【发布时间】:2015-11-18 15:27:47
【问题描述】:
我创建了这个 matrix.cpp 类,其中我重载了一些运算符:
template <typename T>
class MagicObject {
private:
std::vector<std::vector<T> > mat;
unsigned rows;
unsigned cols;
public:
MagicObject(unsigned _rows, unsigned _cols, const T& _initial);
MagicObject(const MagicObject<T>& rhs);
MagicObject();
~MagicObject();
// Operator overloading, for "standard" mathematical MagicObject operations
MagicObject<T>& operator=(const MagicObject<T>& rhs);
// MagicObject mathematical operations
MagicObject<T> operator+(const MagicObject<T>& rhs);
MagicObject<T>& operator+=(const MagicObject<T>& rhs);
MagicObject<T> operator-(const MagicObject<T>& rhs);
MagicObject<T>& operator-=(const MagicObject<T>& rhs);
MagicObject<T> operator*(const MagicObject<T>& rhs);
MagicObject<T>& operator*=(const MagicObject<T>& rhs);
MagicObject<T> transpose();
// matrix/scalar operations
MagicObject<T> operator+(const T& rhs);
MagicObject<T> operator-(const T& rhs);
MagicObject<T> operator*(const T& rhs);
MagicObject<T> operator/(const T& rhs);
// matrix/vector operations
std::vector<T> operator*(const std::vector<T>& rhs);
std::vector<T> diag_vec();
// Access the individual elements
T& operator()(const unsigned& row, const unsigned& col);
const T& operator()(const unsigned& row, const unsigned& col) const;
T& operator[](const unsigned &i);
// Access the row and column sizes
unsigned get_rows() const;
unsigned get_cols() const;
void _print();
};
我避免放置matrix.hpp,因为它很长。在任何情况下,标头都包含正确工作的实现(当然是在 C++ 中)。 现在我想使用 SWIG 创建一个 python 模块。为此,我创建了这个名为 matrix.i 的 SWIG 接口:
%module matrix
%{
#include "matrix.hpp"
%}
%include "matrix.hpp"
// used for the operator[]
%extend MagicObject<T> {
T& __getitem__(unsigned int i) {
return $self[i];
}
}
// used for _print() function
%extend MagicObject<T> {
void __print__() {
print $self;
}
}
%template(intMagicObject) MagicObject<int>;
%template(floatMagicObject) MagicObject<float>;
%template(doubleMagicObject) MagicObject<double>;
我使用这些命令来创建模块:
swig -c++ -python matrix.i
g++ -O2 -fPIC -c matrix.cpp
g++ -O2 -fPIC -c matrix_wrap.cxx -I/Users/dado/anaconda/include/python2.7
g++ -lpython -dynamclib matrix.o matrix_wrap.o -o _matrix.so
当我执行最后一个命令时出现这些错误:
clang: warning: argument unused during compilation: '-dynamclib'
Undefined symbols for architecture x86_64:
"MagicObject<double>::MagicObject()", referenced from:
_wrap_new_doubleMagicObject(_object*, _object*) in matrix_wrap.o
_wrap_doubleMagicObject_transpose(_object*, _object*) in matrix_wrap.o
_wrap_doubleMagicObject___add__(_object*, _object*) in matrix_wrap.o
_wrap_doubleMagicObject___sub__(_object*, _object*) in matrix_wrap.o
_wrap_doubleMagicObject___div__(_object*, _object*) in matrix_wrap.o
_wrap_doubleMagicObject___mul__(_object*, _object*) in matrix_wrap.o
"MagicObject<float>::MagicObject()", referenced from:
_wrap_new_floatMagicObject(_object*, _object*) in matrix_wrap.o
_wrap_floatMagicObject_transpose(_object*, _object*) in matrix_wrap.o
_wrap_floatMagicObject___add__(_object*, _object*) in matrix_wrap.o
_wrap_floatMagicObject___sub__(_object*, _object*) in matrix_wrap.o
_wrap_floatMagicObject___div__(_object*, _object*) in matrix_wrap.o
_wrap_floatMagicObject___mul__(_object*, _object*) in matrix_wrap.o
"MagicObject<int>::MagicObject()", referenced from:
_wrap_new_intMagicObject(_object*, _object*) in matrix_wrap.o
_wrap_intMagicObject_transpose(_object*, _object*) in matrix_wrap.o
_wrap_intMagicObject___add__(_object*, _object*) in matrix_wrap.o
_wrap_intMagicObject___sub__(_object*, _object*) in matrix_wrap.o
_wrap_intMagicObject___div__(_object*, _object*) in matrix_wrap.o
_wrap_intMagicObject___mul__(_object*, _object*) in matrix_wrap.o
"_main", referenced from:
implicit entry/start for main executable
ld: symbol(s) not found for architecture x86_64
clang: error: linker command failed with exit code 1 (use -v to see invocation)
我该如何解决这个问题?谢谢
编辑这里是我的 matrix.cpp 文件(只是一个 sn-p):
#include "matrix.hpp"
#include <iostream>
// Parameter Constructor
template<typename T>
MagicObject<T>::MagicObject(unsigned _rows, unsigned _cols, const T& _initial) {
mat.resize(_rows);
for (unsigned i=0; i<mat.size(); i++) {
mat[i].resize(_cols, _initial);
}
rows = _rows;
cols = _cols;
}
// Copy Constructor
template<typename T>
MagicObject<T>::MagicObject(const MagicObject<T>& rhs) {
mat = rhs.mat;
rows = rhs.get_rows();
cols = rhs.get_cols();
}
// (Virtual) Destructor
template<typename T>
MagicObject<T>::~MagicObject() {}
// Assignment Operator
template<typename T>
MagicObject<T>& MagicObject<T>::operator=(const MagicObject<T>& rhs) {
if (&rhs == this)
return *this;
unsigned new_rows = rhs.get_rows();
unsigned new_cols = rhs.get_cols();
// ..... rest here
【问题讨论】:
-
您显示的链接器错误看起来与您没有在库中的任何位置正确实例化 C++ 中的模板完全一样,但是由于您没有将问题简化为最小的完整示例,因此我无法进一步分析,但可能您根本没有给出默认构造函数的 定义 并且到目前为止还没有发现它。见stackoverflow.com/questions/495021/…
-
%extend 中的
print $self也是错误的,并且应该在出现语法错误的链接器错误之前很久就失败了。 -
@Flexo 对不起,我不太明白你的意思。我编辑了帖子,也许你可以给我一个例子。
-
我猜你没有在 matrix.hpp 的任何地方定义
MagicObject::MagicObject。要么不声明它,要么在适当的地方提供定义。