hybrid programming based on python and C/C++
Python/C API Reference Manual¶
https://docs.python.org/3/c-api/index.html
Extending and Embedding the Python Interpreter¶
https://docs.python.org/3/extending/
Abstract:
Reference counting:
1) Stealing a reference means that when you pass a reference to a function, that function assumes that it now owns that reference,you are not responsible for it any longer;
2) When no ownership is transferred, the caller is said to borrow the reference. Nothing needs to be done for a borrowed reference.
3) in many cases, the returned object is created on the fly, and the reference you get is the only reference to the object.
4) A borrowed reference can be changed into an owned reference by calling Py_INCREF() (create a new owned one)
Details:
Most functions that return a reference to an object pass on ownership with the reference. In particular, all functions whose function it is to create a new object,
such as PyLong_FromLong() and Py_BuildValue(), pass ownership to the receiver. Even if the object is not actually new, you still receive ownership of a new reference to that object.
Many functions that extract objects from other objects also Transfer ownership with the reference, e.g. PyObject_GetAttrString(), which means the extracted obj's ownship transfered
but therr are Eexceptions: PyTuple_GetItem(), PyList_GetItem(), PyDict_GetItem(), and PyDict_GetItemString() all return references that you borrow from the tuple, list or dictionary.
When you pass an object reference into another function, in general, the function borrows the reference from you — if it needs to store it,
it will use Py_INCREF() to become a independent owner. There are exactly two important exceptions to this rule: PyTuple_SetItem() and PyList_SetItem().
These functions take over ownership of the item passed to them — even if they fail!
Extending and Embedding the Python Interpreter
Extend Python Interpreter with c++ : C dynamic lib called in Python script
Embedding Python interpreter in C : invoking Python script in C execuatable
1) an C extension for python is not compatible among different implementations of Python, such as CPython ..
2) you must include Python.h before any standard headers are included.
3) In order to avoid name clashes with other extension modules, all symbols should be static;
4) and symbole which should be accessible from other extension modules must be exported in a different way, that is Capsules machinism
Exceptions:
Exceptions are stored in a static global variable inside the interpreter
An important convention is : when a function fails, it should set an exception condition and return an error value
Set an exception including : exception, associated value, traceback
How to set in extension ? use PyErr_SetXXX or create an error object and then add it to the module manually
Others:
1) Different Python versions contain different lib packages(or syntax), six AND python-future are two packages to provide compatibility layer
2) pip introduce the idea of Requirements files, a list of pip install arguments
3) To support cyclic garbage collection, flag in the type object should be enabled
4) use "pythonX.Y-config --cflags" will give you the recommended flags when compiling/linking
Types:
1) Python runtime sees all Python objects as variables of type PyObject* (base type, only contains the refcount and a pointer to the object’s “type obobject")
2) Tyep object of an Object determine the Object's action(which function get called)
Notes: the follow code was from https://docs.python.org/3/extending/embedding.html
1 embed python sript in C language:
#!/usr/bin/python3
# sample2.pydef func2():
print ("hello world") def func3(s):
print ("hello ", s) class student:
def SetName(self, name):
self._name = name;
def PrintName(self):
print(self._name) # should not run func when this was called by C excutable
# func(
#include <Python.h>
#include <iostream>
#include<string>
using namespace std; void funct()
{
// before call python, initialize with following func first
Py_Initialize();
PyRun_SimpleString("import sys");
PyRun_SimpleString("sys.path.append('./')"); PyObject * pModule = NULL;
PyObject * pFunc = NULL;
PyObject * pClass = NULL;
PyObject * pInstance = NULL; pModule = PyImport_ImportModule("sample2");
if ( pModule == NULL)
printf("dddddddddddddddddddd");
pFunc = PyObject_GetAttrString(pModule, "func2");
PyEval_CallObject(pFunc, NULL);
Py_DECREF(pFunc); pFunc = PyObject_GetAttrString(pModule, "func3");
PyObject_CallFunction(pFunc, "s", "zhengji");
Py_DECREF(pFunc); pClass = PyObject_GetAttrString(pModule, "student");
if (!pClass)
{
cout << "can't find student instance\n";
return ;
} pInstance = PyObject_CallObject(pClass, NULL);
//PyClass
if (!pInstance)
{
cout << "can't create student instance\n";
return;
} PyObject_CallMethod(pInstance, "SetName", "s", "my family");
PyObject_CallMethod(pInstance, "PrintName", NULL, NULL); Py_Finalize();
} int main()
{
funct();
return 0;
}
build it : g++ embedone.cpp -o embed -I/usr/include/python3.4m -lpython3.4m
2 build c/c++ extension for Python:
// file spammodule1.cpp
#include <Python.h>
/* target
>>> import spam
>>> status = spam.system("ls -l")
*/ #ifdef __cplusplus
extern "C" {
#endif // all variable and function should be static
static PyObject *SpamError; // C function to be called
int function(const char * command)
{
return system(command);
} // wrapper for C function
static PyObject *spam_system(PyObject *self, PyObject *args)
{
const char *command;
int sts; if (!PyArg_ParseTuple(args, "s", &command))
return NULL;
sts = system(command);
if (sts < )
{
PyErr_SetString(SpamError, "System command failed");
return NULL;
}
return PyLong_FromLong(sts);
// if return void , use:
// Py_INCREF(Py_None);
// return Py_None;
} // list its name and address in a method table
static PyMethodDef SpamMethods[] = {
{ "system", spam_system, METH_VARARGS,
"Execute a shell command." },
{ NULL, NULL, , NULL } /* Sentinel */
}; // put method talbe in module definition structure
static struct PyModuleDef spammodule = {
PyModuleDef_HEAD_INIT,
"spam", /* name of module */
"spam_doc, which intoduce...", /* module documentation, may be NULL */
-, /* size of per-interpreter state of the module, or -1 if the module keeps state in global variables. */
SpamMethods
}; // The initialization function must be named in the format PyInit_name()
// The init function must return the module object to its caller, so that it then gets inserted into sys.modules.
PyMODINIT_FUNC PyInit_spam(void)
{
PyObject *m; m = PyModule_Create(&spammodule);
if (m == NULL)
{
SpamError = PyErr_NewException("spam.error", NULL, NULL);
Py_INCREF(SpamError);
PyModule_AddObject(m, "error", SpamError);
return NULL;
}
return m;
} #ifdef __cplusplus
}
#endif /* // if this module is called in another C code (enbedding Python), use PyImport_AppendInittab() before it's imported
int function(int argc, char *argv[])
{
wchar_t *program = Py_DecodeLocale(argv[0], NULL);
if (program == NULL)
{
fprintf(stderr, "Fatal error: cannot decode argv[0]\n");
exit(1);
}
// Add a built-in module, before Py_Initialize
PyImport_AppendInittab("spam", PyInit_spam);
// Pass argv[0] to the Python interpreter ,inform the interpreter about paths to Python run-time libraries
Py_SetProgramName(program);
// Initialize the Python interpreter. Required.
Py_Initialize();
// Optionally import the module; alternatively, import can be deferred until the embedded script imports it.
PyImport_ImportModule("spam");
// program was mallocated by Python, so free it with Py_free
PyMem_RawFree(program);
return 0;
} */
// http://www.swig.org/papers/PyTutorial98/PyTutorial98.pdf
build it : g++ spammodule1.cpp -o spam.so -shared -fpic -I/usr/include/python3.4m
or create file setup.py and run : python3 setup.py build
#!/usr/bin/python3 from distutils.core import setup, Extension # Extention is a collection of attributes/every thing that needed to build a module
# init(self, name,sources, include_dirs=NONE, libraries, arguments, macros...)
module1 = Extension('spam',sources=['spammodule1.cpp']) # https://docs.python.org/3.4/distutils/apiref.html?highlight=setup#distutils.core.setup
# name : package name
# packages : A list of Python packages that distutils will manipulate, contain .py files
# py_modules : A list of Python modules that distutils will manipulate, usually .py files
# ext_modules : a list of instances of distutils.core.Extension to be built (module)
setup (name = 'spam1',version = '1.0', description = 'This is a demo package',ext_modules = [module1])
#!/usr/bin/python3
# test.py
import spam
spam.system("ls -a")
3 extended embedding python
// file : pythonembedding.cpp
#include <Python.h>
#include <iostream>
using namespace std; /*
code are copied from https://docs.python.org/3/extending/embedding.html
extend Embedded Python, which means
python extension: accept python objects, funcs are called in python script
embeded python : call python function with c objects
*/
static int numargs = ; /* Return the number of arguments of the application command line */
static PyObject* emb_numargs(PyObject *self, PyObject *args)
{
if (!PyArg_ParseTuple(args, ":numargs"))
return NULL;
return PyLong_FromLong(numargs);
} static PyMethodDef EmbMethods[] = {
{ "numargs", emb_numargs, METH_VARARGS,
"Return the number of arguments received by the process." },
{ NULL, NULL, , NULL }
}; static PyModuleDef EmbModule = {
PyModuleDef_HEAD_INIT, "emb", NULL, -, EmbMethods,
NULL, NULL, NULL, NULL
}; static PyObject* PyInit_emb(void)
{
return PyModule_Create(&EmbModule);
} int main(int argc, char *argv[])
{
PyObject *pName, *pModule, *pDict, *pFunc;
PyObject *pArgs, *pValue;
int i; if (argc < )
{
fprintf(stderr, "Usage: call pythonfile funcname [args]\n");
fprintf(stderr, "examp: ./xx sample getname dd\n");
return ;
} numargs = argc;
PyImport_AppendInittab("emb", &PyInit_emb); Py_Initialize(); // add current path so it can import module in current directory
PyRun_SimpleString("import sys");
PyRun_SimpleString("sys.path.append('./')"); pName = PyUnicode_DecodeFSDefault(argv[]);
/* Error checking of pName left out */ pModule = PyImport_Import(pName);
Py_DECREF(pName); if (pModule != NULL)
{
pFunc = PyObject_GetAttrString(pModule, argv[]);
/* pFunc is a new reference */ if (pFunc && PyCallable_Check(pFunc))
{
pArgs = PyTuple_New(argc - );
for (i = ; i < argc - ; ++i)
{
pValue = PyLong_FromLong(atoi(argv[i + ]));
if (!pValue)
{
Py_DECREF(pArgs);
Py_DECREF(pModule);
fprintf(stderr, "Cannot convert argument\n");
return ;
}
/* pValue reference stolen here: */
PyTuple_SetItem(pArgs, i, pValue);
}
pValue = PyObject_CallObject(pFunc, pArgs);
Py_DECREF(pArgs);
if (pValue != NULL)
{
printf("Result of call: %ld\n", PyLong_AsLong(pValue));
Py_DECREF(pValue);
}
else
{
Py_DECREF(pFunc);
Py_DECREF(pModule);
PyErr_Print();
fprintf(stderr, "Call failed\n");
return ;
}
}
else
{
if (PyErr_Occurred())
PyErr_Print();
fprintf(stderr, "Cannot find function \"%s\"\n", argv[]);
}
Py_XDECREF(pFunc);
Py_DECREF(pModule);
}
else
{
PyErr_Print();
fprintf(stderr, "Failed to load \"%s\"\n", argv[]);
return ;
}
// included in version 3.6
// if (Py_FinalizeEx() < 0) { return 120; }
Py_Finalize();
return ;
}
#!/usr/bin/python3
# sample.py
import emb
print("number of arguments", emb.numargs())
def func(x):
print(" the imput arguments is ", x )
sample.py invokes module "emb" defined in pythonembedding.cpp, meanwhile this python file (module) will be invoked in pythonembedding
build it : g++ pythonembedding.cpp -o pythonembedding -I/usr/include/python3.4m/ -lpython3.4m
run : ./pythonembedding
hybrid programming based on python and C/C++的更多相关文章
- Most simple basic of internet programming based on two different machines sharing the same local net
This blog is just shown for the most simple basic of internet programming based on two different mac ...
- 2018.8.10 programming bat based on python
@echo off REM Current DevProg Version. Match the pip package version (x.y.z)SET currentversion=0.4.0 ...
- Python Socket Programming
本文介绍使用Python进行Socket网络编程,假设读者已经具备了基本的网络编程知识和Python的基本语法知识,本文中的代码如果没有说明则都是运行在Python 3.4下. Python的sock ...
- 使用Boost.Python构建混合系统(译)
目录 Building Hybrid Systems with Boost.Python 摘要(Abstract) 介绍(Introduction) 设计目标 (Boost.Python Design ...
- Introduction of python
"Life is short, you need Python!" Python (British pronunciation:/ˈpaɪθən/ American pronunc ...
- python代码优化技巧
转自:http://www.douban.com/group/topic/31478102/ 这个资料库还有些不错的好文章: http://www.ibm.com/developerworks/cn/ ...
- How To Crawl A Web Page with Scrapy and Python 3
sklearn实战-乳腺癌细胞数据挖掘(博主亲自录制视频) https://study.163.com/course/introduction.htm?courseId=1005269003& ...
- Coursera课程《Python数据结构》中课程目录
Python Data Structures Python Data Structures is the second course in the specialization Python for ...
- Coursera课程《大家的编程》(Python入门)中课程目录
Getting Started with Python Getting Started with Python is the first course in the specialization Py ...
随机推荐
- HDU 1035(走迷宫 模拟)
题意是给定初始位置在一个迷宫中按照要求前进,判断多少步能离开迷宫或者多少步会走入一个长达多少步的循环. 按要求模拟前进的位置,对每一步在 vis[ ] 数组中进行已走步数的记录,走出去或走到已走过的位 ...
- linux常用命令说明
1. 基础快捷键使用 ctrl + c:停止进程 ctrl + l:清屏 ctrl + r:搜索历史命令 ctrl + q:退出 2.常用目录或文件夹命令 说明:安装linux时,创建一个itcast ...
- 转---redshift database ---学习
摘自他人 前沿 根据最近一段时间对redshift的研究,发现一些特性比较适合我们当前的业务. 1 比如它的快速恢复能力,因为这一点,我们可以尽量在redshit里面存放一定生命周期的数据,对过期的数 ...
- 向量空间模型(Vector Space Model)的理解
1. 问题描述 给你若干篇文档,找出这些文档中最相似的两篇文档? 相似性,可以用距离来衡量.而在数学上,可使用余弦来计算两个向量的距离. \[cos(\vec a, \vec b)=\frac {\v ...
- Error:Failed to resolve: com.android.support:recyclerview-v7:26.1.0
修改gradle allprojects { repositories { maven { url "https://maven.google.com" } jcenter() } ...
- Android弹出窗口
protected void PopUp() { final PopupWindow popup = new PopupWindow(TestActivity.this); View popView ...
- SQLite 数据库介绍和基本用法
Ø 简介 SQLite 是一款轻量级的关系型数据库,同时也是一种嵌入式数据库,与 Oracle.MySQL.SQL Server 等数据库不同,它可以内嵌在程序中,是程序中的一个组成部分.所以,经常 ...
- 什么是CRUD
CRUD是指在做计算处理时的增加(Create).读取查询(Retrieve).更新(Update)和删除(Delete)几个单词的首字母简写.主要被用在描述软件系统中数据库或者持久层的基本操作功能.
- [译]kendoui - 方法和事件
原文 为了使用方法和事件,首先要获取到widget实例. 获取widget 一共有3种获取widget的方式. jQuery的data方法 将widget的名作为参数传给jQuery的data方法.( ...
- 本地测试使用Tomcat,生产环境使用GlassFish。
总结:Tomcat8 = javaee7规范(servlet3.1 + jsp2.3 + el3.0 + websocket1.0) + java7 [配置初始化参数使用jdk8编译]conf/web ...