blocks|key|8621|text|看起来你有一个引用计数问题。您需要保留对$input的引用，即使在std::function类型映射完成之后也是如此。否则，一旦完成对registerFunc的调用，它就会丢失一个引用。要做到这一点，最简单的方法是让您的typemap捕获一个std::shared_ptr而不是原始的PyObject，例如：|type|unstyled|depth|inlineStyleRanges|offset|length|style|CODE|entityRanges|data|8622|%25typemap(in)+std::function+{
++++Py_INCREF($input);
++++static+const+auto+decref+=+[](PyObject+*o)+{
++++++++Py_DECREF(o);+//+This+needs+to+be+another+lambda/function+because+Py_DECREF+is+really+a+macro
++++};
++++std::shared_ptr<PyObject>+callable($input,+decref);
++++auto+callback+=+[callable](auto&&...+params)+{
+++++++PyGILState_STATE+state+=+PyGILState_Ensure();
++++++++++++++++++++++++++++++++++++++++++++++++++++++//+Back+to+raw+PyObject++++
+++++++PyObject*+result+=++PyObject_CallFunctionObjArgs(callable.get(),makePyObject(std::forward<decltype(params)>(params))...,+NULL);
+++++++int+retVal+=+-1;

+++++++if+(result)
+++++++{
+++++++++++retVal+=+PyObject_IsTrue(result);
+++++++++++Py_DECREF(result);
+++++++}

+++++++PyGILState_Release(state);
+++++++return+retVal+==+1;
+++};
+++$1+=+std::move(callback);
}|code-block|syntax|javascript|8623|理想情况下，我会使用std::unique_ptr，但即使使用generalised+lambda+captures，它也会阻止复制构造，并强制SWIG在不需要的情况下生成代码，这是一项更多的工作。|8624|不过，我可能也会在类型映射中至少抛出一个PyCallable_Check。|8625|entityMap|0|LINK|mutability|MUTABLE|url|https://stackoverflow.com/a/16968463/168175^0|K|6|X|D|1V|C|3D|F|3Y|8|0|0|A|F|V|R|0|0|K|G|0^^$0|@$1|2|3|4|5|6|7|W|8|@$9|X|A|Y|B|C]|$9|Z|A|10|B|C]|$9|11|A|12|B|C]|$9|13|A|14|B|C]|$9|15|A|16|B|C]]|D|@]|E|$]]|$1|F|3|G|5|H|7|17|8|@]|D|@]|E|$I|J]]|$1|K|3|L|5|6|7|18|8|@$9|19|A|1A|B|C]]|D|@$9|1B|A|1C|1|1D]]|E|$]]|$1|M|3|N|5|6|7|1E|8|@$9|1F|A|1G|B|C]]|D|@]|E|$]]|$1|O|3|-4|5|6|7|1H|8|@]|D|@]|E|$]]]|P|$Q|$5|R|S|T|E|$U|V]]]]

Looks like you have a reference counting problem. You need to retain a reference to <code>$input</code>, even after the <code>std::function</code> typemap has completed. Otherwise it's going to be missing a reference once the call to <code>registerFunc</code> is done. The simplest way to do this is to make your typemap capture a <code>std::shared_ptr</code> instead of the raw <code>PyObject</code>, e.g.:

<pre><code>%typemap(in) std::function {
 Py_INCREF($input);
 static const auto decref = [](PyObject *o) {
 Py_DECREF(o); // This needs to be another lambda/function because Py_DECREF is really a macro
 };
 std::shared_ptr&lt;PyObject&gt; callable($input, decref);
 auto callback = [callable](auto&amp;&amp;... params) {
 PyGILState_STATE state = PyGILState_Ensure();
 // Back to raw PyObject 
 PyObject* result = PyObject_CallFunctionObjArgs(callable.get(),makePyObject(std::forward&lt;decltype(params)&gt;(params))..., NULL);
 int retVal = -1;

 if (result)
 {
 retVal = PyObject_IsTrue(result);
 Py_DECREF(result);
 }

 PyGILState_Release(state);
 return retVal == 1;
 };
 $1 = std::move(callback);
}
</code></pre>

I would have used a <code>std::unique_ptr</code> instead ideally, but even with <a href="https://stackoverflow.com/a/16968463/168175">generalised lambda captures</a> that stops copy construction from being possible and forcing SWIG to generate code without needing that is a bit more work.

I'd probably throw at least a <code>PyCallable_Check</code> in the typemap for good measure though too.

I have written a Python Wrapper for a C++14 library using SWIG.
Within the C++ API I can register std::functions as callbacks.

I have a SWIG typemap for std::function's to pass a lambda expression
which invokes the Python callback:

<pre><code>%typemap(in) std::function {
 auto callback = [$input](auto&amp;&amp;... params) {
 PyGILState_STATE state = PyGILState_Ensure();

 PyObject* result = PyObject_CallFunctionObjArgs($input,makePyObject(std::forward&lt;decltype(params)&gt;(params))..., NULL);
 const int retVal = PyObject_IsTrue(result);

 Py_DECREF(result);
 PyGILState_Release(state);
 return retVal == 1;
 };
 $1 = std::move(callback);
}
</code></pre>

When I run a test script, the following Python expression works
fine:

<pre><code>callback = lambda a,b: self.doStuff(a,b)
self.cppInterface.registerFunc(callback)
</code></pre>

This expression however does not work:

<pre><code>self.cppInterface.registerFunc(lambda a,b: self.doStuff)
</code></pre>

When I pass the lambda directly to the register function,
I get a the following error when the callback is called from C++:

<pre><code>TypeError: 'managedbuffer' object is not callable
</code></pre>

Why is the PyObject $input not a callable?
How do I allow both Python expressions?

Example code:

<a href="https://github.com/nullmedium/python-swig-demo" rel="nofollow noreferrer">https://github.com/nullmedium/python-swig-demo</a>

How to convert a 'managedbuffer' into a Callable in Python

翻译质量差，导致语言生硬或混乱。

没有提供实际的解决方法或示例。

解答不清晰，无法理解或解决问题。

页面排版不美观，阅读体验差。

文章

问答

视频

学习中心

腾讯云实验室

直播

竞赛

腾讯云代码分析专区

腾讯iOA零信任安全管理系统专区

腾讯云架构师技术同盟交流圈

腾讯云数据库专区

腾讯云顾问专区

腾讯云原生专区

腾讯混元专区

腾讯云TCE专区

腾讯云Lighthouse专区

腾讯云HAI专区

腾讯云Edgeone专区

腾讯云存储专区

腾讯云智能专区

腾讯轻联专区 

腾讯云开发专区

TAPD专区

腾讯轻量云游戏服专区

腾讯云最具价值专家

腾讯云架构师技术同盟

腾讯云创作之星

腾讯云开发者先锋

腾讯云代码助手

云原生构建

TAPD 敏捷项目管理

Cloud Studio

SDK中心

API中心

命令行工具

涵盖代码开发、场景应用、自动测试全流程，助你从零构建专属AI助手

一站式MCP教程库，解锁AI应用新玩法

我已经使用SWIG为C++14库编写了一个Python包装器。在std：：C++中，我可以将std：：函数注册为回调函数。我为std：：function提供了一个SWIG类型映射来传递一个调用Python回调的lambda表达式：%typemap(in) std::function {    auto callback...

如何在Python中将'managedbuffer‘转换为可调用的-腾讯云开发者社区-腾讯云

问如何在Python中将'managedbuffer‘转换为可调用的
EN

社区

活动

圈层

关于

腾讯云开发者

热门产品

热门推荐

更多推荐

问如何在Python中将'managedbuffer‘转换为可调用的EN

社区

活动

圈层

关于

腾讯云开发者

热门产品

热门推荐

更多推荐

问如何在Python中将'managedbuffer‘转换为可调用的
EN