutils literal notranslate">two-pass</code>, write everything else
to <code class="docutils literal notranslate">buffer</code>.
</dd>
<dt><code class="docutils literal notranslate">partial-buffer</code></dt>
<dd>Similar to the <code class="docutils literal notranslate">buffer</code> preset, but writes more things to <code class="docutils literal notranslate">block</code>,
only writing the really big chunks of generated code to <code class="docutils literal notranslate">buffer</code>.
This avoids the definition-before-use problem of <code class="docutils literal notranslate">buffer</code> completely,
at the small cost of having slightly more stuff in the block’s output.
Dump the <code class="docutils literal notranslate">buffer</code> near the end, just like you would when using
the <code class="docutils literal notranslate">buffer</code> preset.
Suppresses the <code class="docutils literal notranslate">impl_prototype</code>, write the <code class="docutils literal notranslate">docstring_definition</code>
and <code class="docutils literal notranslate">parser_definition</code> to <code class="docutils literal notranslate">buffer</code>, write everything else to <code class="docutils literal notranslate">block</code>.
</dd>
</dl>
</div></blockquote>
The third new directive is <code class="docutils literal notranslate">destination</code>:
<div class="highlight-none notranslate"><div class="highlight"><pre>destination &lt;name&gt; &lt;command&gt; [...]
</pre></div>
</div>
This performs an operation on the destination named <code class="docutils literal notranslate">name</code>.
There are two defined subcommands: <code class="docutils literal notranslate">new</code> and <code class="docutils literal notranslate">clear</code>.
The <code class="docutils literal notranslate">new</code> subcommand works like this:
<div class="highlight-none notranslate"><div class="highlight"><pre>destination &lt;name&gt; new &lt;type&gt;
</pre></div>
</div>
This creates a new destination with name <code class="docutils literal notranslate">&lt;name&gt;</code> and type <code class="docutils literal notranslate">&lt;type&gt;</code>.
There are five destination types:
<blockquote>
<div><dl class="docutils">
<dt><code class="docutils literal notranslate">suppress</code></dt>
<dd>Throws the text away.</dd>
<dt><code class="docutils literal notranslate">block</code></dt>
<dd>Writes the text to the current block. This is what Clinic
originally did.</dd>
<dt><code class="docutils literal notranslate">buffer</code></dt>
<dd>A simple text buffer, like the “buffer” builtin destination above.</dd>
<dt><code class="docutils literal notranslate">file</code></dt>
<dd>A text file. The file destination takes an extra argument,
a template to use for building the filename, like so:
<blockquote>
<div>destination &lt;name&gt; new &lt;type&gt; &lt;file_template&gt;</div></blockquote>
The template can use three strings internally that will be replaced
by bits of the filename:
<blockquote>
<div><dl class="docutils">
<dt>{path}</dt>
<dd>The full path to the file, including directory and full filename.</dd>
<dt>{dirname}</dt>
<dd>The name of the directory the file is in.</dd>
<dt>{basename}</dt>
<dd>Just the name of the file, not including the directory.</dd>
<dt>{basename_root}</dt>
<dd>Basename with the extension clipped off
(everything up to but not including the last ‘.’).</dd>
<dt>{basename_extension}</dt>
<dd>The last ‘.’ and everything after it. If the basename
does not contain a period, this will be the empty string.</dd>
</dl>
</div></blockquote>
If there are no periods in the filename, {basename} and {filename}
are the same, and {extension} is empty. “{basename}{extension}”
is always exactly the same as “{filename}”.”
</dd>
<dt><code class="docutils literal notranslate">two-pass</code></dt>
<dd>A two-pass buffer, like the “two-pass” builtin destination above.</dd>
</dl>
</div></blockquote>
The <code class="docutils literal notranslate">clear</code> subcommand works like this:
<div class="highlight-none notranslate"><div class="highlight"><pre>destination &lt;name&gt; clear
</pre></div>
</div>
It removes all the accumulated text up to this point in the destination.
(I don’t know what you’d need this for, but I thought maybe it’d be
useful while someone’s experimenting.)
The fourth new directive is <code class="docutils literal notranslate">set</code>:
<div class="highlight-none notranslate"><div class="highlight"><pre>set line_prefix &quot;string&quot;
set line_suffix &quot;string&quot;
</pre></div>
</div>
<code class="docutils literal notranslate">set</code> lets you set two internal variables in Clinic.
<code class="docutils literal notranslate">line_prefix</code> is a string that will be prepended to every line of Clinic’s output;
<code class="docutils literal notranslate">line_suffix</code> is a string that will be appended to every line of Clinic’s output.
Both of these support two format strings:
<blockquote>
<div><dl class="docutils">
<dt><code class="docutils literal notranslate">{block comment start}</code></dt>
<dd>Turns into the string <code class="docutils literal notranslate">/*</code>, the start-comment text sequence for C files.</dd>
<dt><code class="docutils literal notranslate">{block comment end}</code></dt>
<dd>Turns into the string <code class="docutils literal notranslate">*/</code>, the end-comment text sequence for C files.</dd>
</dl>
</div></blockquote>
The final new directive is one you shouldn’t need to use directly,
called <code class="docutils literal notranslate">preserve</code>:
<div class="highlight-none notranslate"><div class="highlight"><pre>preserve
</pre></div>
</div>
This tells Clinic that the current contents of the output should be kept, unmodified.
This is used internally by Clinic when dumping output into <code class="docutils literal notranslate">file</code> files; wrapping
it in a Clinic block lets Clinic use its existing checksum functionality to ensure
the file was not modified by hand before it gets overwritten.
</div>
<div class="section" id="the-ifdef-trick">
<h3>The #ifdef trick<a class="headerlink" href="#the-ifdef-trick" title="Permalink to this headline">¶</a></h3>
If you’re converting a function that isn’t available on all platforms,
there’s a trick you can use to make life a little easier. The existing
code probably looks like this:
<div class="highlight-c notranslate"><div class="highlight"><pre>#ifdef HAVE_FUNCTIONNAME
static module_functionname(...)
{
...
}
#endif /* HAVE_FUNCTIONNAME */
</pre></div>
</div>
And then in the <code class="docutils literal notranslate">PyMethodDef</code> structure at the bottom the existing code
will have:
<div class="highlight-none notranslate"><div class="highlight"><pre>#ifdef HAVE_FUNCTIONNAME
{&#39;functionname&#39;, ... },
#endif /* HAVE_FUNCTIONNAME */
</pre></div>
</div>
In this scenario, you should enclose the body of your impl function inside the <code class="docutils literal notranslate">#ifdef</code>,
like so:
<div class="highlight-c notranslate"><div class="highlight"><pre>#ifdef HAVE_FUNCTIONNAME
/*[clinic input]
module.functionname
...
[clinic start generated code]*/
static module_functionname(...)
{
...
}
#endif /* HAVE_FUNCTIONNAME */
</pre></div>
</div>
Then, remove those three lines from the <code class="docutils literal notranslate">PyMethodDef</code> structure,
replacing them with the macro Argument Clinic generated:
<div class="highlight-none notranslate"><div class="highlight"><pre>MODULE_FUNCTIONNAME_METHODDEF
</pre></div>
</div>
(You can find the real name for this macro inside the generated code.
Or you can calculate it yourself: it’s the name of your function as defined
on the first line of your block, but with periods changed to underscores,
uppercased, and <code class="docutils literal notranslate">&quot;_METHODDEF&quot;</code> added to the end.)
Perhaps you’re wondering: what if <code class="docutils literal notranslate">HAVE_FUNCTIONNAME</code> isn’t defined?
The <code class="docutils literal notranslate">MODULE_FUNCTIONNAME_METHODDEF</code> macro won’t be defined either!
Here’s where Argument Clinic gets very clever. It actually detects that the
Argument Clinic block might be deactivated by the <code class="docutils literal notranslate">#ifdef</code>. When that
happens, it generates a little extra code that looks like this:
<div class="highlight-c notranslate"><div class="highlight"><pre>#ifndef MODULE_FUNCTIONNAME_METHODDEF
 #define MODULE_FUNCTIONNAME_METHODDEF
#endif /* !defined(MODULE_FUNCTIONNAME_METHODDEF) */
</pre></div>
</div>
That means the macro always works. If the function is defined, this turns
into the correct structure, including the trailing comma. If the function is
undefined, this turns into nothing.
However, this causes one ticklish problem: where should Argument Clinic put this
extra code when using the “block” output preset? It can’t go in the output block,
because that could be deactivated by the <code class="docutils literal notranslate">#ifdef</code>. (That’s the whole point!)
In this situation, Argument Clinic writes the extra code to the “buffer” destination.
This may mean that you get a complaint from Argument Clinic:
<div class="highlight-none notranslate"><div class="highlight"><pre>Warning in file &quot;Modules/posixmodule.c&quot; on line 12357:
Destination buffer &#39;buffer&#39; not empty at end of file, emptying.
</pre></div>
</div>
When this happens, just open your file, find the <code class="docutils literal notranslate">dump buffer</code> block that
Argument Clinic added to your file (it’ll be at the very bottom), then
move it above the <code class="docutils literal notranslate">PyMethodDef</code> structure where that macro is used.
</div>
<div class="section" id="using-argument-clinic-in-python-files">
<h3>Using Argument Clinic in Python files<a class="headerlink" href="#using-argument-clinic-in-python-files" title="Permalink to this headline">¶</a></h3>
It’s actually possible to use Argument Clinic to preprocess Python files.
There’s no point to using Argument Clinic blocks, of course, as the output
wouldn’t make any sense to the Python interpreter. But using Argument Clinic
to run Python blocks lets you use Python as a Python preprocessor!
Since Python comments are different from C comments, Argument Clinic
blocks embedded in Python files look slightly different. They look like this:
<div class="highlight-python3 notranslate"><div class="highlight"><pre>#/*[python input]
#print(&quot;def foo(): pass&quot;)
#[python start generated code]*/
def foo(): pass
#/*[python checksum:...]*/
</pre></div>
</div>
</div>
</div>
</div>

</div>
 </div>
 </div>
 <div class="sphinxsidebar" role="navigation" aria-label="main navigation">
 <div class="sphinxsidebarwrapper">
 <h3><a href="../contents.html">Table Of Contents</a></h3>
 <ul>
<li><a class="reference internal" href="#">Argument Clinic How-To</a><ul>
<li><a class="reference internal" href="#the-goals-of-argument-clinic">The Goals Of Argument Clinic</a></li>
<li><a class="reference internal" href="#basic-concepts-and-usage">Basic Concepts And Usage</a></li>
<li><a class="reference internal" href="#converting-your-first-function">Converting Your First Function</a></li>
<li><a class="reference internal" href="#advanced-topics">Advanced Topics</a><ul>
<li><a class="reference internal" href="#symbolic-default-values">Symbolic default values</a></li>
<li><a class="reference internal" href="#renaming-the-c-functions-and-variables-generated-by-argument-clinic">Renaming the C functions and variables generated by Argument Clinic</a></li>
<li><a class="reference internal" href="#converting-functions-using-pyarg-unpacktuple">Converting functions using PyArg_UnpackTuple</a></li>
<li><a class="reference internal" href="#optional-groups">Optional Groups</a></li>
<li><a class="reference internal" href="#using-real-argument-clinic-converters-instead-of-legacy-converters">Using real Argument Clinic converters, instead of “legacy converters”</a></li>
<li><a class="reference internal" href="#py-buffer">Py_buffer</a></li>
<li><a class="reference internal" href="#advanced-converters">Advanced converters</a></li>
<li><a class="reference internal" href="#parameter-default-values">Parameter default values</a></li>
<li><a class="reference internal" href="#the-null-default-value">The <code class="docutils literal notranslate">NULL</code> default value</a></li>
<li><a class="reference internal" href="#expressions-specified-as-default-values">Expressions specified as default values</a></li>
<li><a class="reference internal" href="#using-a-return-converter">Using a return converter</a></li>
<li><a class="reference internal" href="#cloning-existing-functions">Cloning existing functions</a></li>
<li><a class="reference internal" href="#calling-python-code">Calling Python code</a></li>
<li><a class="reference internal" href="#using-a-self-converter">Using a “self converter”</a></li>
<li><a class="reference internal" href="#writing-a-custom-converter">Writing a custom converter</a></li>
<li><a class="reference internal" href="#writing-a-custom-return-converter">Writing a custom return converter</a></li>
<li><a class="reference internal" href="#meth-o-and-meth-noargs">METH_O and METH_NOARGS</a></li>
<li><a class="reference internal" href="#tp-new-and-tp-init-functions">tp_new and tp_init functions</a></li>
<li><a class="reference internal" href="#changing-and-redirecting-clinic-s-output">Changing and redirecting Clinic’s output</a></li>
<li><a class="reference internal" href="#the-ifdef-trick">The #ifdef trick</a></li>
<li><a class="reference internal" href="#using-argument-clinic-in-python-files">Using Argument Clinic in Python files</a></li>
</ul>
</li>
</ul>
</li>
</ul>

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

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

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

</body>
</html>

MR SKB