将ndarray从Python扩展名返回到C [英] Return ndarray from Python-extension to C
问题描述
我有一个C程序,在这里我可以调用Python函数.我希望这个Python函数返回一个ndarray,然后可以在我的C代码中使用它.我需要一个ndarray,因为我想用Python代码记录声音并使用sounddevice.record()将数组发送回C代码.目前,我正在使用PyArrayObject *作为接收的变量类型(pValue).但是我得到的输出都是零.我已经尝试过使用该程序来接收整数,并且确实可以正常工作,所以特别是在传递数组的问题所在.
I have a C-program from where I call a Python function. I want this Python function to return a ndarray which I then could use in my C code. I need a ndarray, as I want to the Python code to record a sound and send the array back to the C code using sounddevice.record(). At the moment I am using PyArrayObject* as the variable type with which I am receiving (pValue). But all I get at the output are Zero's. I already tried this program for receiving integers and this does work, so it is specifically with passing array's where the problem lies.
C代码
#include <stdio.h>
#include <Python.h>
#include <numpy/ndarrayobject.h>
int main(){
PyObject *pName, *pModule, *pFunc;
PyArrayObject *pValue;
Py_Initialize();
import_array();
pName = PyString_FromString("audioPipe");
pModule = PyImport_Import(pName);
Py_DECREF(pName);
if(pModule == NULL) printf("its null\n");
if (pModule != NULL) {
pFunc = PyObject_GetAttrString(pModule, "recordSound");
if (pFunc && PyCallable_Check(pFunc)) {
pValue = PyObject_CallObject(pFunc, NULL);
for(int r = 0; r < 10; r++){printf("%f\n", pValue[r]);}
if (pValue == NULL) {
Py_DECREF(pFunc);
Py_DECREF(pModule);
PyErr_Print();
fprintf(stderr,"Call failed\n");
return 1;
}
}
else {
if (PyErr_Occurred())
PyErr_Print();
fprintf(stderr, "Cannot find function \n");
}
Py_XDECREF(pFunc);
Py_DECREF(pModule);
}
else {
PyErr_Print();
fprintf(stderr, "Failed to load \\n");
return 1;
}
Py_Finalize();
return 0;
}
Python代码(在audioPipe.py文件中)
Python Code (in file audioPipe.py)
import sounddevice as sd
fs = 44100
channels = 2
def recordSound():
recArray = np.arange(20)
print recArray
print type(recArray)
return recArray
输出
PYTHONPATH=. ./audioPassing
[ 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19]
<type 'numpy.ndarray'>
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
0.000000
推荐答案
我终于找到了使用数组的方法.我将其转换为npy_float **类型.(二维,对于我的应用程序是必需的)使用PyArray_AsCarray()函数.从那时起,您就可以使用它了.更大的问题主要是我的打印陈述.我使用了错误的格式,导致读取错误.另一个重要说明是,要使用PyArray_AsCArray函数,应使用PyArray_Descr *而不是typenum.否则,您会遇到细分错误.
I finally found a way to use the array. I converted it to a npy_float** type. (2 dimensional, necessary for my application) using the PyArray_AsCarray() function. From then on you can use it. The bigger problem was mainly with my printstatement. I used the wrong format which caused it to be read wrong. Another important note is, to use the PyArray_AsCArray function you should use PyArray_Descr* instead of a typenum. Otherwise you get a segmentation fault.
PyArray_Descr *descr;
descr = PyArray_DescrFromType(PyArray_TYPE(pValue));
if (PyArray_AsCArray(&pValue, (void*)&array, PyArray_DIMS(pValue), PyArray_NDIM(pValue), descr) < 0) {
PyErr_SetString(PyExc_TypeError, "error converting to c array");
return NULL;
}
重要的是,如果要打印出来,则需要使用它作为格式.
Important is, if you want to print it out you need to use as format.
printf("datapoint: %"NPY_FLOAT_FMT"\n", array[][]);
这篇关于将ndarray从Python扩展名返回到C的文章就介绍到这了,希望我们推荐的答案对大家有所帮助,也希望大家多多支持IT屋!