Develop Reference

Embed Python in C++ (using CMake)

I'm trying to run a python script in c++. For example:
// main.cpp
#include <python3.10/Python.h>
int main(int argc, char* argv[])
{
Py_Initialize();
PyRun_SimpleString("from time import time,ctime\n"
"print('Today is',ctime(time()))\n");
Py_Finalize();
return 0;
}
And I have such CMakeLists:
// CMakeLists.txt
cmake_minimum_required(VERSION 3.21)
project(task_01)
SET(CMAKE_CXX_FLAGS "-O0")
SET(CMAKE_C_FLAGS "-O0")
set(CMAKE_CXX_STANDARD 20)
find_package (Python COMPONENTS Interpreter Development)
find_package(PythonLibs 3.10 REQUIRED)
include_directories ( ${PYTHON_INCLUDE_DIRS} )
add_executable(task_01 main.cpp)
But I get compile errors (I use CLion IDE):
undefined reference to `Py_Initialize'
undefined reference to `PyRun_SimpleStringFlags'
undefined reference to `Py_Finalize'
What am I doing wrong to run a python script?

Prefer imported targets (https://cmake.org/cmake/help/latest/module/FindPython.html#imported-targets):
cmake_minimum_required(VERSION 3.18)
project(task_01)
find_package(Python REQUIRED Development)
add_executable(task_01 main.cpp Utility.cpp Sort.cpp)
target_link_libraries(task_01 PRIVATE Python::Python)

Such configurations in CMakeLists.txt helped me:
cmake_minimum_required(VERSION 3.21)
project(task_01)
SET(CMAKE_CXX_FLAGS "-O0")
SET(CMAKE_C_FLAGS "-O0")
set(CMAKE_CXX_STANDARD 20)
find_package (Python REQUIRED
COMPONENTS Interpreter Development)
add_executable(task_01 main.cpp Utility.h Utility.cpp Sort.h Sort.cpp)
target_include_directories(${PROJECT_NAME} PRIVATE
${Python_INCLUDE_DIRS})
target_link_directories(${PROJECT_NAME} PRIVATE
${Python_LIBRARY_DIRS})
target_link_libraries(${PROJECT_NAME} PRIVATE
${Python_LIBRARIES})

How to link python.h and matplotlibcpp.h in Visual Studio 2019?

From a related post, one installed package matplotlib-cpp through microsoft /
vcpkg . However, when running a simple example,
#include "matplotlibcpp.h"
namespace plt = matplotlibcpp;
int main() {
plt::plot({1,3,2,4});
plt::show();
}
an error returned stating that
Error C1083 Cannot open include file: 'Python.h': No such file or directory Examples
..\vcpkg\installed\x64-windows\include\matplotlibcpp.h 5
I have tried the methods in:
can't include Python.h in visual studio
psycopg: Python.h: No such file or directory and Cygwin gcc issue - cannot find Python.h (which didn't apply since vs managed python environment itself, and there wasn't python-dev option)
A very similiar post C++: matplotlibcpp.h and Python.h linker error also indicated the same issue in Ubuntu environment.
In matplotlibcpp.h file, it mentioned that
// Python headers must be included before any system headers, since
// they define _POSIX_C_SOURCE
#include <Python.h>
What did it mean? How to link python.h and a matplotlibcpp.h in Visual Studio 2019?
Updates:
I was able to locate "Python.h" and "numpy/arrayobject.h" in MS python folder. However, when I manually added those three ".h" files into the project header folder, the issue wasn't resolved. Especially, when I manually fixed the locations in "matplotlibcpp.h", the "Python.h" in ""numpy/arrayobject.h" still could not be located.

Matplotlib-cpp : ImportError: No module named site

I am trying to use matplotlib-cpp for a C++ project compiled by Visual Studio using CMake.
The python distribution I use is Anaconda2 (latest 2.7 version downloaded today). I removed every other python distribution that was on my computer.
I added the paths to the Anaconda2 folder to the system AND user environment variables. (C:\Anaconda2...)
CMake is finding it correctly as when configuring the project with CMake, I have:
Found PythonInterp: C:/Anaconda2/python.exe (found suitable version "2.7.16", minimum required is "2.7")
Found PythonLibs: C:/Anaconda2/libs/python27.lib (found suitable version "2.7.16", minimum required is "2.7")
The project builds correctly with VS (no build or link error) but when I run a quick hello world, I have an error :
Hello World!
ImportError: No module named site
This is the main.cpp:
#include <iostream>
using namespace std;
#include "matplotlibcpp.h"
namespace plt = matplotlibcpp;
int main(int argc, char **argv)
{
cout << "Hello World!" << endl;
plt::plot({ 1,3,2,4 });
plt::show();
return 0;
}
This is the CMakeLists.txt:
cmake_minimum_required(VERSION 3.0)
project(PLT)
set (CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -std=c++14")
# main app
add_executable(
plt
src/main.cpp
)
# Matplotlib
find_package(PythonInterp 2.7 REQUIRED)
find_package(PythonLibs 2.7 REQUIRED)
include_directories(${PYTHON_INCLUDE_DIRS})
target_link_libraries(plt ${PYTHON_LIBRARIES})
I read on this forum that this might be caused by some Path issues, but the path to python directories is set:
I am using Microsoft Visual Studio 2017 to build with CMake's "Visual Studio 15 2017 Win64" generator and I run my program through "cmd" and I think the problem is configuring the default python used by "cmd".
I precise that I tried matplotlib on Anaconda Python's prompt and it is working.
Any idea on how to fix this?

This usually occurs when the PYTHONHOME path is invalid or not set
try:
set PYTHONHOME=C:\Python27

Boost, Python: Linker error with Visual Studio 2015

I basically did the same as the person here:
Building/including Boost.Python in VS2013
However, I used an empty cpp file with only the main function and the inclusion of <boost/python.hpp>
#include <boost/python.hpp>
int main() {
return 0;
}
Now I get the strange linker error (in Visual Studio):
1>LINK : fatal error LNK1104: cannot open file 'boost_python-vc140-mt-gd-1_60.lib'
Which is strange, because I have the lib file I think, however, it is called:
libboost_python3-vc140-mt-gd-1_60.lib

You need to configure your Visual C++ project setting.
The following case operates well.
[debug platform mode] x64
[include directory] (..\;;);C:\boost\boost_1_60_0\;C:\Python35\include\; # add your actual boost and python directory path
[library directory] (..\;;);C:\Python35\libs\;C:\boost\boost_1_60_0\stage\lib; # add your actual boost and python library path

Python.h not found while building sample application with cmake and pybind11

I want to build simple app with pybind11, pybind is already installed in my Ubuntu system with cmake (and make install). I use this simple cmake file:
cmake_minimum_required(VERSION 3.0 FATAL_ERROR)
project(trt_cpp_loader )
find_package(pybind11 REQUIRED)
add_executable(trt_cpp_loader main.cpp)
set_property(TARGET trt_cpp_loader PROPERTY CXX_STANDARD 11)
This is main.cpp:
#include <iostream>
#include <pybind11/embed.h>
namespace py = pybind11;
using namespace std;
int main(){return 0;}
when I build it, I get:
In file included from /usr/local/include/pybind11/pytypes.h:12:0,
from /usr/local/include/pybind11/cast.h:13,
from /usr/local/include/pybind11/attr.h:13,
from /usr/local/include/pybind11/pybind11.h:44,
from /usr/local/include/pybind11/embed.h:12,
from /home/stiv/lpr/trt_cpp_loader/main.cpp:2:
/usr/local/include/pybind11/detail/common.h:112:10: fatal error: Python.h: No such file or directory
#include <Python.h>
^~~~~~~~~~
compilation terminated.
how can I fix this problem? (python-dev and python3-dev are already installed, Python.h is available)

You'll want to use the pybind11_add_module command (see https://pybind11.readthedocs.io/en/stable/compiling.html#building-with-cmake) for the default case of creating an extension module.
If the goal is indeed to embed Python in an executable, it is your reponsibility to explicitly add python headers & libraries to the compiler/linker commands in CMake. (see https://pybind11.readthedocs.io/en/stable/compiling.html#embedding-the-python-interpreter on how to do that)

Following the Wenzel Jakob's answer I want to put an example of CMakeLists.txt for compiling the example provided in this tutorial:
// example.cpp
#include <pybind11/pybind11.h>
int add(int i, int j) {
return i + j;
}
PYBIND11_MODULE(example, m) {
m.doc() = "pybind11 example plugin"; // optional module docstring
m.def("add", &add, "A function which adds two numbers");
}
and
# example.py
import example
print(example.add(1, 2))
and
# CMakeLists.txt
cmake_minimum_required(VERSION 2.8.12)
project(example)
find_package(pybind11 REQUIRED)
pybind11_add_module(example example.cpp)
now in the root run
cmake .
make
now run the python code by
python3 example.py
P.S. I have also written some instructions here for compiling/installing the pybind11.

Maybe just install the Python headers? For example, on Ubuntu you can install the sudo apt-get install python-dev (or python3-dev or pythonX.Y-dev) package. That could resolve this.