Why is the default CPP code different from that in MATLAB official documentation?

[Silver-Fang]

The default code created is like:

#include "mex.h"

void mexFunction(int nlhs, mxArray* plhs[], int nrhs, const mxArray* prhs[]) {
    mexPrintf("hello, world\n");
}

But in MATLAB documentation:

#include "mex.hpp"
#include "mexAdapter.hpp"

class MexFunction : public matlab::mex::Function {
public:
    void operator()(matlab::mex::ArgumentList outputs, matlab::mex::ArgumentList inputs) {
        // Function implementation
        ...
    }
};

[AndyYangjd]

It seems not difference. The default code is c-style, but the code in Matlab documentaion which you show is cpp-style.

[Silver-Fang]

It seems not difference. The default code is c-style, but the code in Matlab documentaion which you show is cpp-style.

I think I have chosen somewhere to use cpp, and the code file itself is cpp.

[gharveymn]

Ah, sorry that I never responded to this.

The C style default is used if you select Matrix API in the wizard. The Matrix API may be used with either C or C++ code, and is backwards compatible with older versions of MATLAB.

Your second example should be the code that is generated if you were to select Data API in the wizard.

I hope that helps explain things.

[AndyYangjd]

I think I have chosen somewhere to use cpp, and the code file itself is cpp.

In fact, i also use cpp, but many people in our country choose to use matlab, so i have to wrap the c/c++ code to matlab/simulink for them.

The operation mechanism in simulink is c-style for embeded development. So if your code is c++(object-oriented, OOD), you have to wrap your cpp code with pointer. This warp can be done in a c file or in s-function defination.